University of Southern California Dissertations and Theses

Organizational barriers to delaying high school start time in independent schools

(USC Thesis Other)

Organizational barriers to delaying high school start time in independent schools

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

Organizational Barriers to Delaying High School Start Time in Independent Schools
by
Matthew Stephen Eisenman

Rossier School of Education
University of Southern California
A dissertation submitted to the faculty
in partial fulfillment of the requirements for the degree of
Doctor of Education

May, 2021

© Copyright by Matthew Stephen Eisenman 2021
All Rights Reserved

The Committee for Matthew Eisenman certifies the approval of this Dissertation

David Cash
Alison Muraszewski
Kimberly Hirabayashi, Committee Chair

Rossier School of Education
University of Southern California
2021

iv

Abstract
This study applied Badura’s (1988) social cognitive theory to examine the organizational barriers
to delaying school start time in independent schools. The purpose of this study was to provide
context for school leaders and recommend a framework for change. This study examined the
knowledge, perceptions, and beliefs of key stakeholders as well as the experiences of school
leaders to guide the recommendations. The study used survey data from 129 parents and 173
educators as well as five interviews to examine the problem and address the research questions.
The findings indicated that stakeholders had a reasonable knowledge of adolescent sleep needs
and that there was strong support amongst the parents and educators surveyed for both the AAP
recommendations and a school start time delay within their own schools. The findings showed
that stakeholders anticipated benefits to student health, happiness, stress, and learning.
Additionally, they anticipated challenges around overall student time commitments as well as
parent and faculty schedules. Stakeholders expected that a delay would impact commutes, but
some saw that as a benefit while others anticipated it as a challenge. Moreover, those leaders
who have guided their schools through such a change discussed the positive response to the
change within their communities. Leaders discussed the benefits to the students and support from
the parents. They also encouraged examining the full range of a student’s commitments
throughout a school day and suggested engaging the faculty in the change process to garner their
support. This study offers a review of the relevant literature, the perspectives of key
stakeholders, and the experiences of other school leaders to shape the recommendations for
school leaders. With faculty involvement, schools should examine both their academic schedules
and the full range of expectations they place on students as they seek to improve student sleep
outcomes.

v

Acknowledgements
It is with the love and support of a great many people that I have arrived at this point in
my academic journey. Completing this doctorate during a pandemic has been challenging, to say
the least, but it is with my deepest gratitude that I thank Dr. Kim Hirabayashi for her guidance
throughout this program. As a committee chair, she is always thoughtful in her feedback and
professional in her engagement. Most importantly, in all our meetings, she always had the effect
of “turning down the volume” on the intensity of this process. I will long be grateful for her
leadership here. Additionally, I would like to thank Dr. Alison Muraszewski and Dr. David Cash
for serving on my committee and giving me meaningful feedback on this work. The foundational
work for this study originated in Dr. M’s class, and she has been a guiding force throughout my
process.
In addition to my committee, there are many that have been instrumental in my success at
the University of Southern California. I took classes with a few key professors for more than one
semester. In addition to Dr. M, Dr. Wilcox and Dr. Malloy have both been tremendously
important to my development as a student and researcher. While great professors have been a
hallmark of my experience, the peers in my cohort are also deserving of my gratitude. To Astin,
Jessica, Moe, Erica, Jen, and the many others with whom I met in class, breakout rooms, and
notetaking groups, without you all, my experience in the program would have been lacking.
In arriving at this moment, I must also thank two exceptional professors I had in college.
Dr. Michael Elliott and Dr. Michael Evenden have been supporters of mine and great mentors
from my first days on campus at Emory. Each took a great interest in me as a student, and I try to
model my pedagogy on theirs. Additionally, I am forever grateful to the mentorship I have
received from Keith Evans, whose advice has been a guiding force through my professional and

vi

academic journeys. Reflecting on my success as a student and as an educator in independent
schools, I am forever grateful to many exceptional teachers I had growing up. My deepest thanks
to Weldon Bradshaw, Betty Hunt, and Ann Griffin. Thank you for always believing in me, even
when I was not a model student. Each of you had the ability to see past my grade (or ability on
the track) and get to know me as an individual. I have tried to keep that in mind with my
students.
Finally, I am most grateful for the love and support of my family throughout my
academic journey. My grandfather, Marx Eisenman, deeply valued education and his vision for
my future has made this all possible. To my parents, Mike and Bonnie Eisenman, thank you for
your continued encouragement, for our chats during my commute, for our appreciation of
palindromes, and for being tremendous grandparents to Reid and Sophie. Most importantly, I
would not have been able to take the time for this work without the dedication of Dana
Eisenman. Of course, we could not have predicted when I started school how the impact of a
second child and a pandemic would have increased the challenges you faced supporting me. It is
not lost on me just how hard you have worked to make this degree possible. Without your love,
and the love of Reid and Sophie, I would not be here today. There are not words to express my
gratitude.

vii

Table of Contents
Abstract .......................................................................................................................................... iv
Acknowledgements ......................................................................................................................... v
List of Tables ................................................................................................................................. ix
List of Figures ................................................................................................................................. x
Chapter One: Overview of the Study .............................................................................................. 1
Context and Background of the Problem ............................................................................ 1
Purpose of the Project and Research Questions .................................................................. 2
Importance of the Study ...................................................................................................... 3
Overview of Theoretical Framework and Methodology .................................................... 4
Definitions ........................................................................................................................... 4
Organization of the Dissertation ......................................................................................... 5
Chapter Two: Literature Review .................................................................................................... 6
Adolescent Sleep Patterns ................................................................................................... 6
Benefits of Delayed School Start Time ............................................................................ 15
Barriers Around School Start Time .................................................................................. 18
Conceptual Framework ..................................................................................................... 23
Chapter Three: Methodology ........................................................................................................ 29
Research Questions ........................................................................................................... 29
Overview of Design .......................................................................................................... 29
Research Setting ................................................................................................................ 31
The Researcher .................................................................................................................. 32
Data Sources ..................................................................................................................... 33

viii

Validity and Reliability ..................................................................................................... 39
Ethics ............................................................................................................................... 40
Chapter Four: Findings ................................................................................................................. 41
Participating Stakeholders ................................................................................................ 41
Research Question 1: What Do Participants Know and What Are Their Perceptions
About Adolescent Sleep Needs? ........................................................................... 43
Research Question 2: What Do Participants Know and What Are Their Beliefs of the
AAP Recommendation for School Start Time? .................................................... 47
Research Question 3: What Challenges and Benefits Do Participants Anticipate in
Meeting the AAP Recommendations for School Start Time? .............................. 51
Research Question 4: How Has Delaying School Start Time Impacted Key Stakeholders
in Schools That Have Changed Time to Meet AAP Recommendations? ............ 78
Chapter Five: Recommendations .................................................................................................. 96
Discussion of Findings ...................................................................................................... 96
Recommendations for Practice ....................................................................................... 107
Limitations and Delimitations ......................................................................................... 112
Recommendations for Future Research .......................................................................... 115
Conclusion ...................................................................................................................... 116
References ................................................................................................................................... 118
Appendix A: Parent Survey Protocol .......................................................................................... 126
Appendix B: Faculty and Leadership Survey Protocol ............................................................... 130
Appendix C: Leadership Interview Protocol .............................................................................. 136

ix

List of Tables
Table 1: Data Sources 31
Table 2: Narrative Table of Interview Participants 79

x

List of Figures
Figure 1: Graphic Representation of Conceptual Framework 27
Figure 2: Estimates of Adolescent Sleep Needs 44
Figure 3: Stakeholder Agreement with AAP Recommendation 48
Figure 4: Parent Perceptions of Impact on Student Health and Safety Outcomes 53
Figure 5: Faculty Perceptions of Impact on Student Health Outcomes 55
Figure 6: Parent Perceptions of Impact on Student Stress and Happiness Levels 57
Figure 7: Faculty Perceptions of Impact on Student Stress Levels 59
Figure 8: Faculty Perception of Impact on Student Learning 61
Figure 9: Faculty Perceptions of Negative Impact on Student Schedules 66
Figure 10: Parent Perceptions of Impact on Family Time 69
Figure 11: Faculty Perceptions of Impact on Their Own Schedules 71
Figure 12: Morning Commute Times 74
Figure 13: Afternoon Commute Times 74
Figure 14: Perceptions of Impact on Commute Times 75

1

Chapter One: Overview of the Study
Adolescent sleep deprivation, exacerbated by early school start times, presents a
significant risk to the health and well-being of high school students. Currently, the American
Academy of Pediatrics (AAP) recommends a start time for high schools after 8:30 a.m. (School
Start Times for Adolescents, 2014). While school start time is not the only factor contributing to
adolescent sleep deprivation, teen sleep is critical to address because well over half of
adolescents get less than 8 hours of sleep per night (National Sleep Foundation, 2006, 2014;
Wheaton et al., 2016). The data shows that later school start times directly correlate to increased
adolescent sleep duration (Paksarian et al., 2015). Subsequently, longer sleep duration minimizes
risk factors that are typically exacerbated by sleep deprivation in teenagers (Pavia et al., 2014;
Winsler et al., 2015). Often the fears of other potential ramifications of a delayed school start
time inhibit other stakeholders, like parents and school leaders, from responding to the evidence
around adolescent sleep (Boylan et al., 2015). This study focused on the knowledge and
perceptions of parents and school leaders of independent schools with start times currently
before 8:30 a.m. and examined the challenges and outcomes faced by schools that have made
delays to start time in recent years.
Context and Background of the Problem
The National Sleep Foundation recommends 8 to 10 hours of sleep for teenagers from 14-
17-years-old (Hirshkowitz et al., 2015), and the National Sleep Foundation’s Sleep in America
poll from 2006 cites 9 hours as “optimal” and 8 hours as a “borderline” amount (p. 7). Two
studies from the National Sleep Foundation (2006, 2014) cite findings indicating that more than
half of teenagers get less than the borderline recommendation for nightly sleep. A notable
problem, adolescent sleep deprivation is linked to a number of negative risk factors, including
2

poor academic performance (Perkinson-Gloor et al., 2012; Winsler et al., 2015), suicidal ideation
and substance abuse (Winsler et al., 2015), increased daytime tiredness, decreased persistence,
and negative attitudes toward school life (Perkinson-Gloor et al., 2012). The evidence highlights
that later school start times correlate with longer sleep duration in teenagers (Paksarian et al.,
2015; Perkison-Gloor et al., 2012; Winsler et al, 2015). For each additional 30-minute delay in
the school start time, students on average gain 20 minutes of sleep (Paksarian et al., 2015;
Perkison-Gloor et al. 2015). Despite the clear data on the benefits of delayed school start times,
the average start time remains approximately 8:00 a.m. (U.S. Department of Education, 2020;
Wheaton et al. 2015; Winsler et al., 2015).
Purpose of the Project and Research Questions
This study examined the possible reasons for the lack of action toward delaying school
start time at the organizational level and identified researched-based solutions to meaningfully
guide schools in delaying high school start time. Topics explored include the potential gaps in
knowledge of the parents and school leaders about adolescent sleep needs, as well as perceived
barriers to change from the decision-making adults. The sleep research on adolescents has been
consistent over the last twenty years and formed the basis of the AAP’s recommendations. This
study sought to identify obstacles that parents and school leaders have to delaying high school
start times in independent schools to better meet the needs of the student population and build on
the experience of school leaders who have led their organizations through this change. The
research questions guiding this study were:
1. What do participants know and what are their perceptions about adolescent sleep
needs?
3

2. What do participants know and what are their beliefs of the AAP recommendation for
school start time?
3. What challenges and benefits do participants anticipate in meeting the AAP
recommendations for school start time?
4. How has delaying school start time impacted key stakeholders in schools that have
changed time to meet AAP recommendations?
Importance of the Study
As stewards of the health and education of children, parents and educators have the
responsibility to act in the best interest of the students. Numerous studies demonstrate the
significant benefits of a delayed start time, including reducing a range of physical and mental
health challenges faced by adolescents (Pavia et al., 2014; Winsler et al., 2015), reducing car
accidents (Danner & Phillips, 2008), and increasing economic growth over the long term (Hafner
et al., 2017). Additionally, the gap between the data and the action taken on that data is
evidenced by the fact that as of the 2011-2012 school year, only 17.7% of public middle and
high schools started after 8:30 a.m. (Wheaton et al., 2015). For the 2017-18 school year, the U.S.
Department of Education found a similar 82% of public high schools started before 8:30 a.m.
(U.S. Department of Education, 2020). In their national study, Dunietz et al. (2017) found that
resistance to delaying school start time is rooted in either a lack of knowledge or an incorrect
perception about the impacts of a potential school start time delay. This problem was important
to study because current school start times lead to more than half of students getting inadequate
sleep (National Sleep Foundation, 2006). This study focused on the specific context of
independent schools and the way school leaders make decisions in that context. The
responsibility of educators is to focus on the academic growth and overall health of our students
4

first, and the concerns of educators, parents, and coaches should be secondary to meeting the
needs of the students.
Overview of Theoretical Framework and Methodology
The theoretical framework for this study is Bandura’s social cognitive theory. In this
theory, Bandura (1988) suggests that behavior, cognitive and personal factors, and the
environment work interdependently to impact the way people think and act. The framework
facilitates the examination of how individuals build their beliefs based on the environment, as
well as the factors involved in increasing their self-efficacy to affect change. Social cognitive
theory aligns with the overall goals of understanding the interactions between the beliefs of
parents, teachers, and administrators regarding school start times, the environmental factors
contributing to organizational barriers, and the self-efficacy of the stakeholders in regard to the
desire to affect change.
To gather this information, the researcher used a mixed-methods approach to answer the
research questions. Surveys to parents, teachers, and administrators included both quantitative
and qualitative questions. Additionally, interviews were conducted with school leaders that have
guided their organizations through a school start time delay.
Definitions
This section provides definitions of key terms in context of this study.
Adolescent is defined as ages 11-17. In their seminal study, the National Sleep
Foundation (2006) defined adolescent as grades 6-12 (ages 11-17). Though this definition
includes middle school grades, the scope of this study was limited to high school.
5

Independent Schools, according to the National Association of Independent Schools, are
generally defined by being mission-driven, financed primarily through tuition and charitable
contribution, and governed by a board of trustees (“What Are Independent Private Schools, n.d)
Phase-delay is a shift in sleep onset and wake times in adolescence, around the onset of
puberty (School Start Times for Adolescents, 2014). Primarily caused by later melatonin release
as well as an altered sleep drive, phase-delay inhibits sleep onset, causing adolescents to take
longer to fall asleep.
School start time delay is defined as moving school start time to later than 8:30 AM,
meeting recommendations from the American Academy of Pediatrics 2014 policy statement
(School Start Times for Adolescents, 2014).
Organization of the Dissertation
In this dissertation, Chapter One includes an overview of the problem of early school
start times. Chapter Two provides a literature review covering the biology of adolescent sleep,
the risks of adolescent sleep deprivation, the benefit of school start time delay, and
organizational considerations about changing school start time. Chapter Two also includes a
discussion of social cognitive theory and the conceptual framework applying the theory to this
study. Chapter Three addresses the mixed-methods approach to the study. Chapter Four
addresses the findings of the study, and Chapter Five provides recommendations as a result of
the findings.
6

Chapter Two: Literature Review
This literature review seeks to examine the research pertinent to a discussion of delaying
high school start time to meet the AAP recommendation of 8:30 a.m. or later for adolescents
(School Start Times for Adolescents, 2014). This research includes a review of the relevant
literature around adolescent sleep patterns, focusing primarily on the biological changes in sleep
during adolescence and the risk factors of adolescent sleep deprivation. Also pertinent is a
review of the literature around the benefits of delaying school start time, including both the sleep
benefits to individual students and the benefits to the community at large. Additionally, this
literature review examines the literature around the organizational barriers to start time delay,
including stakeholder perceptions and the current state of educational policy on this issue.
Adolescent Sleep Patterns
Many factors contribute to the current status of adolescent sleep. A combination of
biological factors and external factors impacts the amount and quality of sleep that adolescents
get on a nightly basis (School Start Times for Adolescents, 2014). Critical to the discussion of
school start times is the understanding that adolescent sleep times are primarily a function of
biology, rather than a choice that teenagers make (Hummer & Lee, 2016; Kelly et al., 2014;
School Start Times for Adolescents, 2014). When combined with the social factors of early wake
times required for many high schools, the biological delays in sleep onset result in adolescent
sleep deprivation (Paksarian et al., 2015). As a result of their changing biology and external
social factors, adolescents often fail to achieve the recommended sleep duration (Hirshkowitz et
al., 2015; National Sleep Foundation, 2006, 2014). While many of the external factors, including
electronics and caffeine, are difficult to address on a systemic level, school start time is one of
the external factors contributing to adolescent sleep health subject to control by school and
7

political leaders (Owens, 2014). Insufficient sleep in adolescents, as documented by many
researchers including the National Sleep Foundation (2006, 2014), leads to the proliferation of
significant negative risk factors that result from sleep deprivation (Danner & Phillips, 2008;
Pavia et al., 2014; Winsler et al., 2015).
Biological Changes to Sleep-Wake Time
Adolescent sleep onset times are dictated primarily as a biological drive, independent of
external factors. In puberty, adolescents experience “phase-delay” (School Start Time for
Adolescents, 2014, p. 642), the delayed release of melatonin, which can shift their sleep cycles
later by as much as two hours from middle childhood. As a result, teenagers often have trouble
going to sleep before 11:00 p.m. or waking before 8:00 a.m. (School Start Times for
Adolescents, 2014). In their landmark study, Carskadon et al. (1993) made the connection
between delayed sleep onset and biological factors. Their study demonstrated that phase-delay
correlates most closely with the onset of puberty, rather than external social or psychological
settings. Unlike younger children who go to bed at similar times on weekdays and weekends, and
tend to wake naturally in the morning, adolescents have notably different sleep and wake times
on weekends and weekdays (Carskadon et al., 1993). In their review of relevant studies,
including the seminal Sleep in America (2006) poll, Crowley et al. (2007) cite numerous studies
that demonstrate the difference between weekend and weekday sleep-wake times in middle and
high school students. A striking change from earlier childhood, where there is little difference
between weekday and weekend sleep-wake times, high school students show significant
discrepancies. While the sleep onset times differ only by an hour or two between weekdays and
weekends, high school students tend to sleep three to four hours later on the weekends (Crowley
et al., 2007). When not faced with school start times that require early wake times, adolescents
8

are much more likely to get the recommended eight and a half hours of sleep, as demonstrated by
their weekend sleep and wake times (Crowley et al., 2007). Further, the difference between
weekend and weekday sleep patterns indicates the need for compensatory sleep over the
weekend to try to overcome the deficit from the school week (Maslowsky & Ozer, 2014;
Paksarian et al., 2015). Adolescent sleep onset time is primarily a function of biology, rather than
choice. In addition to going to bed later, adolescents also have a harder time falling asleep when
they do get to bed.
The biological drive to stay awake can override a student’s attempt to go to bed early
enough to achieve the minimum recommended sleep duration before needing to wake for early
school start times (Kelly et al., 2014). In their study on adolescent sleep, Taylor et al. (2005)
found that postpubertal adolescents took longer, on average, to fall asleep after prolonged
periods of wakefulness. The researchers, in a clinical setting, examined the time for sleep onset
after periods of wakefulness from 14.5 to 18.5 hours and found that older adolescents have a
“slower rate of sleep pressure accumulation” (Taylor et al., 2005, p. 243) making it harder for
older adolescents to fall asleep after a full day. Despite requiring a similar amount of sleep as
younger children, the phase-delay associated with adolescence pushes back their natural bedtime
(Hummer & Lee, 2016). Hummer and Lee (2016) also cite a number of studies across a wide
range of mammals indicating that biological factors during adolescence, such as the delayed
release of melatonin, push the onset of sleepiness later into the evening.
Adolescent Sleep Needs Compared to Sleep Duration
The National Sleep Foundation (NSF) recommends that 8 to 10 hours of sleep is the
appropriate duration for adolescents ages 14 to 17, the typical ages in high school. This
recommendation is rooted in research from hundreds of studies on the appropriate amount of
9

sleep needed at each stage of development (Hirshkowitz et al., 2015). Similarly, the NSF Sleep
in America polls conducted in 2006 and 2014 use the designations of “insufficient sleep” to
indicate less than 8 hours, “borderline sleep” between 8 and 9 hours, and “optimal sleep” to
indicate more than 9 hours. Similarly, the American Academy of Pediatrics recommends 8.5 to
9.5 hours of sleep per night (School Start Times for Adolescents, 2014). The consensus on the
sleep needs of adolescents is fairly consistent across recommendations. The research points to an
optimal amount of sleep as about 9 hours, with room for variation. Also consistent is that for
adolescents, sleeping fewer than 8 hours a night consistently creates the conditions for sleep
deprivation (Hirshowitz, et al., 2015, School Start Times for Adolescents, 2014).
As the sleep needs of adolescents are clear, there is also similar consensus around the fact
that a significant portion of adolescents are not getting the recommended amount of sleep
(Hirshkowitz et al., 2015; Ming et al., 2011; National Sleep Foundation, 2006, 2014). In their
national study, the NSF (2006) found that 45% of adolescents report receiving insufficient sleep
on a regular basis. Additionally, another 31% report getting a borderline amount of sleep.
Similarly, in their 2014 poll (Sleep in America, 2014), the findings show that parents report 56%
of 15–17-year-olds get less than 8 hours of sleep per night and only 10% meet the nine-hour
recommendation. Using the Centers for Disease Control and Prevention’s data from their Youth
Risk Behavior Surveys from 2007, 2009, 2011, and 2013, Wheaton et al. (2016) identify lack of
sufficient sleep at consistently high rates. Over the four years of the survey, the data is
remarkably similar, with between 68.4% and 69.1% of respondents reporting seven hours of
sleep or less per night (Wheaton et al., 2016). While the researchers identify slight variations
between gender, ethnicity, and grade level, in all demographic breakdowns, the data indicate that
insufficient sleep is a notable problem impacting a significant portion of high school students.
10

One of the challenges with much of the sleep data is that it is either self-reported or based on
parental report or observation. In their meta-analysis of studies using actigraph data, Galland et
al. (2018) find that mean sleep duration for adolescents aged 15-18 is 7.4 hours, significantly
below the medical recommendations of sleep duration. The challenge facing teenagers in getting
adequate sleep has far-reaching consequences to their health and well-being.
Risk Factors Associated with Adolescent Sleep Deprivation
Significant risk factors for adolescents are associated with sleep deprivation. Those risks
include increases in daytime sleepiness and fatigue (Kilgore, 2010; Pavia et al., 2014; Perkinson-
Gloor et al., 2012), decreased academic performance (Galvan, 2020; Ming et al., 2011, Urria et
al., 2017), an increased risk for depression, suicide, and drug use (Clark & Harvey, 2012; Owens,
2014; Winsler et al., 2015), an increase in high-risk behavior (Wheaton et al., 2016) and an
increase in motor vehicle accidents (Danner & Phillips, 2008, Martiniuk et al., 2013, Vorona et
al., 2011, Vorona et al., 2014). These increased risk factors associated with adolescent sleep
deprivation necessitate close examination and action by the adults, on behalf of the students
involved.
Increasing Daytime Sleepiness and Minor Risk Factors
The increase of daytime fatigue and decreased alertness associated with sleep deprivation
has a notable impact on the school day of a typical adolescent (Kilgore, 2010; Pavia et al., 2014;
Perkinson-Gloor et al., 2012). In their study of adolescent sleep and health risks, Pavia et al.
(2014) noted that increases in headaches, fatigue, body pains, irritability, and sadness all increase
with sleep deprivation. Similarly, their study corroborated other national studies on the typical
sleep patterns of adolescents, which includes a decrease in sleep duration that correlates with
age. In a comprehensive review of the literature to date, Kilgore (2010) documents similar
11

findings in alertness, attention, and emotional processing. Perkinson-Gloor et al. (2012) use their
study of more than 3,000 students in Switzerland to deepen the research on adolescent sleep
patterns and habits that students need to be successful. Their study highlights the pattern that for
students getting less than 8 hours of sleep per night, the rates of daytime tiredness increased, and
students showed less behavioral persistence, as well as a less positive attitude toward life overall
(Perkinson-Gloor et al., 2012). These negative risk factors, while generally minor, also have
notable impacts on academic performance.
Academic Impacts
Numerous studies cite a correlation between decreased sleep and decreased academic
performance (Ming et al., 2011, Perkinson-Gloor et al., 2012, Urria et al., 2017). Because student
learning is often measured by academic performance, it is critical to understand how sleep, and
the lack thereof, impacts academic output. In addition to their measures of tiredness, persistence,
and attitude, Perkinson-Gloor et al. (2012) also measured student results in their mathematics
and German classes. Their study demonstrates that students sleeping less than 8 hours a night are
also likely to have lower scores in those subjects than their counterparts sleeping for longer than
8 hours. In their study, Urria et al. (2017) measure weekday time in bed, as well as weekend
sleep and wake times, and their correlation to gray matter volume in the brain. In addition to
showing that shorter weekday time in bed, as well as later weekend wake times, leads to a
smaller gray matter volume, they also show the correlation between poor grades and lower gray
matter volume in particular areas of the brain (Urria et al., 2017). Though they are not studying
the way adolescents use late weekend wake times to try to make up for weekday sleep debt, as
other studies have shown (Maslowsky & Ozer, 2014; Paksarian et al., 2015), they do show that
the pattern of longer sleep on the weekends compared to the weekdays might be a contributing
12

factor to lower gray matter volume in certain areas of the brain and lower academic performance
as a result. Similar results were found by Ming et al. (2011) in their study of nearly 2,000 New
Jersey high school students. Their study shows the correlation between multiple sleep health risk
factors and academic performance. As students have more negative sleep risk factors like short
weekday and weekend duration, early rise times, and late bedtimes, they are more likely to have
poorer academic performance overall (Ming et al., 2017). While it is easy to talk about academic
performance and student learning as the key goals of education, parents and educators also have
a responsibility to student health and well-being beyond the classroom. Beyond academic
performance, insufficient adolescent sleep is linked to serious negative outcomes impacting the
health and wellness of children.
Depression, Suicide, Substance Abuse and Other High-Risk Behaviors
Multiple studies show a connection between adolescent sleep deprivation and depression,
suicidal ideation and attempts, and substance abuse (Chen et al., 2012; Clark & Harvey, 2012;
Owens, 2014; Winsler et al., 2015). In their study, Chen et al. (2012) looked at the sleep patterns
of girls with no history of psychopathology, but with varying levels of risk for depression based
on their mothers’ histories. Their study found that girls with higher risk for depression are also
more likely to have sleep related challenges (Chen et al., 2012). Similarly, for students dealing
with depression, sleep is critical in the treatment of the condition. In their study of adolescents
with depression and insomnia, Clark and Harvey (2012) make the claim that improved sleep
outcomes for adolescents lead to an increased ability to treat their depressive symptoms.
Additionally, with less sleep per night, students are more likely to struggle with hopelessness,
suicidal ideation, and suicide attempts (Winsler et al., 2015). In their study, Winsler et al. (2015)
found that students who average six hours of sleep per night are twice as likely to experience
13

feelings of hopelessness, consider suicide, or attempt suicide than those students who meet the
recommended 9 hours per night of sleep. Their study also shows that teen tobacco, alcohol, and
illicit drug use correlate directly with sleep duration (Winsler et al., 2015). As students get more
sleep, they are less likely to partake in those risk-taking behaviors. Similarly, using the data from
the Centers for Disease Control and Prevention’s national Youth Risk Behavior Survey,
Wheaton et al. (2015) examined the more than 50,000 results over four administrations of the
survey in 2007, 2009, 2011, and 2013 to investigate the correlation between sleep and injury-
related risk behaviors. They found the sleep patterns remained similar over the four
administrations of the survey and that the less sleep adolescents receive on average leads to an
increase in risky behaviors like drinking and driving, texting and driving, and riding with a drunk
driver (Wheaton et al., 2015). While adolescent sleep deprivation contributes to mental health
challenges and risky behavior for teenagers, they also face significant risks by operating motor
vehicles tired throughout the day, and those risks have an impact on the community at large.
Automobile Safety
Drowsy driving and adolescent car accident rates increase with insufficient sleep,
increasing risk to both the adolescent drivers themselves and the community at large (Danner &
Phillips, 2008). A number of studies show the correlation between adolescent sleep deprivation,
school start time, and motor vehicle accidents (Danner & Phillips, 2008, Martiniuk et al., 2013,
Vorona et al., 2011, Vorona et al., 2014). In their study of more than 20,000 young drivers, ages
17-24, in New South Wales, Australia, Martiniuk et al. (2013) showed that drivers who sleep less
than six hours per night are at an increased risk for single-car and multi-car motor vehicle
accidents. Danner and Phillips (2008) compared the accident data from one county which made a
one-hour delay in school start time to the state averages. In the four-year period they studied,
14

while the state rate of adolescent car accidents increased slightly, the county that made the delay
in start time saw adolescent car accidents decrease by more than 15% in the two years after
making the delay in start time (Danner & Phillips, 2008). Similarly, Vorona et al. (2011)
compared the neighboring towns of Virginia Beach, Virginia and Chesapeake, Virginia using
crash data from 2007 and 2008. Virginia Beach high schools started over an hour earlier than
Chesapeake high schools, and the crash data indicates that students in Virginia Beach were
significantly more likely to have car accidents. Vorona et al. (2014) followed up the 2011 study
by examining the rates of traffic accidents in Chesterfield and Henrico counties between the
2009-2010 and 2010-2011 school years. Like in the previous study (Vorona et al., 2011), this
study found that Chesterfield county, with an earlier high school start time, also had an increased
rate of traffic accidents in drivers ages 16 to 18 (Vorona et al., 2014). By comparing similar
counties in each study and using two separate studies in significantly different parts of the state,
Vorona and his teams of researchers demonstrated that adolescent sleep deprivation, confounded
by early school start times, has a significant negative impact on teen drivers and the surrounding
community. Because of the myriad negative risk factors associated with adolescent sleep
deprivation, it is critical to examine ways to increase adolescent sleep and reduce the negative
outcomes for high school students.
In summary, biological and social changes in adolescence have a significant impact on
natural sleep and wake times (Hummer & Lee, 2016; Kelly et al., 2014; School Start Times for
Adolescents, 2014)). As a result, many adolescents struggle to get a sufficient amount of sleep on
a regular basis (Paksarian et al., 2015). Students who do not sleep long enough each night face
higher instances of fatigue and daytime tiredness (Kilgore, 2010; Pavia et al., 2014; Perkinson-
Gloor et al., 2012), are at a higher risk for suicide and drug use (Chen et al., 2012; Clark &
15

Harvey, 2012; Owens, 2014; Winsler et al., 2015), are less likely to do well in school (Ming et
al., 2011, Perkinson-Gloor et al., 2012, Urria et al., 2017), and are less safe in their cars (Danner
& Phillips, 2008, Martiniuk et al., 2013, Vorona et al., 2011, Vorona et al., 2014).
Benefits of Delayed School Start Time
The changes in adolescent sleep patterns and the risks associated with adolescent sleep
deprivation discussed previously should be enough to make parents and educators look for
effective solutions to get teenagers more sleep. While there are many factors that contribute to
the total amount of sleep a teenager gets on an average night, one factor that school and political
leaders can control is start time (Owens, 2014). The evidence suggests that later start times result
in more sleep (Boergers et al., 2014, Carskadon et al., 1998; Owens et al., 2010; Paksarian et al.,
2015). Research also indicates that there is the potential for a significant economic benefit to
delaying school start time (Hafner et al., 2017, Jacob & Rockoff, 2011).
Delayed Start Increases Sleep
While many factors contribute to total adolescent sleep, later school start times directly
correlate to increased sleep duration. Even a 30-minute delay can lead to significant sleep gain.
In a large-scale study of more than 10,000 students, researchers found that for every 30-minute
delay in start time, students gain on average 20 more minutes of sleep (Paksarian et al., 2015).
When schools delay the start time, even by 30 minutes, more students meet the recommended 8
hours of sleep. In their study of 201 students, Owens et al. (2010) found that the 30-minute delay
resulted in nearly an 80% reduction in the number of students who slept fewer than seven hours
per night and the rate of students getting the recommended 8 hours jumped from 16.4% to
54.7%. In a study of schools in one region with a range of start times, Perkisnson-Gloor et al.
(2012) found that a delayed start time correlated with more sleep and less daytime tiredness.
16

Specifically, a 20-minute delay was enough to lower the rate of daytime tiredness and increase
sleep by an average of 16 minutes. Minges and Redeker (2015) undertook a systematic review of
the recent scientific, peer-reviewed studies of delayed start time and adolescent sleep. In their
review, they highlight the fact that each of studies they examined found that a delay in start time
delays wake time and increases total sleep (Minges and Redeker, 2015). The data shows that
later school start times can increase adolescent sleep duration. With an increase in sleep duration,
there is a reduction in the risk factors associated with adolescent sleep deprivation discussed in
the previous section. Beyond increasing total adolescent sleep, delayed start times also have the
potential to benefit the community at large.
Community Benefits of Delayed Start Time
The benefits of a delayed high school start time extend beyond the personal benefits to
the individual students at hand. The whole community has the potential to benefit from a delayed
school start time. Jacob and Rockoff (2011) projected a benefit-to-cost ratio of 9:1. In their
study, they look at the potential gains in lifetime earnings of students because of gains in
academic performance associated with improved sleep and compare it to the potential costs
associated with changes in bus schedules to arrive at their ratio. When Hafner et al. undertook
their study published by the RAND Corporation (2017), the Jacob and Rockoff (2011) study was
the only other published study that examined the economic benefits of delayed school start time.
Unlike the previous study, which was based on a one-hour delay, Hafner et al., (2017) use a
model which projects an 8:30 a.m. start time, corresponding to the recommendations by the
AAP. The other significant difference in the Hafner et al. (2017) study is that rather than
projecting a lifetime benefit-to-cost analysis, they try to examine the year-over-year impact of
moving all middle and high school start times to 8:30 a.m. In their study, they compare the
17

benefits from academic performance increase and reduced auto accidents to the increased costs
of changes to the bussing system. While they find a more modest benefit-to-cost ratio, Hafner et
al. (2017) still model an economic benefit to both individual students and to each state when
considering school start time delay. The study also acknowledges that it does not factor other
potential physical and mental health benefits of increased sleep into their model. While economic
benefit is important, the delay in school start time also has the potential to benefit the health and
safety of the greater community as well.
The reduction in adolescent traffic accidents associated with delayed school start times
(Danner & Phillips, 2008, Martiniuk et al., 2013, Vorona et al., 2011, Vorona et al., 2014) does
not just have an impact on the teenagers behind the wheel, but on the whole community. As each
of these studies highlights the way increased sleep improves adolescent driving, they also imply
a benefit to the community. Traffic accidents do not happen in a vacuum. While some are single
car accidents, many accidents impact those in the community beyond the adolescent drivers.
Well-rested adolescent drivers make it safer for all other drivers to be on the road. A reduction in
adolescent car accidents makes driving safer for everyone.
In summary, delaying school start time directly increases the adolescent sleep duration
(Owens, 2014; Owens et al., 2010; Paksarian et al., 2015). Increased sleep leads to a reduction in
the risk factors identified in the first section of this literature review. While the cost of delaying
school start time is relatively low (Edwards, 2012), there is a significant economic benefit to
doing so (Hafner et al., 2017; Jacob and Rockoff, 2011). There is a macro-economic benefit to
states over the long term as well as individuals (Hafner et al., 2017). Additionally, the roads are
safer for everyone (Danner & Philips, 2008; Vorona et al., 2014).
18

Barriers Around School Start Time
The most significant barrier to change is a lack of knowledge about adolescent sleep
needs amongst parents and school leaders (Owens, 2014; School Start Times for Adolescents,
2014). Often, the perceptions about negative impacts on after-school jobs and activities (Boylan
et al., 2015; Owens, 2014) and overall cost, despite studies showing net benefit (Hafner et al.,
2017; Jacob and Rockoff, 2011), coupled with the lack of knowledge and adherence to the status
quo (Kohl Malone et al., 2017), lead to a significant barrier to change. Generally, school start
time is decided at local levels and on a school-by-school basis for independent schools (Troxel &
Wolfson, 2017). However, a 2019 California state law moved the decisions about school start
time from the local to state level, despite exempting independent schools from the legislation
(Pupil attendance: School start time, 2019). This legislation mandates that high schools,
including charter schools, start no earlier than 8:30 a.m., while leaving actual start time and
organizational change to the local districts (Pupil attendance: School start time, 2019). The
barriers to change and the potential agents of change are rooted in the perceptions of individual
stakeholders, as well as local and state legislation.
Community Knowledge and Perceptions
Many parents lack a clear understanding of the sleep needs of their adolescent children
and the research-based solutions that might benefit them. In their seminal study of teen sleep, the
NSF (2006) found that there is a discrepancy between the percentage of high school students
who report getting enough sleep and the beliefs of their parents. Their national study found that
despite data showing that 87% of high school students fail to get the recommended 8.5 hours of
sleep, 64% of high school parents believe that their students are indeed getting enough sleep. The
AAP cites a lack of education as the key barrier to improving sleep outcomes for adolescents
19

(School Start Times for Adolescents, 2014). Even when the sleep research data is available,
parents simply do not know the information and are unresponsive to the potential changes that
would benefit their students. According to a national study done by C.S. Mott Children’s
Hospital at the University of Michigan (Davis et al., 2015), only 40% of parents believe that a
delayed start time would help their adolescent student get more sleep and almost half, 49%,
would not support a change in start time. Additionally, only 20% of parents were aware of the
recommendations from the AAP in regard to school start time (Davis et al., 2015). Without clear
parental knowledge of the sleep research that would impact their adolescent children, there is
little push for a delayed school start time. In addition to a lack of knowledge regarding
adolescent sleep needs, misconceptions around the barriers to change often lead stakeholder
groups to resist meaningful change.
The perceptions of negative impacts of a change in school start times are often barriers to
making a meaningful change, despite the research not supporting those negative impacts. Parents
can be resistant to changing school start times because of concerns over interference with jobs
and after-school activities. In their longitudinal study, Boyland, et al. (2015) found that over the
course of a three-year period spanning the year before a start time delay and the two subsequent
years, parents and other stakeholders had a more positive perception of the change in each of the
three years of the study. Similarly, in an NSF article on the barriers to changing start time, they
identified that some of the obstacles included worries over transportation costs and athletic
performance (Eight Major Obstacles to Delaying School Start Time). Contrary to perception,
after making a change to start time, some school districts actually saved money by purchasing
public bus passes for high school students rather than supplying their own fleet of school busses.
Rather than a negative impact of reduced practice time, districts saw improved athletic
20

performance from a student body that had healthier amounts of sleep (Eight Major Obstacles to
Delaying School Start Time; Portantino, 2017). Kohl Malone et al. (2017) looked at barriers to
change through a lens of behavioral economics and highlighted that people are more likely to
favor the status quo simply because it is what has been done before. The authors suggest that to
shift the status quo, districts could be required to justify current early start times on an annual
basis, rather than the normal process of lengthy justification for changing. Simply knowing that
many of the concerns that parents have ahead of transitioning to a later school start time end up
being unfounded or rooted in behavioral science rather than in the students’ best interest might
help stakeholders make better informed decisions. Because some of the barriers to change are
reflected in an uneducated population, stakeholder education is one component of meaningful
behavior change.
Parent education is critical to improving sleep outcomes and delaying school start time.
According to the AAP (School Start Times for Adolescents, 2014), the responsibility for
educating stakeholders about the risks of early school start times and the benefits of middle and
high school start times falls on pediatricians and school leaders. In their national study, Dunietz
et al. (2017) found that roughly half of parents underestimate the sleep needs of their children
and that 77% reported not recalling conversations with their pediatricians about sleep health and
hygiene during their child’s well-visits. The AAP claims that in addition to educating parents,
pediatricians are responsible to reaching out and educating school health professionals,
administrators, and coaches as well (School Start Times for Adolescents, 2014). To improve the
quality of this education outreach, in their application of behavioral economic research to this
issue, Kohl Malone et al. (2017) suggest that improving the salience of the messaging and
education on this issue is an important way to affect meaningful change. They offer the
21

suggestion of color-coding data that correlates negative student risk factors on academic
performance and teenage automobile fatalities with school start times. For schools with start
times before 8:00 AM, they suggest using red, for those with times between 8:00 and 8:30 using
yellow, and using green for start times that meet the AAP’s recommendation. Additionally,
highlighting data that is particularly meaningful to parents, e.g., test scores and automobile
accidents increases parental self-efficacy toward engaging in the change process. In order to
enact meaningful change, stakeholders must know the real data around adolescent sleep needs as
well as have their misconceptions corrected through comprehensive education efforts. In some
cases, change must be initiated in a grass-roots way, with parents and educators as change
agents. This is particularly true for independent schools. In other cases, legislation at the local or
state level has the potential to set students up for success.
Education Policy and School Start Time
Typically, school start time is decided on the district level, but recently several states
have introduced or passed legislation to examine and act on the issue (Troxel & Wolfson, 2017).
The national organization, Start School Later, documents legislation introduced and passed
around this issue. While the issue has been introduced in a number of states, only a few have
passed legislation to undertake studies, and only California has mandated an earliest possible
start time for high school at 8:30 a.m. (Start School Later, 2020).
The California legislation was introduced by Senator Anthony Portantino in 2018, and
while it passed the state legislature, it was ultimately vetoed by Governor Jerry Brown
(Legislation, 2020). Portantino reintroduced the bill in 2019; it passed in October of that year and
takes effect in July of 2022 (Pupil attendance: school start time, 2019). Much of the research
presented in this literature review was presented to the California State Legislature as
22

justification for the passage of the bill in October of 2019. In the California bill, school districts
have some power to make their own decisions within the framework provided by the state (Pupil
attendance: school start time, 2019). While they are mandated to start no earlier than 8:30 a.m.,
there is flexibility for rural districts’ schools to have activities before school for small groups that
do not factor into attendance or state funding applications (Pupil attendance: school start time,
2019). These regulations apply specifically to public and charter schools and omit independent
schools, the primary focus of this dissertation.
Beyond the California legislation, there is little broad action on school start time from a
policy perspective. Generally, each district is on its own to consider the research, respond to
parent feedback, or hold hearings on this issue. In the research justifying the California bill, State
Senator Portantino (2017) identified adolescent sleep deprivation as a statewide public health
crisis which requires a statewide response. To date, this perception has not been widely adopted
on a state or national level, and start time is decided by individual districts. Furthermore, by their
very nature, independent schools determine start time outside of district or state policy. As
recently as the 2017-18 school year, the U.S. Department of Education found in its National
Teacher and Principal Survey that public schools, on average, start at 8:00 a.m., indicating little
movement towards the recommendation from the AAP.
In summary, a lack of education around adolescent sleep amongst stakeholders (Owens,
2014; School Start Times for Adolescents, 2014) and negative perceptions about costs and
broader impact (Boylan et al., 2015; Hafner et al., 2017; Jacob and Rockoff, 2011; Owens, 2014)
have led to inaction on the issue of high school start time. California is the only state thus far to
pass statewide legislation on high school start time (Legislation, 2020), which is otherwise left to
each district to determine (Troxel & Wolfson, 2017). Despite the ongoing research into this
23

issue, public high school start time still averages 8:00 a.m. (U.S. Department of Education,
2020).
Conceptual Framework
The theoretical framework for this study is Bandura’s social cognitive theory. Bandura
(1988) suggests that individual behaviors, personal knowledge and perceptions, and external
environmental factors interact in a way that influences an individual’s thoughts and actions. The
framework allows the researcher to examine how individuals build their beliefs based on their
perceptions and environment, and then use that interplay to make decisions. In particular, self-
efficacy, or an individual’s belief that they can successfully generate particular outcomes as a
result of their own action (Bandura, 1977), becomes a critical component of behavior change in
Bandura’s model of social cognitive theory. When key stakeholders feel empowered with the
requisite knowledge and skills, they are more likely to successfully implement behavior change.
In connection to this study, the primary aim of the research is to ascertain the barriers to
change in school start time and provide recommendations based on the experience of school
leaders that have guided their independent schools through a delay in school start time. Through
the lens of social cognitive theory, the researcher examined the factors that lead to the specific
behavior in question, that is, taking action to delay high school start time to meet the
recommendations of the AAP and the sleep needs of high school students. The decision to delay
start time is impacted by personal factors such as stakeholder knowledge and student health
needs, as well as environmental factors such as traffic patterns, extracurricular scheduling, and
the start times at other schools. Additionally, stakeholder self-efficacy, or their belief in their
ability to act on their knowledge and perceptions, plays a role in effecting this organizational
change.
24

The literature review highlights many of the core concepts in both the personal factors
and environmental factors around delaying school start time. Ultimately, the decision to delay
start time falls to school leaders and administrators. However, to successfully implement
organizational change, those leaders need the support of both parents and faculty. One of the
most significant personal factors impacting stakeholder support for the decision to delay start
time is their knowledge of adolescent sleep needs. The AAP recommends robust parent and
school leader education about adolescent sleep as a critical component of implementing a change
in school start time (School Start Times for Adolescents, 2014). A clear understanding of the
significant risk factors associated with adolescent sleep deprivation, including physical and
mental health (Chen et al., 2012; Clark & Harvey, 2012; Owens, 2014; Winsler et al., 2015), car
accidents (Danner & Phillips, 2008, Martiniuk et al., 2013, Vorona et al., 2011, Vorona et al.,
2014), and academic performance (Ming et al., 2011, Perkinson-Gloor et al., 2012, Urria et al.,
2017), would help parents and educators see the value in changing school start time. When
parents and educators see the value in the delay, their support can increase a school leader’s self-
efficacy. If the stakeholders see the value in the change, leaders would engage in the action with
confidence, believing themselves able to effectively produce change in a situation that they could
ordinarily find intimidating (Bandura, 1977). The NSF (2006) cites the discrepancy between the
reported sleep from adolescents and the beliefs their parents have about their children’s sleep.
Davis et al. (2015) identified that 40% of parents do not believe that late start will help students
get more sleep, despite the studies indicating otherwise (Owens et al., 2010; Paksarian et al.,
2015). Additionally, there are other notable environmental factors impacting perceptions of start
time change for families, educators, and leaders in independent schools.
25

Beyond stakeholder knowledge and perceptions, a number of environmental factors
contribute to stakeholder willingness to effect change on school start time. Some of the
environmental factors that impact the behavior of educators and school leaders are those that
school start time would have on extracurricular activities, as well as the impact on a school
where multiple divisions might have different start times. Additionally, those school leaders are
influenced by the start times at independent schools that are perceived to be in the competing
market. Because of the impact of the start time of neighboring schools, this study examined the
beliefs of the stakeholders in the Atlanta Area Association of Independent Schools (AAAIS),
considering the perceptions of schools that start before and after the AAP recommendation of
8:30 a.m. (School Start Times for Adolescents, 2014). Bandura (1977) addresses modeling as a
component of behavior change, and the experiences of schools within the AAAIS that have
effectively made a change can serve as an example to schools considering such a delay. These
models also have the effect of increasing the self-efficacy in school leaders at peer organizations.
Similarly, families that choose independent schools are influenced by a number of
relevant environmental factors. In Atlanta, traffic during morning school and job commutes can
alter travel time significantly. Because independent schools tend to draw from a wide geographic
area, small changes in departure time from home can lead to notable differences in total
commute time given Atlanta traffic patterns. Additionally, many families have two working
parents. Job start time often dictates the time parents need to drop children off at school. Many
families also have students in multiple school divisions at the same school. A change to high
school start time would present a notable environmental barrier to families with students that
have multiple different start times. Similar to the challenges that school leaders face, parents also
consider the way a start time delay would affect their child’s extracurricular activities and job
26

opportunities. The combination of personal and environmental factors also impacts stakeholder
self-efficacy.
A lack of knowledge about the negative risk factors of adolescent sleep deprivation and
the misperceptions of adolescent sleep deprivation, coupled with the perceived impact from the
environmental factors, act as inhibitors to self-efficacy. Stakeholders are less likely to act on the
delay of high school start time if they lack knowledge of adolescent sleep needs and the impact
of school start time delay or if they are worried about the ramifications of other environmental
factors. Kohl Malone et al. (2017) looks at this issue through the lens of behavioral economics
and suggests a number of ways to influence efficacy, including changing the default and making
the education messaging salient. As stakeholder knowledge and beliefs change with education,
they might have more self-efficacy around changing the start time, even with the considerations
from the impacts of the environmental factors.
Bandura (1988) discusses the interactions between behavior, personal factors, self-
efficacy, and environmental factors in social cognitive theory. It is the interactions of these
factors that impact whether or not an independent school will change behavior through a delay in
school start time. Social cognitive theory offers a compelling lens through which to explore the
knowledge and perceptions that stakeholders have about adolescent sleep and the way a change
in start time would impact their daily lives. Figure 1 illustrates the conceptual framework
described in this section.

27

Figure 1
Graphic Representation of Conceptual Framework

In summary, adolescent sleep deprivation is a significant problem (Hirshkowitz et al.,
2015; National Sleep Foundation, 2006; National Sleep Foundation, 2014) that is affected by
both biological factors (Hummer & Lee, 2016; Kelly et al., 2014) and environmental factors
(Paksarian et al., 2015). Adolescent sleep deprivation promotes negative risk factors, including
negative impacts on academic performance (Ming et al., 2011, Perkinson-Gloor et al., 2012,
Urria et al., 2017), student mental health and well-being (Chen et al., 2012; Clark & Harvey,
2012; Owens, 2014; Winsler et al., 2015), and automobile safety (Danner & Phillips, 2008,
Martiniuk et al., 2013, Vorona et al., 2011, Vorona et al., 2014). Later school start times allow
adolescents to get more sleep (Owens, 2014; Owens et al., 2010; Paksarian et al., 2015),
reducing the impact of the negative risk factors. Later school start times also have a direct
positive benefit on the local economy (Hafner et al., 2017, Jacob & Rockoff, 2011). The most
significant barriers to change are a lack of knowledge amongst the stakeholders (Owens, 2014;
28

School Start Times for Adolescents, 2014) and misconceptions of the impacts of the change
(Boylan et al., 2015; Hafner et al., 2017, Jacob & Rockoff, 2011). Currently, the school start time
is primarily set at the individual independent school or district level. Some states are exploring
statewide legislation (Legislation, 2020), and California is the only state that has passed
legislation mandating that public and charter high schools start no earlier than 8:30 a.m. (Pupil
attendance: school start time, 2019). To study this issue in independent schools, the researcher
used the lens of social cognitive theory (Bandura, 1988) to look at the interactions between
personal and environmental factors, self-efficacy, and behavior change.

29

Chapter Three: Methodology
Chapter three introduces the methodology of the study. It addresses the study’s purpose
of understanding organizational barriers to and lack of action around delaying school start times
in independent schools. This chapter includes the specific research questions, as well as a
discussion of the surveys and interviews used in the study. Additionally, this chapter discusses
the overall nature of the study, as well its validity, reliability, and ethics.
Research Questions
1. What do participants know and what are their perceptions about adolescent sleep
needs?
2. What do participants know and what are their beliefs of the AAP recommendation for
school start time?
3. What challenges and benefits do participants anticipate in meeting the AAP
recommendations for school start time?
4. How has delaying school start time impacted key stakeholders in schools that have
changed time to meet AAP recommendations?
Overview of Design
In order to examine the research questions, the study used a mixed-methods approach with a
combination of surveys and interviews. The surveys have both qualitative and quantitative
components, while the interviews are qualitative. In order to gauge the stakeholders’ knowledge
and perceptions about adolescent sleep, as well as their beliefs about the AAP recommendations
and their perceptions of the challenges and benefits of delaying school start time, this study
deployed a parent survey and a school faculty and leadership survey. By surveying a broad
population from a large number of schools and multiple stakeholder groups, the researcher
30

assessed the knowledge and beliefs of stakeholders in institutions that have a start time that does
not meet the AAP recommendation. The surveys were administered broadly to all willing
participants at traditional AAAIS high schools. To examine the fourth research question, the
researcher conducted interviews with school leaders at AAAIS schools that have delayed school
start time to meet the AAP recommendation. By interviewing school leaders who have guided
their institutions through this organizational change, this study provides not only an
understanding of the potential barriers to change, but also a roadmap to school leaders on the
process and long-term ramifications. Taken together, these two methods provide a
comprehensive and meaningful perspective on what stakeholders know and believe in addition to
gleaning key insight from experienced leaders.

31

Table 1
Data Sources
Research Questions Surveys Interviews
RQ1: What do participants know
and what are their perceptions
about adolescent sleep needs?
X
RQ2: What do participants know
and what are their beliefs of the
AAP recommendation for school
start time?
X
RQ3: What challenges and
benefits do participants anticipate
in meeting the AAP
recommendations for school start
time?
X
RQ4: How has delaying school
start time impacted key
stakeholders in schools that have
changed time to meet AAP
recommendations?
X X

Research Setting
This study is situated in the Atlanta Area Association of Independent Schools (AAAIS).
The AAAIS is comprised of approximately 70 schools serving roughly 35,000 students (About
AAAIS, 2018). Of those 70 schools, 31 have traditional high school programs (School Search,
2018). This study included participating stakeholders from eight of the 31 AAAIS high schools.
Using a broad range of independent schools with varying sizes, locations, and missions within
the metro Atlanta area provided context beyond a single school location and allowed for larger
sample populations for the surveys. Additionally, it afforded the opportunity to use the
32

experience of leaders who have already stewarded their organizations through this change as a
meaningful component of the research process. Of the 31 traditional high schools in AAAIS,
only five have a published start time at 8:30 a.m. or later. Surveying the high school parents, as
well as faculty and school leaders, about their knowledge, perceptions, beliefs, and experiences
around high school start time created a robust participant pool to address the research questions.
Similarly, the researched interviewed leaders from two of the schools with a start time at 8:30
a.m. or later. This approach to the setting secured a robust and wide-reaching participant
population in the Atlanta area. Rather than selecting a single school, focusing on the AAAIS
increased sample populations and allowed for validity and reliability across a large population at
a diverse range of independent schools in the Atlanta area.
The Researcher
The researcher is a doctoral candidate in the Organizational Change and Leadership
program of the Rossier School of Education at the University of Southern California.
Additionally, the researcher is an independent school educator and leader at one of the AAAIS
schools that was part of the study. In considering positionality, the researcher’s role was
exclusively as doctoral candidate and researcher. Particularly when stakeholders from the
researcher’s own organization participated in the study, it was important to distinguish between
the role as researcher from the researcher’s position at the school. This was done by only
communicating about the study through the researcher’s University of Southern California email
address and using the same introductory language at all participating schools. The research itself
was not undertaken at the request of the researcher’s institution of employment, nor was it
necessarily indicative of any potential action that the administration of the school may be
considering in terms of scheduling. However, when engaging in research with other AAAIS
33

schools, the researcher addressed both of these positions in the initial communication with the
school leaders. The researcher’s role as an educator within the Atlanta independent school
community in addition to that as a graduate student lent credibility with the school leaders that
were asked to engage in the research process. Merriam and Tisdale (2016) address the
inside/outside nature of the researcher. Because the researcher is both conducting the study and
an independent school leader and educator, they understand both the theoretical nature of the
research as well as what it might take for an independent school to address this organizational
change.
Additionally, the researcher addressed the biases they brought to the research on this
particular topic (Creswell, 2014). Through the review of the relevant literature, the researcher
believed in the value of a delayed high school start time. From anecdotal experience, which is
supported in the research cited in the previous chapter, the researcher seen high school students
suffer as a result of sleep deprivation, of which early school start time is a contributing factor. To
mitigate that bias in the research, the researcher purposefully explored questions that address
participants’ knowledge, perceptions, and experiences around this topic, rather than exploring the
value of changing school start time (Malloy, 2011; McEwan & McEwan, 2003). Further, to
mitigate bias in the interviews and surveys, interview questions were written to remove bias and
survey questions were primarily adapted from existing metrics (Merriam & Tisdale, 2016),
allowing respondents to feel comfortable expressing their knowledge, perceptions, and
experience.
Data Sources
The research was conducted in two ways: surveys addressed all four research questions
and interviews addressed RQ4 in more depth. These two methodologies were conducted
34

concurrently, as there was little overlap in the populations surveyed and interviewed. The
research was conducted between September 14, 2020 and November 17, 2020.
Surveys
Surveys were used to address participants’ knowledge and perceptions of adolescent
sleep, their beliefs around the AAP recommendation of an 8:30 a.m. start time, and the
anticipated challenges and benefits of a change in school start time.
Participants
The participants for the surveys were parents, teachers, and school leaders from AAAIS
schools. The participants were selected as a convenience sample (Creswell, 2014, Merriam &
Tisdale, 2016). In order to generate a participant list, the researcher contacted school leaders at
all 31 AAAIS schools to ask if they would be willing to send the survey instruments to their
relevant populations on my behalf. Of the 31 schools contacted, ten did not respond to any of the
three email communications. Thirteen declined the offer to participate in any portion of the
study. One school was willing to participate in interviews but not in either survey. Seven schools
participated in the surveys. Three schools distributed the surveys to both faculty and parent
groups and an additional four schools distributed it only to their faculty.
Instrumentation
There were two versions of the survey, one to address the research questions in the
context of parents of high school students (see Appendix A) and the other to address the
educators’ and school leaders’ perceptions of the problem of practice (see Appendix B). The
perceptions of these two groups were important because both groups need to be engaged in the
change process. For both surveys, the instruments were adapted from two existing surveys. The
first portion, on perceptions of adolescent sleep and knowledge and beliefs about the AAP
35

recommendations, was adapted from a survey administered by the C.S. Mott Children’s Hospital
National Poll on Children’s Health (Davis et al., 2015). Some questions were modified and some
added in this part of the survey to address the specific context of independent high schools. The
second part of the survey was adapted from a survey from Boyland et al. (2015). In their study,
they surveyed parents and faculty over three years, with the first year “pre-change” and the two
subsequent years after the delay in school start time. Though Boyland et al. (2015) conducted a
longitudinal study about stakeholder perceptions of the challenges and benefits associated with
school start time change over time, this study only addressed the initial “pre-change” perceptions
of the stakeholder groups. This survey was modified to specifically be relevant to stakeholders in
independent schools. To address the two stakeholder groups, the survey instruments are similar,
but contextualized for the different experiences of parents and educators in relationship to the
research questions and problem of practice.
Data Collection Procedures
Survey data was collected through online surveys using Qualtrics. They were sent as a
link in an email to school administration, who then sent them out to potential participants. Both
participant groups typically have access to email, as it is a primary source of communication
between independent schools and their faculty and parent groups. The researcher did not
anticipate any issues regarding equity and access for participants. The parent survey consists of
25 questions and the faculty and leadership survey has 21 questions. Both were able to be
completed in under fifteen minutes. As a method to prevent survey fatigue and maximize
responses, data was collected between September 30, 2020 and November 10, 2020, after the
initial push to open the school year and well before the end of the semester. The researcher gave
the participants a window of a few weeks to complete the survey, but most respondents take
36

surveys when they receive the initial email or a reminder notification (Robinson & Leonard,
2019).
Data Analysis
The parent survey generated 149 total responses, of which 129 were used in the study.
Because the parent survey collected data on multiple children within a household, there were 163
unique responses to the questions that asked about the habits of each child in the family. The
faculty and leadership survey generated 196 total responses, of which 173 met the threshold for
use in the study. The threshold for use was those surveys for which all of the required questions
had been answered. The only optional questions on either survey were the open-ended responses
and the demographic data collect. As a result, all surveys used in the study included the complete
set of quantitative data.
Because the surveys collected both qualitative and quantitative data, the surveys were
analyzed through a variety of methods. The qualitative data was gathered and coded to identify
trends in the responses covering a range of anticipated benefits and challenges. Those categories
included benefits for students around sleep and health, stress and happiness, and academic
performance. They also included potential challenges like negative impacts on student, faculty,
and family schedules. The quantitative data was filtered and analyzed using the software built
into Qualtrics and in some cases exported to Microsoft® Office Excel®. Questions pertaining to
specific categories, including student health, student learning, student stress and happiness,
negative impacts on the family, and negative impacts on the faculty were grouped to get a
subscale mean for those categories. Data from the quantitative portion was used to create
descriptive statistics and frequencies that reflect the overall knowledge, perceptions, and beliefs
of the sample population.
37

Interviews
The interviews were used to gather experiential data from school leaders who either
shepherded their schools through the delay in school start time and who lead school programs
that meet the AAP recommendation. The interviews investigated primarily the fourth research
question that addresses the way in which the later start time impacts the key stakeholder groups
from the perspective and experience of the interview subject. Some of the questions also address
the participants’ experiences with stakeholder knowledge, perceptions, and beliefs addressed in
the first three research questions.
Participants
The interview participants were school leaders at schools with start times at or later than
the AAP’s 8:30 a.m. start time. Because there are so few of the AAAIS schools that meet the
AAP recommendation, leaders at all five schools that met the criteria were invited to participate.
Two schools offered leadership staff to participate in the study and a total of five interviews were
conducted. Because of the researcher’s role as an insider in the independent school world, as
well as a researcher (Merriam & Tisdale, 2016), they leaned on their network of connections and
their credibility in the field in order to generate participants in the interview portion of the study.
Instrumentation
The interview protocol (see Appendix C) addressed the specific experiences of the
participants. It is a semi-structured protocol (Merriam & Tisdale, 2016), which allowed the
researcher to ask probing questions and continue to engage the participant in an in-depth
conversation around the 14 questions. If the leader was involved in the start time delay, a number
of questions addressed the steps taken to implement the change and the participant’s perceptions
of the way the other stakeholder groups engaged in the process. For all participants, whether they
38

were part of the change process or came to the school after the change, the next part of the
interview addressed the impacts the current start time had on the school and its stakeholders,
including parents, faculty, and students. Interview questions were written allowing for
consideration of various perspectives and allowing the participants to address both perceived
benefits and challenges for the organization and its stakeholders.
Data Collection Procedures
The interviews were each approximately 30-minutes and were conducted on Zoom as a
requirement of the University of Southern California’s institutional review board process and
given the state of the Covid-19 pandemic at the time of the interviews. The interviews were
conducted between November 4, 2020 and November 17, 2020. The interviews were recorded
via Zoom and transcribed using the transcription services of Scribie.com. The researcher then
revised the transcripts for accuracy and grammar.
Data Analysis
To analyze the interview data, the researcher used ATLAS.ti to review the transcripts,
sorting and coding the responses to the interview questions. The codes included demographic
information on the participant and their school, the anticipated and actual benefits and
challenges, the impact on parents and faculty, and the change process and advice for other
schools. The coding then enabled the identification of trends and patterns in the experiences of
the participants, leading to discussion of the impact on key stakeholders, which is the primary
focus of the fourth research question. The trends identified in the interviews with leaders at
schools that have made a start time delay are a key component of the recommendations presented
in Chapter 5.
39

Validity and Reliability
In order to maximize validity and reliability, this study sought to align the research
questions, conceptual framework, and methodological approach. Though this is not a
generalizable study, due to the specific nature of the data collected in the Atlanta area, it might
provide a framework for similar studies in other areas or lend insight to an independent school
considering a start time delay. The survey and interview protocols were designed specifically
with the intention of mitigating researcher bias (Maxwell, 2013). Additionally, the researcher
aimed to collect “rich data” (Maxwell, p. 126, 2013) through robust interviews, including full
transcripts of the interviews. Triangulating the interviews with surveys conducted with two
stakeholder groups and a robust literature review also helped assure validity in the study.
Additionally, the protocols for the study were reviewed by a community of experts before being
implemented.
The interview instrument was developed by the researcher to examine RQ4. The survey
protocols were adapted from two existing studies, with a few additional questions as discussed in
the specific methodology for the survey. To maximize the sample population, all of the AAAIS
schools with a traditional high school program were invited to participate in the study. The
surveys were then distributed to stakeholders at participant schools by school leadership. To
maximize response rate, the time of year the survey was distributed as well as the overall length
of the survey were both accounted for. However, as discussed in the Limitations section in
Chapter 5, the conditions of the COVID-19 pandemic had the effect of dampening schools’
willingness to participate in the study.
40

Ethics
This study exclusively uses human participants in the research. The initial request of
school leaders included the information sheet as directed by the university. Though it included
some demographic data, there was no personal identifying information collected by the
researcher (Creswell, 2014). The distribution lists remained under the control of the individual
school leaders and was not in the researcher’s possession at all. Additionally, all participants in
the interviews gave their permission to be recorded. They were recorded digitally through the
researcher’s University of Southern California Zoom account and stored digitally in the
university’s cloud storage. Pseudonyms for the individual participants and for their institutions
were used in the publication of this study. All participants could opt out of the study at any point
(Creswell, 2014). Finally, there is no physical, social, or emotional risk to the participants by
participating in this study, nor is there any tangible benefit to the participants.

41

Chapter Four: Findings
The chief aim of this study was to examine the organizational barriers to delaying high
school start time in independent schools to meet the AAP’s recommended start time of 8:30 a.m.
or later. While there is a significant body of research indicating the benefits of a delayed start
time to improved sleep outcomes for adolescents, of the 31 traditional independent schools in the
AAAIS, only five have a start time that meets the AAP recommendation. These findings reflect
the knowledge, perceptions, and beliefs about adolescent sleep and school start time held by the
primary decision-making stakeholders in independent schools: parents, faculty, and school
leadership. To that end, these findings reflect the survey results of parents and faculty at the
schools that participated in the study as well as the opinions of the interview participants at
schools with a delayed start time.
Participating Stakeholders
The 31 traditional independent high schools in the AAAIS were invited to participate in
the research study. Of those schools, eight agreed to participate in some capacity. Those schools
represent a geographically diverse set of schools spread throughout the Atlanta area, and the
survey populations include both religious and secular schools. Three schools agreed to share the
parent survey, and seven schools agreed to share the faculty and leadership survey. Of the five
schools that currently have a start time of 8:30 a.m. or later, two schools agreed to participate in
interviews. The interviews sought to gain perspective from schools that already met the AAP
recommendation. A total of five participants from those two schools participated in interviews.
To protect those participants and their schools, interviewees and schools have been given
pseudonyms and some of the positions have been generalized to avoid participants being
identified by the specific nature of their job title.
42

School Start Times as Reported by Participants
The start time of the school day is a contributing factor to whether or not students are
getting enough sleep. When asked, “On a typical school day, what time does the academic day
begin,” the faculty and leadership reported a range of times from before 7:00 a.m. to 8:30 a.m. or
later. Thirty-five (20.2%) of the faculty reported a start time before 8:00 a.m., including six who
reported starting before 7:30 a.m. The majority of faculty, 101 (58.4%), reported starting
between 8:00 and 8:29 a.m., and the remaining 37 (21.4%) reported starting at 8:30 a.m. or later.
Because none of the AAAIS schools publish a start time before 7:45 a.m., there might be
variation in the way the question was interpreted, with some faculty and leadership reporting
their regular arrival to school or their workday start time, rather than the start of the academic
school day or “first bell.” However, even with the variation in interpretation, 79% of faculty
reported a start time before the AAP recommendation of 8:30 a.m. This is appropriate, given the
schools participating in the survey, as only one of them has a published time at 8:30 a.m. or later.
The parent data on academic day start time follows a similar pattern, with slightly less
variation. While all of the schools who participated in the parent survey publish a start time
between 8:00 and 8:29 a.m., 16 responses (about 10%) reported their high school child has a start
time before or after that range. The other 147 (90%) reported a start time in this expected range.
This variation could be due to a number of factors, including students who are expected to arrive
before the start of the regular academic day or students who have an open period first period and
are not required to report to school for the first bell. Current start time is foundational in
considering how parents will respond to questions asking them about sleep needs and
perceptions of a delayed start time.
43

Research Question 1: What Do Participants Know and What Are Their Perceptions About
Adolescent Sleep Needs?
To examine participants’ knowledge and perceptions about adolescent sleep needs,
participants in both the faculty and parent surveys were asked to estimate the amount of sleep an
average teen needs per night. Additionally, parents were asked to approximate the amount of
sleep their high school students actually get on an average school night. While teachers cannot
make estimates of the amount of sleep their students get regularly, they were asked to identify
the frequency with which their students fall asleep in class and if that happens more frequently in
the morning or in the afternoon. Both stakeholder groups were asked if they agreed their students
were getting enough sleep.
Stakeholders Estimate Adolescent Sleep Need at NSF Borderline Designation
When asked to estimate the needed sleep duration for teenagers, stakeholders estimated a
sleep duration in the National Sleep Foundation’s “borderline” designation. In their Sleep in
America polls, the NSF (2006, 2014) cites less than 8 hours as an insufficient amount of sleep, 8
to 9 hours as borderline, and 9 hours or more as optimal. When the AAP recommended an 8:30
a.m. start time, they also recommended that the appropriate amount of sleep for a teenager is
between 8.5 and 9.5 hours per night (School Start Times for Adolescents, 2014). When asked to
estimate nightly sleep needs, the data show that parents and educators estimate a sleep need that
fits in the NSF’s borderline category and is near the minimum recommendation from the AAP.
The mean sleep duration on the parent survey was approximately 8 hours and 37 minutes.
Similarly, on the faculty and leadership survey, the mean was approximately 8 hours and 25
minutes. These averages, in the NSF’s borderline category and on the low end of the AAP’s
44

recommendation, indicate that stakeholders’ estimates are close to adolescent needs, but that
there is still an opportunity for better stakeholder education.
Figure 2 shows the sleep needs estimates made by parents and faculty. Nineteen percent
of faculty and 12% of parents identified an amount that the NSF would define as insufficient.
The average of the responses fell in the NSF’s borderline category, including 36% of faculty and
38% of parents who identified a needed amount between 8 and 9 hours. Less than half of faculty
(45%) and half of the parents (50%) identified an optimal amount of sleep, more than 9 hours per
night. Similarly, almost half of the respondents to the surveys identified adolescent sleep needs
below the AAP’s recommendation, with 46% of faculty and 41% of parents identifying a total
amount below the recommended 8.5 hours. Because knowing the recommendations and
implementing those recommendations at home are not the same, parents were also asked to
identify an average night’s sleep for their children.

Figure 2
Estimates of Adolescent Sleep Needs

0
5
10
15
20
25
30
35
5-5.4
5.5-5.9
6-6.4
6.5-6.9
7-7.4
7.5-7.9
8-8.4
8.5-8.9
9-9.4
9.5-9.9
10-10.4
10.5 or more
Percentage of Responses
Hours Needed
Sleep Needs Estimates
Parents
Faculty
45

Parent Estimates of Adolescent Sleep as Insufficient
The majority of parents estimated that, on average, their students do not meet the NSF’s
borderline designation for sleep duration. The 129 respondents reported data for 163 high school
students. The responses indicate that their children are generally not getting the recommended
amount of sleep. Parents reported that only 47 of 163 (29%) students get even the borderline
recommendation of 8 hours of sleep per night. Of those 47 students, only 13 (8%) meet the AAP
recommendation of 8.5 to 9.5 hours per night. The median nightly sleep duration reported is
seven hours, well below the recommendations from the NSF and AAP. Taken together, these
two questions indicate that parents generally know how much sleep their children should be
getting and know that their children are not getting enough. While educators cannot assess how
much sleep children are getting at home, a lack of sleep can manifest by students falling asleep in
class.
Students Are Sleeping in Class
When asked to report the number of students who fall asleep in class in a normal week,
about 60% of educators indicated that students fall asleep in their class at least once a week, and
25% indicated that at least 3 students a week fall asleep in class. Additionally, a majority of
faculty, roughly 60%, observe that their students fall asleep more frequently in their afternoon
classes. This measure is also impacted by teachers’ classroom practices. One teacher indicated in
the open-ended comments that “I marked that students don’t fall asleep in my class because I
don’t let them” and suggested assessing sleepiness would be a better estimate. To that, students
sleeping in class is not a perfect estimator of student sleep health, but taken with the reported
data from parents, this metric may support the notion that students are not getting enough sleep.
46

Students Are Not Sleeping Enough
Both parents and educators were asked the extent to which they agreed that their students
are getting enough sleep, and their responses show that they do not think their students get
enough sleep. Of the 173 educators surveyed, only seven (4%) agreed with the statement:
“Currently, I believe my students are getting enough sleep.” Of the other 166 educators (96%),
91 (52.6%) disagreed with the statement and 75 (43.4%) strongly disagreed. Similarly, parents
were asked to what extent they agreed with the statement: “Currently, I believe my child is
getting enough sleep.” Of the 163 children reported on in the sample, parents only agreed or
strongly agreed with the statement for 42 students, 26% of the sample population. For 121
students (74%) parents disagreed or strongly disagreed with the statement. A clear majority of
parents (74%) and educators (96%) do not believe that their students are getting enough sleep.
The work of the NSF and the AAP also reflect the general belief that high school students do not
get enough sleep.
Summary
The data around parent and educator knowledge and perceptions of adolescent sleep
needs indicate a discrepancy between what the key stakeholders know about adolescent sleep
and the amount of sleep students are actually getting on a regular basis. In general, the decision-
making stakeholders are able to nearly estimate the amount of sleep teenagers need. They also
agree that their students and children are not getting enough sleep. Despite having a good idea of
how much sleep students need and understanding that their students are not getting that amount
of sleep, the majority of participants are at schools where the start time that does not meet the
AAP recommendation.
47

Research Question 2: What Do Participants Know and What Are Their Beliefs of the AAP
Recommendation for School Start Time?
This research question sought to evaluate the extent to which participants knew about the
AAP’s recommendation and then how the participants feel about the AAP’s recommended start
time once they were given the information. To examine their knowledge of the AAP
recommendation for school start time, participants on both surveys were first told that the AAP
made a recommendation in 2014 and then were asked to estimate the recommendation. They
were then given a statement specifying the actual AAP recommendation and asked to what
extent they agreed with the statement. Finally, participants were asked the extent to which they
would support changing their school’s start time to meet the AAP’s recommendation.
Stakeholders Accurately Estimate AAP Recommendations for Start Time
Moving from participants’ knowledge of adolescent sleep needs to knowledge and beliefs
about the AAP recommended start time, participants were told that the AAP made a
recommendation for earliest start time in 2014. They were then asked to estimate the AAP’s
recommendation for earliest start time. On average, parents and faculty were able to correctly
estimate the recommendation from the AAP. Of the 172 responses on the faculty and leadership
survey, 80% entered a time at or after 8:30 a.m. A majority of parents, 71%, estimated a time at
or after 8:30 a.m. Only five parents and three educators suggested a time before 8:00 a.m. The
mean estimate of the AAP’s recommended start time on the parent survey is about 8:40 a.m. and
8:47 a.m. on the faculty and leadership survey. This indicates that generally, when asked what
they think the AAP would have recommended, a majority of parents and educators are able to
make an estimate close to the actual recommendation.

48

Figure 3
Stakeholder Agreement with AAP Recommendation

Most Stakeholders Agree with AAP Recommendations
More than 90% of parents and faculty expressed agreement with the AAP’s
recommendation. After offing their estimate of the AAP’s earliest recommendation, participants
were shown the following statement: “The AAP urges high schools and middle schools to start at
8:30 am or later, allowing students to get 8.5-9.5 hours of sleep, in order to improve physical and
mental health, safety (by preventing drowsy driving crashes), academic performance and quality
of life.” They were asked to what degree they agreed with the statement. As shown in Figure 3,
the overwhelming response was that parents and faculty agree or strongly agree with the AAP’s
recommendation. Eighty-three faculty members (48%) and 81 parents (63%) selected “strongly
agree,” while another 45% of faculty and 29% of parents agreed with the statement. Taken
together, 93% of faculty and 92% of parents were in some level of agreement with the AAP’s
recommendation. Of the respondents who disagreed with the recommendation, only one
0%
10%
20%
30%
40%
50%
60%
70%
Strongly Disagree (1) Disagree (2) Agree (3) Strongly Agree (4)
Percentage of Responses
Stakeholder Agreement with AAP Recommendation
Parents
Faculty
49

participant on each survey strongly disagreed. The responses on this question reflect a very
strong support for the AAP’s recommendation.
Stakeholders Support Start Time Change
More than 90% of parents and educators indicated that they would support a change in
start time to meet the recommendation. This is a similar percentage to those who agree with the
AAP’s recommendation. However, there were respondents in both surveys who agreed with the
AAP recommendation, but did not support a start time change. On each survey, there were seven
respondents who either agreed or strongly agreed with the recommendations but would not
support a change to their school’s start time. Those who agreed with the recommendation but did
not support a change only represent about 5% of the total sample size; however, their responses
point to the idea that despite a general agreement with the medical recommendations, there might
be inhibiting factors beyond adolescent sleep needs in making a change to school start time. One
parent shared that there would be no benefit because their child would still need to be dropped
off at the same time and another shared that their work schedule would prevent their child from
seeing a benefit. These factors are addressed in further subsequent research questions.
Many educators and parents addressed their support for the AAP’s recommendations
when asked to consider the benefits and challenges of changing start time or if they had
additional comments in the open-ended portion of the survey. Fifty-two educators and 53 parents
indicated that they had no concerns about delaying school start time. Repeated comments on the
idea of delaying start time included those who “love it” or “strongly support a change.” A
number expressed the feeling that they were “all for” the change. One parent expressed that they
“wish it would have happened when my older son was in high school” and that a change
“couldn’t happen soon enough.” Addressing the research that is already published, faculty and
50

parents both expressed a desire to respond to that research. One educator said, “I do not
understand why we haven’t done [it] this year. It would be so beneficial,” and another noted that
“the research behind sleep shows only benefits to students getting more sleep.”
Multiple parents also leaned on the existing research. One said, “Dozens of stud[ies]
show this is better for teens, not sure what the issue is.” Another discussed the shift in sleep
rhythms and the “need to sleep longer in the mornings.” One parent noted that they were a little
surprised at the recommendation because “I thought the recommended start time would be later.”
To that, there were both parents and faculty that indicated that the AAP recommendation might
not go far enough.
Of the responses that expressed specific support for the AAP recommendations, four
educators and nine parents specifically suggested times later than the AAP recommendation. One
parent noted that while “8:30 would be an improvement, but 9:00 is ideal for students’ well-
being and health.” One faculty response also talked about the need to push even later, saying:
I’m not sure that 8:30 is late enough. For some students this might work out to be 15-30
more minutes of sleep, which may prove negligible. If a school was willing to make
adjustments to its schedule, I’d push for an even later start time than this.
Of the response that suggested later start times than the AAP recommendation, most simply
indicated that a 9:00 a.m. start would be preferable to the 8:30 a.m. recommendation. These
comments bolster the more than 90% of parents and educators who would support a delay in
school start time to meet the AAP recommendations.
Summary
To this research question, the survey data is unambiguous. There is strong agreement
from both stakeholder groups about the AAP’s recommendation for an 8:30 a.m. school start
51

time. Additionally, over 90% of participants would support a change in their school’s start time
to meet the recommendation. Not only do parents and educators have a reasonable understanding
of adolescent sleep needs, they also overwhelmingly support changing school start time to meet
medical recommendations.
Research Question 3: What Challenges and Benefits Do Participants Anticipate in Meeting
the AAP Recommendations for School Start Time?
Building on the findings in the previous research questions, this section addresses the
potential benefits and challenges that participants anticipate in meeting the AAP’s
recommendation for school start time. The potential benefits and challenges are wide-ranging,
covering topics including student health and wellness, learning, and schedules, as well as
potential negative impacts on the students, parents, and faculty. These responses were collected
on a Likert scale made up of four potential responses: strongly agree, agree, disagree, and
strongly disagree. For the purposes of quantifying these responses, a “strongly agree” was scored
as a four and a “strongly disagree” was given a one. Participants’ scores on each subscale were
used to find an average score for the participants in each of the primary categories covering the
potential benefits and challenges of start time delay. In addition to the quantitative responses on
the Likert scale, participants in the survey were given three open-ended questions. They were
asked to identify potential benefits and challenges and were given an opportunity to make
additional comments.
While the anticipated benefits and challenges cover a wide range of topics and reflect
significant perspectives on both sides of the conversation, this data should be viewed through the
lens of overwhelming support for school start time delay. The responses in both the quantitative
and qualitative data were wide ranging and reflected real potential benefits and real concerns. In
52

examining this data, it is important to recall that more than 90% of parents and faculty responded
that they would support a start time delay. This indicates that those challenges can generally be
viewed as problems to be solved, rather than as compelling reasons not to make a change in start
time at all, and they will be addressed in the recommendations in the following chapter.
Stakeholders Anticipate an Improvement in Student Health and Safety Outcomes
A significant majority of both parents and educators felt that a delayed start time would
have a positive impact on overall student health. Participants were asked a series of questions to
assess their perceptions on how school start time delay might improve health and safety
outcomes for students. The quantitative data and the open-ended responses showed that parents
and faculty felt that delaying start time would have a benefit to student health and safety
outcomes.
Parents were asked four questions that pertained specifically to student health. They were
asked if they agreed that a start time of 8:30 or later would lead to more sleep, improved health, a
reduction in adolescent traffic accidents, and an overall healthier child. The mean score for
parent responses on the student health and safety section was 3.3, or somewhere between “agree”
and “strongly agree” that a delay in start time would improve overall student health outcomes.
As shown in Figure 4, the responses to the four questions that pertain to student health
indicate that parents believe a delay in start time would improve health outcomes for their
children. Of the 129 responses, 112 (87%) believed the delay would improve their child’s overall
health and 113 (88%) agreed that a start time delay would give them a healthier child. When
asked if they thought a delay would allow their child to get more sleep, 117 (91%) agreed or
strongly agreed with the statement. More parents strongly agreed with the improved sleep
outcomes than any of the other questions in this category. The health outcome that parents
53

thought they were the least likely to see is the reduction in adolescent car accidents. Seventy-four
percent of parents believed that a delayed start time might reduce the rate of adolescent traffic
accidents. While this still represents a majority of parents, it is the question in this section with
the most disagreement. Thirty-four parents (26%) disagreed or strongly disagreed that a delayed
start time would reduce the rates of adolescent car accidents. While there was some variation in
the responses to these four statements, overall, the parents surveyed believed that a delayed start
time would benefit the health and safety of their children.

Figure 4
Parent Perceptions of Impact on Student Health and Safety Outcomes

0%
10%
20%
30%
40%
50%
60%
Strongly
Disagree (1)
Disagree (2) Agree (3) Strongly Agree
(4)
Percentage of Responses
PARENT PERCEPTIONS OF STUDENT HEALTH AND
SAFETY OUTCOMES
Improve My Child's Health
Give Me a Healthier Child
Allow My Child to Get More Sleep
Reduce Rates of Car Accidents
54

In their open-ended responses about the potential benefits of a delayed start time on
student health and safety, 95 of the 115 responses indicated that they saw a potential health and
safety benefit. As a component of their responses, 76 of 95 that discussed potential health and
safety benefits specifically named that more sleep or a better rested child was a potential benefit
to start time delay. This included comments as simple as “more sleep” along with those that
extrapolated the benefits of more sleep to overall health. One shared that “giving children more
time to sleep (even if its [sic] just 30 minutes more) would help them be healthier mentally
through the day.” Three addressed the changes in adolescent circadian rhythm that impact sleep
onset and wake times. Responses also pointed to other areas in overall health, including five who
specifically mentioned their child being more likely to eat breakfast. One parent said, “It would
easily give my child extra time sleeping and the ability to stay up later when needed due to
academic demands, as well as improving the possibility of my teenager sitting down and eating a
healthy breakfast.” In addition to the general health benefits, 12 responses talked about driving
conditions. While the quantitative data showed some skepticism towards there being a potential
driving safety benefit, there were some who felt strongly enough to identify it in the open-ended
response. Those that mentioned driving in the morning discussed both the potential increased
alertness as a result of more sleep and the added safety of driving in daylight, which would be
more likely all year with a later start. As on the parent surveys, educators also identified the
potential health and safety benefits to students.
The faculty and leadership survey had a similar set of questions about student health and
safety, with three statements asking about student health and safety. The first asked about
aligning with students’ natural sleep schedules, the second asked about allowing students to get
more sleep, and the third asked about improving students’ overall health. The mean score for this
55

category was 3.3, between “strongly agree” and “agree,” indicating that the educators surveyed
believe that a delayed start time would benefit student health overall. As shown in Figure 5, on
each of the three questions, between 86 and 90 percent of faculty members who responded agree
or strongly agree that the change would be a benefit to the particular health outcome asked.

Figure 5
Faculty Perceptions of Impact on Student Health Outcomes

0%
10%
20%
30%
40%
50%
60%
Strongly
Disagree (1)
Disagree (2) Agree (3) Strongly Agree
(4)
Percentage of Responses
FACULTY PERCEPTIONS OF STUDENT HEALTH
OUTCOMES
Improve My Students' Health
Align with Natural Sleep Schedule
Allow my Students to Get More
Sleep
56

In their open-ended responses, 110 of the 152 respondents addressed a benefit to some
element of student health and safety. Similar to the parent survey, the faculty primarily identified
the benefit of increased sleep. In line with the parent responses, the educators also addressed the
changes in adolescent sleep, the potential to eat a healthy breakfast, and potential benefits to
driving slightly later in the morning. One response noted that “from the sleep-in days we have
had, the students arrive to their first few classes with an elevated amount of energy. They’ve
likely eaten breakfast and had time to collect themselves before the start of the school day.”
Another pointed directly to the clear research around the benefits of school start time delay,
shifting from personal perceptions to ideas rooted in adolescent sleep data.
Respondents believed that a delayed start time would allow students to get more sleep,
eat a better breakfast, and have safer morning driving conditions. Two also mentioned the health
and wellness benefits to the students of potentially having time to exercise before the start of the
school day. One shared that “in addition to sleep, some may find time for exercise, meditation,
[and] proper nutrition.” Along with the positive impact on student health, participants addressed
the potential impact on student stress and happiness. Managing student stress, particularly in the
independent schools studied here, presents a challenge for parents and educators alike. Stress
levels also impact overall health and well-being. As with improved health and safety outcomes,
parents and educators alike believe that a delayed start time would have a positive impact on
student stress and overall happiness.

57

Delayed Start Will Reduce Student Stress and Increase Overall Happiness
Both parents and educators anticipated that benefits to a delayed start time would include
the reduction in student stress levels and a general increase in happiness. The parent survey
asked two questions about the impacts of a delayed start time on stress and happiness. They were
asked the extent to which they agreed that a delayed start time would reduce student stress and
give them a happier child. The mean on the stress and happiness questions was 3.2, between
“strongly agree” and “agree.” As noted in Figure 6, seventy-nine percent of parents indicated that
they “agree” or “strongly agree” that delaying start time would reduce student stress, while 88%
believed that delaying start time would give them a happier child overall. Even if a delay did not
have a tangible benefit on student stress levels, parents generally believed that their children will
be happier with a later start to the morning.

Figure 6
Parent Perceptions of Impact on Student Stress and Happiness Levels

0%
10%
20%
30%
40%
50%
Strongly
Disagree (1)
Disagree (2) Agree (3) Strongly Agree
(4)
Percentage of Responses
Parent Perceptions of Student Stress and Happiness
Levels
Decrease My Child's Stress
Give Me a Happier Child
58

In their open-ended responses, parents generally addressed a reduction in stress, while
also naming a few specific things that might lead to that reduction. First, 15 parents addressed
the potential for a more relaxed start to the morning, with less stress around getting their children
up and moving for school. One noted that “anything that reduces the morning chaos that requires
them to be “dragged” out of bed “before the sun even gets up sounds like a plus to me.” Because
of the current Covid-19 pandemic, the different modes of learning available this year have given
parents a chance to observe some of the differences. One noted of their son that “on days he is
learning remotely (and thus not having to leave the house so early), it is easier for him to get up
and get started.” Despite being asked in separate quantitative questions, many of the open-ended
responses also linked the overall health of their child to their overall levels of stress and
happiness. It is worth noting that these open-ended responses might have also been as a result of
being prompted by the quantitative questions.
The educators surveyed were also asked their perceptions regarding the potential impact
of a school start time delay on student stress. As shown in Figure 7, about 70% of the educators
surveyed believed that a delayed start time would also have the effect of reducing student stress.
The mean score on that question was 2.9, firmly in the “agree” category. Though the educators
do not anticipate the reduction in stress will be as significant a benefit as other potential benefits
identified in the survey, most believed that a delay in school start time would be beneficial to
student stress levels.

59

Figure 7
Faculty Perceptions of Impact on Student Stress Levels

In their open-ended responses, faculty pointed to similar potential benefits as parents.
They primarily indicated a general reduction in stress levels, mentioning specific benefits of
added time in the morning beyond more sleep. One noted that “better rested, happier students
[…] are more capable of handling the curveballs that life throws at them.” The implication is not
that it is just a general reduction in stress, but that better rested students are more able to handle
stressful situations. Like the parents, educators also addressed the potential benefit of a morning
that feels less rushed and less stressed as students go to start their day. Two also noted the
potential mental health benefits of being more in line with the natural daylight when they get up
and going in the morning.
As with the potential health benefits, parents and educators saw the potential benefits to
student stress levels and overall happiness. They anticipated that a change would make students
better rested and less stressed. By increasing student sleep, reducing morning stress, and
0%
10%
20%
30%
40%
50%
Strongly
Disagree (1)
Disagree (2) Agree (3) Strongly Agree
(4)
Percentage of Responses
Axis Title
Faculty Perceptions of Impact on Student Stress Levels
Decrease My Students' Stress
60

improving happiness, parents and educators looked toward the potential academic benefits of a
delayed school start time. Both stakeholder groups anticipated a potential benefit to student
learning by delaying school start times.
Stakeholders Anticipate Benefits to Student Learning and Academic Performance
When asked about potential benefits to student learning and academic performance,
parents and educators alike believed that a delayed start time would improve academic outcomes
for their students. The parents were asked only one question about the impact on academic
performance. They were asked the extent to which they agreed that a delayed start time would
improve students’ performance in school. The mean score on this question was 3.3. About 85%
of parents believed that there would be a benefit to student performance, with 59 (46%)
respondents strongly agreeing with this statement and another 50 (39%) agreeing. This reflects a
sizeable majority that believes delaying school start time might have a positive effect on
academic performance.
Fewer parents specifically addressed student learning and academic performance than
those that addressed health and stress levels. However, the 12 that did discuss student learning
point to a few critical components of the topic. Beyond the general comment that it might
improve general academic performance like grades and test scores, parents addressed the
potential for students to be more alert and more focused, and to pay better attention in class. One
suggested that more sleep would lead to “better cognitive awareness.” Three parents also
mentioned that a delayed start time, rather than allowing for more sleep, might provide students
and teachers more time to meet for additional support, noting that there might be “more
opportunity for teacher office hours and volunteer tutorials prior to the first class.” Even if
students were not sleeping more, there may be the added benefit of collaborative time with
61

faculty and peers to improve academic outcomes, particularly since after-school availability for
students and teachers is often limited by sports and extracurriculars.
Like the parents, faculty anticipated an academic benefit to a delayed start time. In
addition to being asked the same question as parents about improving school performance, the
educators were also asked about a delay aligning the school day with the best times for student
learning. The mean score on the two questions taken together was 3.2, indicating that the faculty
were somewhere between “strongly agree” and “agree” about potential academic benefits to
delaying school start time. As presented in Figure 8, on the question about the extent to which
they believed a delay aligns with the best times for learning, 43% strongly agreed with the
statement and 47% agreed. This 90% agreement is higher than the 84% that believed it would
improve student academic performance. Additionally, 25 more people selected “strongly agree”
about aligning with the best times for student learning than they did about the potential for
student improvement. This indicates that while educators think a delay is better for student
learning, they are not as confident that it will actually improve performance.

Figure 8
Faculty Perception of Impact on Student Learning

0%
10%
20%
30%
40%
50%
60%
Strongly
Disagree (1)
Disagree (2) Agree (3) Strongly Agree
(4)
Percentage of Responses
Faculty Perception of Impact on Student Learning
Improve my Students' School
Performance
Take Advantage of the Best Times
for Learning
62

In their open-ended responses, 31 of the educators surveyed discussed potential benefits
to student learning. While their quantitative responses indicated that they are less sure it would
directly benefit performance, faculty noted a range of potential academic benefits. Fifteen of the
comments about student learning indicated the belief that students are more likely to be more
awake, more engaged, and more focused as a result of a delayed start time. One educator said
that they think students “would be better rested and better able to focus in class.” They
continued, adding that they “see this translating into better grades and learning outcomes.” Being
more awake, alert, engaged, and focused in class may not directly improve grades, but might set
students up to be able to be more successful in general. As with the parents, two faculty
responses addressed the possibility that the time before school could also be used to meet with
students for extra help. While this might not improve student sleep outcomes, the shift in
schedule might create the space to improve student academic performance.
The stakeholders perceived that a delayed start time would have tangible benefits to
student health and safety, student stress and happiness, and student learning and academic
performance. Across both stakeholder groups and all three categories, no single statement had
less than 70% agreement with the idea that a delayed start time could improve outcomes for
students. In most cases, the agreement was between 85% and 90%, reflecting the percentages of
the two stakeholder groups’ willingness to support a delay in school start time. While improving
outcomes for students is the primary goal in this study, the surveys also identified a few areas
where faculty might also benefit from a delayed start time.
A Delayed Start Would Also Benefit Faculty
While the chief aim of the study was to ultimately benefit high school students, faculty
also identified personal benefits to a delayed start. When asked the extent to which they agreed
63

with the statement that a later start time would allow them to get more sleep, two-thirds of the
faculty respondents strongly agreed or agreed with that statement. The mean score was a 2.8,
indicating that, on average, faculty believed they, too, would get more sleep with a delayed start
time. In the open-ended responses about potential benefits, 28 educators identified potential
faculty benefits of a later start time. Of those, half listed more sleep for faculty as a potential
benefit. Additional responses included those who addressed the potential benefit of shifting some
work time before school, saying there would be “more time in the morning to grade and plan”
and there would be “more time to plan and collaborate.” Teachers also named the ability to have
meetings before school or meet with students for extra help as a potential benefit of the delayed
start time. Finally, three addressed other personal benefits, including a less rushed and stressful
morning schedule with their own children and the potential to exercise before school. Though a
delay in school start time would be implemented with the purpose of improving outcomes for
students, educators also saw potential personal benefits in such a change.
Skepticism on Actual Impact on Sleep Gains
While 85% of faculty and 90% of parents agreed in the quantitative portion of the survey
that a delayed school start time would lead to sleep gains for their students, there were open-
ended responses that question the efficacy of such a delay. One comment from a faculty member
posed a key question:
Students getting more sleep depends on their willingness to go to bed, [and] not think
that they can stay up even later or put off starting homework even longer because school
starts later. So what needs to be done to insure [sic] that a later start time efficaciously
brings about its goal?
64

Parents and educators asked similar questions about whether “the kids would use it for more
sleep” or “an excuse to stay up later.” Nine parents and more than 20 faculty members
included a comment about the actual sleep benefit of a delay, most worried that it would just lead
students to feeling like they could stay up later because they did not have to be up as early in the
morning. Responses included a discussion of other factors, beyond adolescent brain
development, that might impact a student’s sleeping patterns. Primarily from, but not limited to,
the faculty survey, these comments included a discussion of cell phone usage, social media apps,
streaming services like Netflix, and online gaming. This all serves to point to the fact that no
single factor can encompass all potential reasons for adolescent sleep duration. Beyond the
questions around students potentially choosing to stay up later, the study examined the potential
negative impacts on the students’ broader daily schedules.
How Can Students Fit It All In?
When considering the broader impacts of a start time delay on student schedules,
stakeholders questioned students’ ability to fit in all of their commitments if they delay the start
to the school day. Both parents and faculty were asked about the potential impacts of a delay on
students’ after-school athletics and extracurriculars as well as students’ ability to hold an after-
school job. On the parent survey, the mean score on the subscale for the two student schedule
questions was 2.0, showing a general disagreement that a delayed start would negatively impact
student schedules. On the question about after-school athletics and activities, 81% parents
disagreed or strongly disagreed that it would have a negative impact on those activities.
Similarly, 80% of parents disagreed or strongly disagreed that it would reduce their child’s
ability to have an after-school job.
65

The faculty survey included one question on student schedules that was not on the parent
survey. Before being asked about specific reasons, faculty were asked generally the extent to
which they agreed that they were concerned that a start time change would have a negative
impact on students’ schedules. They were then asked about negative impacts on extracurricular
activities and after-school jobs. The subscale mean for the three questions about potential
negative impacts on student schedules was a 2.3, slightly leaning towards disagreeing with the
assertion that a delay would have a negative impact on student schedules. However, the specific
responses on those three questions offered a clearer picture. When initially asked the general
question about a negative impact on students’ schedules, the mean response was a 2.0, or a
“disagree” that the start time delay would have a negative impact. However, when asked later in
the survey about specific ways a student schedule might be impacted, the responses indicated a
higher level of concern. Like the parents, faculty were also asked specifically about athletics and
extracurriculars, as well as after-school jobs. As shown in Figure 9, when asked the extent to
which they agreed that a start time delay would limit students’ time for athletics and
extracurriculars, 52% of faculty agreed or strongly agreed that it would limit students’ time for
those activities. Similarly, faculty were much more ambivalent about a student’s ability to hold
an after-school job. The mean on this question was 2.5, with almost 45% agreeing or strongly
agreeing that a delayed start would reduce a student’s ability to have an after-school job. The
ambivalence in the responses of the faculty implies the reality that any change in the academic
day schedule will have broader implications for the full schedule of commitments in a student’s
waking hours.

66

Figure 9
Faculty Perceptions of Negative Impact on Student Schedules

In their open-ended responses, faculty and parents addressed similar concerns about the
potential negative impacts of a start time delay on student schedules. The biggest concern,
reflected in 37 of the parents’ open-ended responses and 36 of the faculty responses, was that a
delay in start time would simply push the whole school day back, including athletics, mealtimes,
and homework. As one parent put it:
I think a later start time means later dismissal. With homework, extra curriculars,
studying and eating, there is already so little time after school to get it all done. I think
starting later will cause kids to stay up later finishing work.
This concern is present in many of the specific comments. Both parents and teachers were
concerned that a later start to the day implied a later end time, keeping all other scheduled
commitments the same. This would then push everything the student had to do later into the
evenings. In addition to the evening implications of start time delay, many expressed concerns
0%
10%
20%
30%
40%
50%
60%
Strongly
Disagree (1)
Disagree (2) Agree (3) Strongly Agree
(4)
Percentage of Responses
Faculty Perceptions of Negative Impact on Student
Schedules
Negative Impact on Students'
Schedules
Limit Time for Extracurricular
Activities
Negative Impact on Ability to Hold
an After-school Job
67

about the way students might be asked to use the morning times, potentially negating the benefits
of a start time delay.
Because students at independent schools are often highly involved in a wide range of arts,
athletics, and community extracurricular activities, some have started asking students to
regularly attend morning meetings and arts or athletics practices before school. When asked how
many days a week students are expected to arrive more than 30 minutes before the start of the
academic day because of extra curriculars, athletics, academic support, or other personal factors,
parents reported that more than half (56%) of the students need to arrive early at least one day a
week. Forty-seven of the 163 (29%) students described a need to arrive early at least three days a
week. In their responses to the open-ended questions, both faculty and parents discussed the need
to be thoughtful about how additional time before school is used. One parent noted that an “8:30
start time may result in schools scheduling activities before school” and another discussed how
the “official” start time is not the actual start time because many students are expected to be there
early “due to athletics, clubs, extra help, etc.” The faculty surveys also discussed the challenges
and temptations of filling time in the mornings.
The faculty addressed the specifics of scheduling activities in the morning. One
suggested that while a later start opens the possibility of morning help sessions with students, it
would negate the sleep benefits of a later start time. Additionally, the comments addressed the
way the athletics departments tend to use the time before school. One wondered if the athletics
department “would be required to push their morning workout routines ahead to accommodate
the later start time” and shared their personal experience that when the school has late start days,
athletics practices still run at their normal “early” times. Another suggested that schools impose
an earliest start time allowed for sports practices and noted that in the “nonstop race toward
68

college,” students are required to have such a broad range of commitments that they look to fill
all possible times with activities, including before school. While the faculty surveys reflected
concern about the potential that the school day would run later and that everything would be
extended, they seemed similarly concerned about the possibility of coaches and teachers
expecting students to use that time for things other than extending their sleep duration. In
addition to challenges for the students with the myriad expectations during the school day, a
change in a student’s schedule also has potential ramifications for the entire family.
Parents Report They Anticipate Minimal Impact on Family Time
Because high school students do not operate in a vacuum, school start time delay
potentially impacts the entire family. Generally, the parents did not feel a change would have a
significant impact on the family. Parents were asked a series of questions about the potential
negative impacts of a school start time delay on the family. They were asked to what extent they
agreed with the possibility of a delayed start time interfering with mealtimes, negatively
affecting student transportation plans, and interfering with their own work schedule. The mean
score for the family impact subscale was 1.8. This reflects the general feeling that they did not
believe that a start time delay would have a negative impact on the family in these ways. As
presented in Figure 10, only 9 (7%) parents believed that a delayed start might potentially
interfere with mealtimes. Slightly more parents expressed concern about the possibility that a
delayed time would negatively affect student transportation. About 15% of the parents surveyed
agreed or strongly agreed that the change would have a negative impact on their child’s
transportation. When the questions shifted to ask about negative impacts on the parents
themselves, a slightly higher percentage agreed with the statements. About 20% of parents
agreed or strongly agreed that a change in start time might interfere with their work schedules.
69

While this numerical data does reflect a significant potential challenge, it is worth noting that for
each of the four statements, between 80% and 93% percent disagree or strongly disagree with the
potential negative impacts of a start time delay, which is reflected in the mean “disagree” in the
subscale score. Though the numerical data seems to indicate that generally the impacts of a start
time delay would be minimal on the family in general, a number of specific concerns were
addressed in the open-ended responses to potential challenges to meeting the 8:30 a.m. start time.

Figure 10
Parent Perceptions of Impact on Family Time

0%
10%
20%
30%
40%
50%
60%
Strongly
Disagree (1)
Disagree (2) Agree (3) Strongly Agree
(4)
Percentage of Responses
Axis Title
Parent Perceptions of Impact on Family Time
Interfere with Mealtimes
Negatively Alter Transportation
Plans
Interfere with Parent Work Schedule
70

Parents expressed concerns about potential challenges including the impact on family
schedules, siblings’ transportation, and their own work schedules. Because of the potential
impacts on overall family schedules, eight parents mentioned that, for a variety of reasons, their
child would not be able to benefit from a start time delay. One specific challenge mentioned
multiple times was the potential that different divisions within an independent school would have
different start times. One parent noted that “if my younger children’s division begins earlier, my
high schooler is[n]’t able to take advantage of the later start time.” This implies that families
would expect to make one drop off at the school, regardless of the ages of their children, or that
driving-age teens would be expected to bring their siblings to school. The single biggest obstacle
within the family schedule was the expected work start time for parents. About 20 parents
mentioned that a delay would present a challenge for them to get their kids to school and then to
work on time. Three specifically addressed the fact that their child would have to get to school at
the same time as their current schedule in order for the parent to meet their work obligation. One
shared that their “child would still need to be dropped off at the same time in order for [them] to
get to work in a timely manner.” Others shared that while it might impact their work schedules,
they “will work around it.” Additionally, though it would not impact them directly, two parents
addressed the potential negative impacts a delay might have on coaches and other school
employees involved in extracurricular activities.
Faculty Do Not Believe a Change Will Negatively Impact Their Schedule
Eighty-four percent of faculty and leadership surveyed reported that they did not think a
delayed start would negatively impact their schedule. Like the parents, the educators surveyed
were given space both in quantitative survey questions and open-ended responses to address the
potential negative impacts on their own schedules. Faculty were asked the extent to which they
71

agreed that a delay in start time would negatively impact their schedule, affect their after-school
extracurricular commitments, and interfere with their childcare plans. As shown in Figure 11,
just under half of the participants, or 85 individuals, in the faculty and leadership survey
responded to the question about a delay negatively impacting childcare. Of those respondents,
only about 22% agreed or strongly agreed that it would have a negative impact on their childcare
plan. For the 149 participants who responded about their after-school commitments, about 40%
believed that the delay in start time would have those commitments affected. These concerns
were also reflected in the open-ended responses that discussed potential challenges for faculty in
a start time delay.

Figure 11
Faculty Perceptions of Impact on Their Own Schedules

Note. For the questions asking about extracurricular commitments and childcare, participants
were given an option to select “Not Applicable” in the section that asked about extracurriculars
0%
10%
20%
30%
40%
50%
60%
Strongly
Disagree (1)
Disagree (2) Agree (3) Strongly Agree
(4)
Percentage of Responses
Faculty Perceptions of Impact on Their Own Schedule
Negative Impact on Faculty
Schedule
Affect Extracurricular Commitments
After School
Interfere with My Childcare Plan
72

and childcare, because not all respondents might have those commitments. Figure 11 reflects the
percentage of responses to that prompt, not the percentage of total participants.

Faculty addressed the concern for their own schedules, that similar to student schedules,
their commitments would simply be pushed back later into the afternoon and evening. Ten
responses addressed the potential challenges of delaying everything and getting off of work later.
One who discussed that concern said that “most of my job is done after school. This would
trickle down to effect [sic] my family.” Another discussed that a potential challenge would be the
“scheduling of activities for both students and my own family obligations.” They also expressed
concern that the school day might go later, but they would still be expected to arrive at the
previous time, particularly if faculty were “required to do morning supervision duty” for students
who needed to arrive early because of their parents’ commitments. Additionally, five educators
expressed the desire to allow the start time delay to actually be a delay for them as well, instead
of causing “administration to add more faculty meetings to our schedule resulting in less sleep
for me.” While they overwhelmingly support the delay in start time generally, faculty did not
seem interested in that delay leading to a longer workday for them.
Beyond the length of their workday, faculty also raised the potential challenge of the loss
of instructional minutes. While some were concerned about the day being pushed later, three
specifically addressed their concerns that if the start time was later and the end time remained the
same, they would lose teaching time. One teacher expressed “some concern about lost
instructional time if the school day ends around 3:00 p.m.” Like the responses on the parent
survey, three faculty responses also noted the challenges of different start times in multiple
73

divisions within the school, as well as the potential challenges of managing the schedules of their
own children in other divisions or at other schools.
Difficulty Anticipating the Impact on Traffic Patterns and Commute Times
While many expressed their feelings about the potential impacts of a delayed start time
on traffic and commute times, there was little consensus about how exactly traffic patterns might
be impacted. While this issue is tied to parent work schedules and commute times, it is being
addressed independently because of the broad range of feelings about how a delay in start time
might impact traffic patterns and commutes. Because independent schools tend to draw students
from a wider area and have more limited bus systems than public schools, the duration of one’s
commute becomes critical to the school day. Parents and faculty were both asked to estimate
their normal morning and afternoon commutes. As shown in Figure 12 and Figure 13, those
commutes ranged from less than ten minutes to more than an hour for faculty, with 94% having a
commute of less than 30 minutes in the morning and less than 40 minutes in the afternoon.
Similarly, the parents reported that about 90 % have less than a 30-minute commute both in the
mornings and afternoons.

74

Figure 12
Morning Commute Times

Figure 13
Afternoon Commute Times

0%
10%
20%
30%
40%
50%
60%
10 min
or less
11-20
min
21-30
min
31-40
min
41-50
min
51-60
min
More
than an
hour
Percentage of Responses
Morning Commute Times
Student Morning Commute
Faculty Morning Commute
0%
10%
20%
30%
40%
50%
10 min
or less
11-20
min
21-30
min
31-40
min
41-50
min
51-60
min
More
than an
hour
Percentage of Responses
Afternoon Commute Times
Student Afternoon Commute
Faculty Afternoon Commute
75

Figure 14
Perceptions of Impact on Commute Times

Note. There were only 145 faculty responses, short of the 173 total participants, primarily
because one of the schools surveyed is a boarding school where many faculty members live on
campus.

Both stakeholder groups were asked the extent to which they agreed that a delayed start
time would interfere with their commute times. The mean response on the parent survey was 2.0,
and it was 2.1 on the faculty and leadership survey. As noted in Figure 14, on both surveys about
75% of respondents disagreed or strongly disagreed with the idea that the delay would interfere
with their commute. However, when given the opportunity to offer comments about potential
benefits or challenges of school start time delay, the discussion of traffic and commute times
appears regularly on both the challenges and benefits comments.
When addressing potential benefits and challenges, similar numbers of parents and
faculty identified potential traffic and commute benefits as did those who saw potential
0%
10%
20%
30%
40%
50%
60%
Strongly
Disagree (1)
Disagree (2) Agree (3) Strongly
Agree (4)
Percentage of Responses
Perceptions of Impact on Commute
Interfere with Parent Commute
Interfere with Faculty Commute
76

challenges to traffic and commute timing. Twelve parents and 11 faculty identified specific
benefits of a later start time. Those comments included more school days where driving would be
in the daylight, as mentioned in the discussion of student safety, as well as lighter traffic. One
parent noted that, “When I did drive my kids on late start days in the past it was nice that most of
the traffic had already passed. It was a much easier commute to school with less traffic.” A
faculty member noted that with a later start time, “commuters will avoid public school buses and
bus stops.” A number generally pointed to lighter traffic when there was a later start time.
However, this seems fairly unscientific, as almost the same number cited traffic and commute
issues as a potential challenge.
When discussing potential challenges with a delayed start time, 13 parents and 12 faculty
members cited the belief that increased traffic and commute times would present a challenge.
Most generally noted that with a later start, traffic is worse as more people are trying to get to
work. One faculty member noted the risk of putting “student drivers in the thick of Metro ATL
rush hour” and another asked, “Have you seen metro Atlanta traffic around 0830 and 1600???”
and then asserted that “commute times for faculty will increase.” One parent noted that “traffic is
far worse at 8:30 than at 7:30.” Another parent put it this way:
My older daughter drives herself and it takes a while with traffic. I think even if school
started later, she’d like[ly] still have to leave our house at the same time to arrive in time
for school; otherwise, she’d just end up spending that extra ½ hour in traffic.
Finally, one parent noted that “people have already forgotten their pre-covid commutes, and how
much time was built into them.” The implication here is that because there are fewer people
commuting to work this school year, people may not remember how long their commutes took
77

before this school year, and as a result they may not see this as the challenge it might be when
things return to normal.
Unlike the other data collected there is no clarity offered in the discussion of traffic.
Stakeholders feel strongly that a delay in start time would have an impact on their commute time
and the traffic that they may encounter, but there was no clear consensus on what that impact
might be. Both the respondents who saw the change as a commute benefit and those who saw it
as a challenge projected confidence in their responses. The fact that it is so evenly split in the
open-ended discussion indicates that there is ambiguity about how traffic patterns might be
impacted, but many were confident that there would be an impact in some way.
Summary
When the participants were asked to anticipate potential benefits and challenges
associated with school start time delay, their responses were wide-ranging. Parents and faculty
expected that a delay in start time would have a benefit to students’ overall health, stress levels,
happiness, and academic performance. They anticipated challenges in the way a delay might
impact the students’ daily schedules as well as their own. Stakeholders anticipated that with a
start time delay, students would have the intended benefit of increased sleep, but they also
worried about myriad potential obstacles to actually increasing student sleep duration as well as
the potential schedule ramifications of a change. They were also deeply considerate of the idea
that traffic patterns would impact their commute times, though there is no consensus as to
whether that would lead to a benefit or a challenge. To address the conversation on anticipated
benefits and challenges in experience, the next research question seeks to address the experience
of educators at schools that have already made a delay to meet the AAP’s recommendation of an
8:30 a.m. or later start time.
78

Research Question 4: How Has Delaying School Start Time Impacted Key Stakeholders in
Schools That Have Changed Time to Meet AAP Recommendations?
In order to address this research question, interviews were conducted with school leaders
at schools with start times that met the AAP recommendation for 8:30 a.m. or later. Additionally,
37 respondents to the faculty survey identified that they were at a school with a late start time, so
the open-ended survey responses from those faculty participants also provide meaningful
insights toward addressing this research question. Like any decision in a large organization, the
specific context of the change in start time was different at the two participating schools. The
interview participants shared a wide range of perspectives that addressed the comments shared in
the anticipated benefits and challenges presented in the previous section. In each interview,
participants were given an opportunity to discuss the specific context of their school setting, the
reasons for initiating a change process, the change process itself, the anticipated benefits and
challenges of a start time delay, and the outcomes they have seen since making a change.
Additionally, participants were asked to offer any advice they might have for schools looking to
pursue such a change, or if, upon reflection, they would do anything differently in their own
process. While the participants had a variety of perspectives on the process as well as the
outcomes, they uniformly supported the change and did not express any desire to go back to the
previous start times.

79

Table 2
Narrative Table of Interview Participants
Interviewees
(Pseudonym)
School (Pseudonym) Position Years at current
school
Allison Dunhill Academy Director of Student Support 10-15 years
John Dunhill Academy Dean of Academics 25-30 years
Richard Dunhill Academy Director of Faculty
Development
10-15 years
Sam Berwick School VP of Finance and
Operations; Executive
Director of Online Academy
10-15 years
Melissa Berwick School Director of Academics More than 30
years

School Context and Background
Both Dunhill Academy and Berwick School are well-established, large, faith-based
schools in the metro-Atlanta area. Both schools have student populations between 1,500 and
2,000 students and include elementary, middle, and high school divisions. In addition to similar
demographic profiles, interview participants addressed the academic rigor of the institutions.
Ahead of the start time delay, both schools had similar schedules, with an 8:00 a.m. start with
one “late start” day per week.
Interview participants also addressed the traditional nature of their academic programs
and schedules. Allison described Dunhill Academy as “rigorous, fast paced, [and] high energy,”
while John described the students and faculty as “bright and curious, motivated, pretty hard
driving, [with] high expectations for themselves and for others.” When talking about the
traditional nature of the school and their schedule, Allison mentioned that they had employed the
same schedule “probably even for as long as 30 or 40 years.”
Similarly, Melissa addressed the academic focus of Berwick School. She shared that “we
have a really strong, rich academic component to what we do,” and that “rigorous education is
80

not something that we think has to be set aside from a [religious] education.” Sam addressed
some of the traditional nature of Berwick School, noting that in 50 years, they were only on their
third head of school, which reflects the school’s stability. He did go on to share that while he
perceives that many independent schools are “averse to change,” at Berwick School they are “a
little bit different…in that we do push and we do try to lean into change.” He continued, “We
want to be obsessed with finding a better way, and we want to be yes before no.” While on paper
the schools appear similar, there might be cultural climate factors that influenced the process and
potential outcomes.
Reasons for Initiating the Change Process
When Dunhill Academy changed their start time, it was in the context of a much larger
schedule overhaul. In addition to delaying school start time to 8:30 a.m. four days a week and to
9:15 a.m. on Wednesday, they also fundamentally shifted the school-year academic calendar as
well as the daily schedule. John shared that:
There are two big goals that we had. One was to reflect the current research about sleep
and learning and basically create better, greater balance in the lives of students. And then
the second goal was to align with … a strategic plan goal around teaching and learning.
To accomplish these goals, leaders at Dunhill Academy examined a wide variety of neuroscience
and learning research to set up more effective school days for their students. Allison shared that
“there was just a renewed effort to look at what science says is the best way to educate kids and
build something responsive to that.” She continued that that included a “host of things like fewer
class periods per day…, longer class periods…and less homework to prep for the next day.” She
added that “related to that was just the science around teenage need for sleep.” Richard shared
that the school spent time “looking at some of the neuroscience research that informs how we
81

should teach and how students learn, and that along with also understanding that…kids really
could benefit by starting school a little later.”
In addition to the delayed start time, Dunhill Academy employed a schedule that shifted
from six periods per day to four. Also, though it is not particularly relevant to this study, the
other significant change they made was the addition of a winter term where high school students
spend three weeks in classes designed by faculty across departments to “explore new and
different ideas.” While start time was one part of the desire for an overall schedule change at
Dunhill Academy, the lengthening of the individual classes, the reduction in number of classes
per day, and the addition of the winter term were all significant factors in moving to a new
schedule.
At Berwick School, the reasons for pursuing a change were much more focused on the
outcomes for students and families around sleep and school start time. Melissa shared that “we
had noticed and been listening to our families, as the rat race of life was becoming unbearable for
them.” She continued, “What they were giving up to accomplish everything they felt they needed
to accomplish – and I say that because it was a choice to be heavily involved in everything – they
were giving up sleep.” Melissa shared that they first tried the one late start day, but that proved
not to be enough. She added that contributing to the burden was the fact that many families were
“having to leave at 6:00 a.m., 6:30 a.m. to get to school on time with an 8:00 a.m. start.” When
Sam discussed the reasons for examining a change, he also cited the traffic, noting that their
“average family lives eight miles away, which, in Atlanta traffic, means 40 minutes to an hour.
In addition to the student loss of sleep because of commute time, Sam also addressed the “lack of
sleep because of increasing schedules and workloads.” At Berwick School, the driving factors
82

for the change were to ease the burden on overcommitted students by allowing them to sleep a
little more and spend a little less time in traffic.
Dunhill Academy and Berwick School were both hoping to improve student health and
learning outcomes for their students. Dunhill Academy’s was part of a much wider reaching
schedule overhaul and Berwick School was specifically aiming to address issues around student
sleep loss as a result of the schedule and commute times. The change in start time had the
potential to impact all stakeholders, including the students, families, faculty, and the broader
school community.
Impact on Students
The primary goal of this study was to benefit students by engaging the stakeholders that
make decisions on their behalf. When considering how the change in start time impacted
students, the interview participants primarily discussed gains in student sleep, how students felt
about the delayed start, and the way the schedule change impacted their schedules more broadly.
While leaders felt as though students have benefitted from the additional sleep, they also
addressed the way the schedule change impacts other areas of student life, including
extracurriculars before and after school.
Leaders Observe Positive Sleep Outcomes for Students
Interview participants from both schools believe the change has led to positive sleep
outcomes for their students. At Berwick School, Melissa shared that even though some were
skeptical about the impact of a 30-minute delay, “We’ve seen that it made a huge difference.”
She continued saying:
83

A five-day 8:30 start made the difference. We heard, and you could feel it in all the
surveys that we gave, there was a decided difference in approaching life and how they
were able to get the rest they needed and do the things they needed to do.
Similarly, Allison shared that “I think they really liked it, and I really do think it benefitted the
kids.” John elaborated on his observations at Dunhill Academy, saying, “I do think kids are
better rested…They have lived in this schedule for so long they don’t know anything else.” He
also addressed that even though he has seen benefits over time, students are in an age where
there is a rise in mental health challenges across the board. He continued that in “the first year or
two or three of the schedule change, where we still had kids who remembered and had lived the
old schedule, the feedback was super positive. We can’t imagine going back.” Sam, who looks at
some of the school-wide data collected also addressed the fact that in the years since his school
made the change, student responses about stress levels have “continued to improve over that
timeframe.”
In their open-ended responses about benefits and challenges, about two thirds of the
faculty who have a late start addressed student sleep increases, rooted in their experience. One
shared that “most of my students are able to get additional sleep because of the later start,” and
another added that they have no concerns because “we start at 8:30 – it’s the best thing we ever
did.” Another shared that “it’s been great for our school, and we’ve established healthier
rhythms.”
While the response of the participants was generally positive, they did also address
challenges from myriad competing factors. One interview participant shared that they are not
sure the change has fully had the desired effect of increasing student sleep because of those
external factors. They shared:
84

So what people did, which I think is part of what happens in a fast-paced community
where people are always wanting to sort of get 1% better every day, is that things like
sports teams then just started having morning lifts and clubs starting to meet in the
morning. We tend to have kids who do lots of extra activities. They’re in multiple clubs,
they play multiple sports, they may play a year-round sport plus a school sport, and so
instead of letting it be a way for people to get more rest, it was almost like the
community, as a whole, took it as a way to put more into a day. When you have kids
showing up at 6:00 in the morning to lift weights, they’re certainly not getting extra
sleep.
Because the milieu of the independent school is one of highly driven students looking to engage
in myriad academic and extracurricular activities, the participants all had to address the broad
range of student commitments as they looked to make their schedule changes.
Schools Must Consider the Impact of a Student’s Full Schedule of Commitments
When delaying school start time, the surveys and interviews indicated that stakeholders
expect a school to look at the impact on a student’s schedule to include the student’s full
commitments before, during, and after school. One of the primary concerns that parents and
faculty had in their survey responses was about the potential impact that a start time delay might
have on the time at school before and after the school day. As the participant previously quoted
addressed, many at their school have seen the delay as time that can be used for something
deemed productive, rather than as protected sleep time. Looking beyond the scheduled school
hours, Melissa shared that while “academics are important, and what we do here is important,”
the school also had the responsibility “to look at the whole child and the whole family” to
consider if the full expectations of the schedule are “limiting their ability to rest appropriately.”
85

Participants from both schools addressed the time both before and after school. Both Sam
and Allison shared that one meaningful outcome of the time before school was that it had created
space for students who chose to come in early to see teachers for extra help. Allison also shared
that “giving teachers that protected time felt really good to the teachers.” She continued, “They
weren’t having to stay ‘til 4:00 when the workday had actually ended at 3:00 to do a meeting.”
Furthermore, both parents and faculty addressed in the surveys the fact even at their schools that
did not have a late start, teams were already having practices before school. One parent discussed
this in their survey response, saying that they do not think people realize that “students are
required to come to school well before the ‘official’ start time due to athletics, clubs, extra help,
etc. The official start time is not really the reality of what most high school students experience
on most days.” A start time delay that also pushed back practice start time before school would
be a benefit, even if students still needed to arrive to school earlier than the official start time.
While there are other positive uses for the time before school, participants expressed the feeling
that sleep should be the primary use of the time and examined ways to not impact the end of the
day.
At Dunhill Academy, the schedule change did not fundamentally shift the school day end
time, so, as a result, the practices and athletics contests still generally ran on the same schedule
and did not shift back any later. In order to combat a potential delay in those things at Berwick
School, participants shared that the leadership implemented a policy that all practices, arts and
athletics alike, were required to end by 6:00 p.m. as part of an overall school policy of limiting
time spent on campus and improving outcomes for families. As reflected in the survey responses
of potential challenges to a delay, faculty from schools that have made such a delay generally
indicated that it is important for a school to consider the overall amount of time that they are
86

asking students to be on campus and to engage the athletics and arts departments so that all are
working together to improve outcomes for those students. While positive student outcomes are
the primary goal in such a change, the change also has an impact on the whole family.
Positive Parental Feedback
When reflecting on the feedback they have received from parents throughout the process,
all interview participants indicated that the feedback they had received from parents was
positive. The unanimous response about positive parental support was noteworthy, given the
candor with which the interview participants addressed the challenges raised by faculty. When
asked if there had been any pushback from the parents since making the change, the responses
included, “It’s really a non-issue,” “I think the parents have been really happy with the schedule
itself,” and that parents are “100% in support of” the change the school made. Allison elaborated
that:
The second you say to parents, “We think teens are stressed. We think teens don’t sleep
enough. We think teens have heightened anxiety,” you’re gonna have very few parents
who fight you on those fronts. And then when you’re saying that “Research says an extra
hour of sleep can be transformative,” you’re going to have very few parents fight you.
Sam shared similar insight from his experience at his school. He described some of the annual
school surveys that track overall parent satisfaction, likelihood to recommend the school, and
leadership quality. While he acknowledged that the delay in start time was not the sole driver of
these metrics, he thinks the change has led to an increase in those scores in the years since they
made the change. He also shared an anecdote of a parent who expressed that one of the reasons
they chose the school was the letter that the leadership sent making the change in start time. He
said, “What it spoke to her was [that] we were grounded in the research, and the decisions that
87

were made weren’t simply decisions that were flashy, but actually grounded in the research.”
The interview participants were unanimous in their perception that parents have supported them
in the change throughout the process.
While the goal of a start time delay is primarily to improve outcomes for the students, a
school schedule change for the student has the potential to impact the family in terms of how the
high school schedule fits with parents’ jobs and siblings in other divisions or in other schools. In
the surveys, parents addressed the potential challenges associated with the need to get their
children to school and then to their jobs. The interview participants, when addressing this
question, did not find that start time delay had posed a significant challenge to families. Sam’s
general observation was that the main feedback they received was that “the timing was better for
families.” In his perspective, this meant that help sessions could be done exclusively in the
morning and they were “not having to start at 7:00 a.m. or 7:30 a.m.” Now they were “able to
start at 7:30 a.m. or 8:00 a.m. and their teachers were not having to leave to go to a practice or
here or there.” Allison also specifically addressed the idea that students might still need to arrive
on campus at their previous arrival times because of parents’ jobs or siblings in other divisions.
She shared, “We have lots of places in the mornings kids can go sit and do work.” None of the
leaders indicated that any of the major concerns of the parents in the surveys impacted those
stakeholders in the negative ways they anticipated. One major concern raised in the survey
responses, which the leaders at Berwick School did address specifically, was traffic.
While the response about the impact of traffic was mixed in the surveys, with many
anticipating it as a benefit and many sure that it would present a challenge, the experience of
Berwick School indicates that the delay has improved traffic and commute outcomes for
families. Melissa shared that “we didn’t want families to continue to feel so rushed and like they
88

were spending all their time sitting in traffic to get here.” Sam also shared that “we were trying
to find ways to help with traffic, find times when traffic patterns were a little bit better.” While
reflecting on the change, he shared that “in Atlanta, with traffic, it helps.” Speaking generally
about the stress of waking up, getting out of the house, and traffic, Melissa reflected that it has
“relieved the burden of the morning from our families.” When sharing the feedback they had
received from parents, no participants mentioned a negative impact of the traffic patterns as a
critical response from a parent.
Faculty Are Resistant to Change and Concerned About Instructional Time
The Administrators interviewed reported that school start time delay did not
fundamentally change the length of the workday for faculty. In line with the concerns of the
faculty in the surveys, the interview participants addressed how the change in start time has
impacted faculty. In both schools, faculty are still generally expected to arrive to campus about
the same time they were expected to before the start time delay. At Dunhill Academy, this allows
two days a week for small group faculty collaboration, two days where they are available for
extra help sessions, and their one later start day where they use the time for faculty meetings.
This means that their workday ends after school. As a result, Allison believed the faculty “felt
seen, [and] valued.” Similarly, Sam discussed how the delay allowed space for students to come
in for extra help while still being able to leave home at a reasonable time.
Interview participants also addressed the concern raised in the surveys about the loss of
instructional minutes. As Melissa said, “either you are gonna have to shrink your instructional
time or you’re gonna have to get out later.” While Berwick School did shift the official school
day a little later, the hard-stop at 6:00 p.m. for campus activities meant that instead of losing
instructional minutes, they ended up reducing the length of after-school practices for
89

extracurriculars. Melissa continued that what shrunk, “was the practices in the afternoon, and I’m
not certain that shouldn’t have shrunk anyway. I think when kids go to practice at 3:30 and get
out at 6:30 or 7:00, that’s a pretty lengthy practice.” In addition to the delayed start, Dunhill
Academy was facing a larger impact to their instructional time as a result of the three-week
winter term as well. This combination proved particularly stressful for AP teachers with a limited
time frame to cover a broad curriculum expected by the College Board. While this was stressful
for the faculty, John reminded them that “we all take the exam at the same time, but schools in
New England start often after Labor Day and we’re starting August 15
th
. So, it just felt like we
could do it.” Richard also addressed this topic, saying, “AP teachers have … a little harder time
with it because losing three weeks into the winter term was challenging for them. They gained
some time back with longer classes, but they lost time with the winter term.” Another step they
took to manage instruction minutes, which one interview participant referred to as “sort of
sleight of hand,” was to eliminate transition times between classes. While students do still have
to walk between classes, this change operates under the theory that “when you give them five
minutes, everyone takes 10 anyway, so give them none and they’ll get there faster.” The ideas
shared by the school leaders who participated in the interviews are also reflected in the feedback
received on the survey from faculty who currently have a delayed start.
The survey responses from faculty who already have a later start time presented similar
insights about the length of day, use of morning time, and instructional minutes. One noted that
“I still came to work at 7:15 when we changed from an 8 start time to an 8:30 start time. It may
help kids, but it didn’t change much about my own schedule as a teacher.” Another addressed
this, pleading with their administration through the survey to “let the teachers come in later, too.”
In line with the interviews, one faculty survey response highlighted the increased time for
90

meetings and for students to meet with teachers before school. On the topic of actual teaching
time, one faculty member shared that the “biggest loss is in instruction time. We start at 8:30
a.m. and it just eats a little into the instruction time that we have, because we are still out at
3:10.” These responses from the teachers reflect the perceptions of the school leaders’
interviews, and on the whole, do not reflect a significant concern about having made a shift to a
later start time.
Leaders from both schools addressed the idea that they felt their faculty were relatively
averse to change and rooted firmly in what they had been doing. By talking them throughout the
process and leaning on the research, the interview participants felt as they were able to bring the
faculty along, even if there was some resistance at first. Allison addressed this directly, within
the context of longer class periods, saying, “We explained it to the faculty and staff. We gave
them time to do the education, and then they had the spring to really ramp up.” Melissa also
addressed getting faculty on board in her discussion of the observation data from their first
period classes. She said, “There was a decided difference in productivity outcomes. And so,
when they were able to see that, they all got on board.” John shared that “I think on the whole,
the folks are happy. There’s still math and science [chuckle]. They might like to go back still, a
few of them.” He continued that “I think what people would say is, this is on the whole better for
students, even if there are certain sacrifices that I made in my own discipline or department.”
Similarly, Melissa expressed that “I think if we went back to an early start time now, people
would be like, ‘What? No way’ ‘cause I think it also provides time for our faculty to not have to
get up so early and not have to be here so early and that sort of thing.” While they acknowledge
the challenges and benefits to their faculty, all of the school leaders believe that it has, on the
whole, been a positive change for the faculty. While the interview participants discussed the
91

impacts of the change on the students, parents, and faculty, they also brought up the way it
impacts the broader school community, particularly in independent schools with multiple
divisions.
Change Impacts on Multiple Divisions
Like many of the traditional AAAIS schools in this study, both Dunhill Academy and
Berwick School contain elementary, middle, and high school divisions within the same school.
One of the central concerns reflected in the surveys was how a change in high school start time
would impact families and faculty members with children in multiple divisions. The two schools’
approaches to start time delay for the high school were drastically different. At Berwick School,
they made the change for all divisions at once. While this did have the effect of delaying the end
time slightly in the elementary schools, leaders in each division were tasked with studying the
benefits and challenges and collaborated to make a school-wide decision. At Dunhill Academy,
the schedule change process was exclusively conducted by the high school.
Dunhill Academy’s high school operated independently in their schedule change. Richard
shared that there was “no pressure and push to get a more coordinated effort across the
divisions.” When discussing this topic, John reflected that at least in terms of aligning with the
middle school, their high school acting independently has added some challenges. These
challenges were expected, but he expressed that if he were to do it again, he would likely try to
align the middle and high school schedules. As a result of an independent high school change,
they lost the ability to have cross-over teachers, who teach in multiple divisions. This has a
significant impact from a budget perspective, as they previously had teachers in subjects like
Latin and Mandarin that could teach in both divisions and as a result of divesting the schedules,
they had to hire additional faculty. Another challenge created by not aligning with the middle
92

school was that a number of shared spaces, specifically the dining and arts spaces, are now
significantly more difficult to schedule and share.
While these two schools took totally different approaches to aligning a school with
multiple divisions, all of the participants addressed the need to at least consider the school-wide
impact when making a change. A change in one division does impact other divisions. For
Dunhill Academy, two of the three participants addressed this challenge as the most significant
thing that they would address differently undertaking a new schedule study. Being a
multidivisional school means individual families might be impacted by multiple divisions.
Advice for Other School Leaders
When asked to offer advice to others about making a change in school start time and if
there was anything they would do differently if they had to do it again, four of the five leaders
discussed specific elements of the process that they would suggest other schools lean on. Those
leaders discussed the importance of process and the need to root the change in the data, either
from outside research or from data collected within the institution. Allison spoke specifically to
the need to present a clear rationale for the decisions being made and the importance of buy-in
from the key stakeholders. She said:
If you’re gonna do it, do it, and really advertise, acknowledge, celebrate, and push the
rationale, and on the front end do everything you can do to get buy-in for the concept and
the rationale behind it, and then probably in the execution it should be pretty seamless.
Richard suggested “take your time. Do a lot of research.” Elaborating on the process
components, John spoke to the positives of including students in the process and doing site visits
at other schools around the country. Melissa said she would not do anything differently, as they
“actually went through all of the roles to give everybody a voice and buy-in, and we listened and
93

we collected data and we didn’t just do it on a whim. I think we did everything very thoughtfully
and carefully.” She offered that she would advise schools that even though “academics are
important and what we do here is important,” it is equally important “to look at the whole child
and the whole family, and if the schedule that you’ve got in place is limiting their ability to rest
appropriately, you need to seriously consider if that’s the wisest path.” What these leaders shared
in common was about the depth of their process.
While other leaders spoke to the details in their process, Sam had a strikingly different
perspective. He shared that at times he worries “that schools make too big of an issue out of
something that’s not that big a deal.” While he shared the others’ interest in presenting the
research and rationale, he suggested that being direct with clear communication is the best way
forward. He even specifically said that this issue is not one that requires “all kinds of focus
groups” or the gathering of teacher and parent committees. He suggested that schools consider
the potential challenges and engage with a few leaders in the faculty and parent groups, but that a
larger process was unnecessary. He ended the interview saying, “It (start time delay) makes a
difference. And so, for crying out loud, it ain’t that big a deal. Just do it…You can use that
colloquial “ain’t” within there too. I did that for emphasis.” Sam’s perspective is reflective of the
quantitative data showing more than 90% of stakeholders in favor of a start time delay. While the
others spoke to the typical independent school process, Sam spoke like a leader who would look
at the data and make a reasonable change that the data suggests, without the traditional
independent school change process.
In summary, the interviews, first and foremost, discussed the observed benefits for the
students of a delayed start time. They also all indicated that despite potential logistical
challenges, they faced almost no pushback from parents and felt like they were able to
94

reasonably mitigate the concerns of the faculty. Finally, each interview offered meaningful
perspectives and suggestions, including being mindful of the way students are asked to use the
time before and after school, as well as how one division’s schedule change might present
logistical challenges for other divisions within the school. On the whole, even when the
participants discussed the lengthy process each school went through to make a change in start
time, Sam’s words seem to be an undercurrent in all of the conversations about the barriers to
school start time delay. “For crying out loud, it ain’t that big a deal. Just do it.”
Summary
In summary, the findings of this study indicate that, in general, parents and educators
estimate adolescent sleep needs to be near the minimum recommendations from the NSF and
AAP. They do not think their students are getting enough sleep, and they believe that the AAP
recommendations are reasonable ones. More than 90% of the just over 300 teachers and parents
who participated in the surveys would support a school start time delay to meet the AAP’s
recommendation of 8:30 a.m. or later. The survey participants anticipated a wide range of
benefits and challenges for students. Those benefits primarily focused on student outcomes of
increased sleep, improved health, and reduced stress. The challenges included potential impacts
to the students’ full daily schedules and the ways a change might affect parents and faculty.
Though the participants anticipated an array of potential benefits and challenges, they can be
viewed through the lens of overwhelming support for a start time delay.
Supplementing the surveys, the interviews demonstrated the positive outcomes for
students and reflected many of the same considerations addressed in the surveys. The schedule
changes at Dunhill Academy and Berwick School were rooted in the research on adolescent
sleep as well as their own observations within the school community. While each change
95

required problem-solving a range of potential challenges, the experiences each participant
discussed indicated that the outcomes of the start time delay had been a benefit to their students.
While their interviews might help other school leaders avoid potential pitfalls, they would all still
make the same change today if they had to do it over again.
Both the quantitative and qualitative data in the study indicate that the key stakeholders in
independent schools, the parents and educators, generally understand adolescent sleep needs.
They know their kids are not getting enough sleep. They anticipate myriad potential benefits and
challenges, and they are ready to make a change to reflect the current research and best practices
around adolescent sleep and school start time. Finally, those who have made a change affirm the
benefits and identify the challenges as manageable.

96

Chapter Five: Recommendations
This chapter presents a discussion of the key findings of the study, situating those
findings within the context of the literature review and Bandura’s (1988) social cognitive theory.
The findings, which indicate support for school start time delay in AAAIS schools, are then used
to make recommendations for school leaders looking to make a change in their school’s start
times. The intention of this study, along with the subsequent recommendations, was to provide
school leaders with the relevant literature around school start time, the findings of parents and
educators in AAAIS schools, and a set of recommendations for the change process. Taken
together, this study will help school leaders enact a start time delay in their schools to meet the
AAP recommendation of 8:30 a.m. or later.
Discussion of Findings
The most significant single finding in this study is that parents and educators are
overwhelmingly in favor of school start time delay that meets the AAP recommendations. More
than 90% of the parents and educators surveyed responded that they would support their school
delaying start time. As a result, school leaders can frame the potential challenges raised as
problems to be solved rather than as reasons not to engage in the conversation or delay start time.
Subsequently, the experience of school leaders who have led a change in start time experienced
no pushback from parents in response to the delay in start time. Even though parents have
concerns about the logistical challenges they might face, they generally support this particular
change. Similarly, the educators surveyed discussed potential challenges, particularly in the areas
of the daily schedule and instructional time, and these concerns should be part of a conversation
around delaying start time but need not be barriers to change.
97

Parent Perception Has Shifted but Start Time Has Not
The findings of this study reflect a shift in parent perception of school start time in recent
years. In the 2015 study conducted by C.S. Mott Children’s Hospital at the University of
Michigan, researchers found that only 51% of parents surveyed would support a change in start
time and only 60% believed that such a change would increase adolescent sleep duration (Davis
et al., 2015). However, in this study, 91% of parents both support a start time delay and believe
that it will lead to increased sleep for their adolescent students. This reflects a notable shift in
perceptions around adolescent sleep in the time between studies. While feelings about school
start time delay might have changed in recent years, start times have not shifted considerably.
Of the schools identified as possible participants for this study, 84% do not meet the APP
recommendation in the 2020-2021 school year. This percentage is similar to the percentage of
public schools that do not meet the recommendation in two previous national studies. Wheaton et
al. (2015) found that for the 2011-2012 school year, 82.3% of public middle and high schools did
not meet the recommendation. Similarly, the U.S. Department of Education (2020) found that
82% of public high schools started before 8:30 a.m. More importantly, this average has been
static for the last 10 years, while the findings of this study point to a notable shift in parental
perception. Although the perceptions of start time delay have shifted, the findings reflect a need
for improved stakeholder knowledge of adolescent sleep needs and the associated risks
associated with adolescent sleep deprivation.
Stakeholders Need to Deepen Their Knowledge of Adolescent Sleep Needs and Risks
In their statement making the recommendation for a delayed start time, the AAP also
recommended engaging in a robust process to educate parents and school leaders about the sleep
needs of adolescents and the risks associated with adolescent sleep deprivation (School Start
98

Times for Adolescents, 2014). The findings of this study support the AAP’s recommendation for
increased stakeholder education as a means to effect change. While the findings of this study
show that parents and educators have a general sense of the appropriate sleep duration for
teenagers, their responses about sleep needs and actual sleep duration, the benefits of a start time
delay, and the risks associated with adolescent sleep deprivation indicate a need for increased
stakeholder education.
In their 2006 study of adolescent sleep, the National Sleep Foundation highlighted a
discrepancy between the sleep duration reported by teenagers and the perceptions of their
parents. In that study, their data showed that only 13% of teenagers got the recommended 8.5
hours of sleep, but 36% of parents believed that their children were indeed getting enough sleep.
In this study, the data reported by parents also shows that only 13% of their children meet the
8.5-hour recommendation, the same percentage published by the NSF 15 years ago. The median
sleep duration as reported by parents in this study was 7 hours. However, while the NSF’s 2006
study found a majority of parents believe their children to be getting enough sleep, this study
found almost three-fourths (74%) of the parents surveyed do not believe their children are getting
enough sleep. While the percentage of parents who seem to know that their children are not
getting enough sleep has increased significantly, the fact that the same percentage of students
still do not meet the recommended sleep duration indicates that parental knowledge alone is not
enough to increase adolescent sleep duration.
Start Time Delay Increases Adolescent Sleep Duration
An important place to deepen stakeholder knowledge is in the efficacy of a start time
delay on adolescent sleep duration. One of the most important findings in previous research is
that a delay in school start time increases adolescent sleep duration (Boergers et al., 2014,
99

Carskadon et al., 1998; Owens et al., 2010; Paksarian et al., 2015, Perkinson-Gloor et al., 2012).
While these studies varied in scope, they each demonstrated that a delay in start time improves
sleep outcomes for adolescents. Paksarian et al. (2015) found that a 30-minute delay led to
approximately 20 more minutes of sleep. Similarly, Perkinson-Gloor et al. (2012) found that a
delay of 20 minutes led to 16 more minutes of sleep. In their study, Owens et al. (2010) showed
that a 30-minute delay significantly reduced the number of students sleeping less than seven
hours per night and significantly increased the number of students getting eight hours or more.
However, despite these studies that repeatedly demonstrate improved sleep duration in
conjunction with delayed school start time, parents and educators both expressed skepticism
about the efficacy of such a delay.
The findings of this study indicate that 91% of parents and 86% of faculty believed that a
delay would improve sleep outcomes. Though the quantitative responses reflect the general
perception that a delayed start time will increase adolescent sleep, some open-ended responses
express skepticism. As discussed in the previous chapter, nine parents and 20 educators
addressed concerns that a delayed start time would simply lead to students staying up later.
These statements reflect a need for increased education on the efficacy of a start time delay as
well as the biological changes that happen at the onset of puberty that shift adolescent sleep
patterns (Hummer & Lee, 2016; Kelly et al., 2014; School Start Times for Adolescents, 2014).
Adolescent sleep onset time and sleep duration are not solely a result of choices that teenagers
make, but rather a product of their biology and the time they need to be at school. An
improvement in stakeholder education could help to alleviate this skepticism in the stakeholders
that have input in the decision-making process. In addition to the biological changes and the
100

sleep benefits of a start time delay, stakeholder education needs to improve in the understanding
of the real risks associated with adolescent sleep deprivation.
Findings Do Not Address the Dire Risks Associated with Adolescent Sleep Deprivation
In general, the findings of this study highlight that stakeholders believe a delay will lead
to improved health and wellness outcomes for their students; however, these findings do not
address the direst health and safety risks adolescents face when sleep deprived. Parents and
educators alike believe that a delay has the potential to improve health outcomes, increase sleep,
and improve outcomes for overall stress and happiness levels. Beyond the noted impact on stress
and happiness reflected in the findings of this study, adequate sleep leads to a reduction in
depression, suicidal thoughts and actions, drug use and abuse (Clark & Harvey, 2012; Owens,
2014; Winsler et al., 2015), and other high-risk behaviors (Wheaton et al., 2016). In addition to a
reduction in these risks, delaying school start time also reduces the rate of adolescent traffic
accidents (Danner & Phillips, 2008, Martiniuk et al., 2013, Vorona et al., 2011, Vorona et al.,
2014).
While parents agreed in high percentages with the possibility of improved health
outcomes, they were the least confident in the impact of start time delay on automobile safety.
When asked specifically about automobile safety outcomes for their teens, 26% of parents did
not believe that a delayed start time would have the impact of reducing teen traffic accidents.
However, numerous studies compare teen traffic accident rates in districts with a delayed start
time to those with a traditional early start time. These studies show that there is a reduction in the
rate of traffic accidents in districts with start times that meet the AAP recommendation (Vorona
et al., 2011, Vorona et al., 2014). Though most students discussed in the survey drive less than
30 minutes each way to and from school, parents indicated that some of those teens drive
101

younger siblings. Driving safety is a concern reflected generally in the discussions of traffic and
commute. While the parents did not agree about what a delay would mean for the amount of
traffic or length of commute, many indicated that they wanted a start time that improved safety
outcomes for their children. If more parents knew the details of the association between start
time and these significant risk factors, they might feel stronger about encouraging their school
leaders to implement this change.
Increasing Self-Efficacy in School Leaders Can Effect Change
Ultimately, it is school leaders who establish the academic daily schedule and execute the
implementation of a start time delay. In order to effect change, there needs to be an increase in
the self-efficacy of those school leaders to make them more confident in their ability to
implement such a delay. Those leaders need to feel equipped and empowered with the skills and
knowledge pertinent to implementing start time delay. As noted in Figure 1, social cognitive
theory addresses the impacts of personal factors, such as stakeholder knowledge and self-
efficacy, and external factors on behavior change. If school start time delay is the desired
behavior change, school leaders must address both the personal and external factors when
implementing such a change. For school leaders, this study may provide an opportunity to
examine stakeholder perceptions around the benefits and challenges of a school start time delay.
The subsequent recommendations, rooted in the responses from the participants, might help
provide administrators the procedural knowledge to implement such a change and increase their
self-efficacy when it comes to making such a change.
102

School Leaders Need a Broader Understanding of the Impacts of Start Time Delay on the
Community
In addition to the broad importance of parent and faculty education as a means of
increasing stakeholder knowledge, school leaders also need more knowledge on this topic. The
findings of this study provide meaningful insights for school leaders about the feelings of their
key stakeholder groups. First and foremost, school leaders are looking for the best outcomes for
their students. The literature reviewed for this study presents myriad ways that school start time
can improve outcomes for their students. Additionally, the findings of this study reflect the
stakeholders’ beliefs that delaying school start time will improve outcomes for their students. Of
import to school leaders, these findings also demonstrate that a significant majority of parents
and educators would support a school in making a start time delay. Armed with this knowledge,
school leaders are more likely to believe that they would have the support of the community in
doing so. Supporting the quantitative findings in this study and leaning on the experiences of the
school leaders interviewed can also promote self-efficacy in school leaders.
The perspectives of the leaders interviewed for this study support the quantitative
findings. As Sam at Berwick School colloquially put it, “It (start time delay) makes a difference.
And so for crying out loud, it ain’t that big a deal. Just do it.” The other leaders interviewed for
the study discussed their processes of research, committees, and the communication of their
decision. However, when given the opportunity to offer advice to other schools or changes they
would make to their process, none of the participants brought up changing the start time back.
Their advice included the consideration of other external factors but noted that the support from
parents had been overwhelmingly positive. Additionally, over time faculty had been able to get
on board with the change. Taken with the survey data, the act of benchmarking their process with
103

the experience of peer schools within the AAAIS can improve self-efficacy in leaders at schools
within the organization and potentially other private schools.
Finally, in addition to the support from parents and faculty demonstrated in the surveys
and the positive experiences of the other school leaders, there is supporting evidence in the
findings of previous studies. In their longitudinal study, Boyland et al. (2015) surveyed parents
and educators about their perceptions of the change in the year before making a change and in
each of the two subsequent school years. Their study found that parents’ and teachers’
impressions of the change became more favorable over time. This is supported in the experience
of the school leaders interviewed for this study. Considering the Boyland et al. (2015) study in
conjunction with the high rate of support demonstrated in the findings of this study, school
leaders should feel confident in the long-term support for school start time delay.
External Factors Should Not Inhibit Behavior Change
Though nine out of 10 stakeholders offered support for a school start time delay in the
quantitative responses, their open-ended comments addressed myriad external factors that are
potential barriers to change. For leaders exploring a start time delay, considering these external
factors in the planning and implementation process of the schedule change and then clearly
communicating those considerations will be critical to building support for a start time delay.
Though these external factors are important and are potential challenges to implementation, the
significant support for a start time change implies stakeholders are looking for schools to find
solutions for these potential barriers. Stakeholders are ready for the change. They also want to be
sure that a school considers the broader ramifications for students, their families, and the faculty
when implementing a change. The longitudinal study from Boyland et al. (2015) also presents
data indicating that the major concerns parents and educators had ahead of making the start time
104

delay ended up not being as significant as they expected when they were surveyed in subsequent
years about the change. For school leaders, the support of the parents and faculty, the experience
of other school leaders, and the findings of previous research, and the specific recommendations
around the change made in this study should have the effect of improving the self-efficacy. It is,
however, important for those leaders to keep in mind the broader implications of one change on a
student’s full set of time commitments at school on a daily basis.
School Start Time Delay Does Not Happen in a Vacuum
The goal of a start time delay is to improve adolescent sleep outcomes. Subsequently,
schools need to consider the full range of a student’s time commitments to school on a daily
basis. The conversation about school start time is often in the context of the academic daily
schedule and is framed in a discussion of the time from first bell in the morning to the dismissal
bell in the afternoon. Increasingly, however, students are committed to campus and school life
much longer than the academic day. Sports teams and clubs hold morning practices and
meetings. Students spend long hours after school engaged in extracurricular activities and then
come home to nightly homework. While school start time is one factor that can allow students to
get more sleep, the school is responsible for looking at the other ways they are asking students
for their time and how those requests might impact student sleep. In examining the full schedule
implications for a student, schools should consider a number of questions about how they ask
students to use their time.
What Is the Unofficial Start Time?
As noted in the findings of this study, students are regularly being asked to arrive on
campus more than 30 minutes before the official start of the academic school day. Clubs have
meetings and sports teams have morning practices. As one school leader identified, some at the
105

school view the delay as more time to fill in the never-ending quest to improve college resumes
and athletics teams. That leader shared that if they were making the change again today, they
would implement firmer limits on the way the time before school can and cannot be used. One
faculty member noted in their survey response the need to prioritize students’ health and
academics over their extracurricular activities. This speaks to a school’s need to examine the
impact of those morning activities on student sleep. With that said, if schools that have a start
time that does not meet the AAP recommendation are currently asking students to arrive early to
campus for clubs and athletics, a delay in start time could potentially still allow for those
morning activities to start just a little later, even if nothing else changes.
Will a Delay Just Push Back All Activities?
Another repeated concern in the open-ended comments on the survey was that a delay in
start time will simply push back all afternoon activities, meals, and homework, shifting a
student’s wake and sleep times without increasing sleep duration. While this was a theme in the
survey responses, the experienced leaders in the interviews shared the ways their schools
combatted this challenge. At Dunhill Academy, the academic schedule design, which involved a
shift to fewer, longer classes a day in addition to the start time delay, meant that they were able
to leave the end time of the school day relatively untouched. As a result, they did not experience
a shift in after-school timing. Similarly, Berwick School did extend the end of the academic day
slightly, but implemented a “hard stop” end time to campus activities at 6:00 p.m. As Melissa
shared, this policy had the effect of slightly shortening some after-school practices; it also
preserved the opportunity for families to eat dinner and for students to complete their work and
go to sleep at an appropriate time. The campus end time prevented the reported anxiety of
stakeholders about simply delaying sleep onset rather than improving sleep duration.
106

What if a Delay Does Not Benefit Everyone?
One concern raised in the surveys from parents and faculty is that because of the schedule
commitments of parents and siblings, as well as traffic implications, there are some students who
might not reap the benefits of a school start time delay. It is plausible that a parent or sibling
would still need to drop a student off at the same time, regardless of the school start time. Faculty
also asked about the potential for extra responsibilities before school for students who need to
arrive to school significantly before the first bell. While this is a valid concern, Allison addressed
the experience at Dunhill Academy, noting that they have plenty of places on campus for
students to go before class starts. Additionally, she noted that in their experience, not all of the
families in their population were limited by having two working parents, so for some it was a
non-issue. Sam offered that students at Berwick School use that time meet with teachers for extra
help if needed. None of the interview participants noted this anticipated challenge as one that
turned out to be significant. It is also prudent to note that while a start time delay may not benefit
everyone, it would not have a significant negative impact on those students arriving to campus at
their previous time. A school leader should not avoid a start time delay simply because it may
not benefit everyone.
Summary
The findings of this study demonstrate support for a school start time delay. The study
highlights places where increased stakeholder education might improve stakeholder response to a
school making a start time delay. Additionally, the findings of the study both in the qualitative
and quantitative elements can be used to increase self-efficacy in school leaders considering
making a change in their school’s start time. Those leaders can lean on the experiences of leaders
at peer schools within the AAAIS, as well as on the data as reported by parents and faculty at
107

other AAAIS schools. Finally, this study highlights the importance of school leaders considering
the goals and potential outcomes of a start time delay within the broader context of the full range
of commitments that students have to school life on a daily basis.
Recommendations for Practice
These recommendations reflect the confluence of the relevant literature, conceptual
framework provided by social cognitive theory, and the findings of this study specifically. When
school leaders look to these recommendations, they can be carried out and implemented in order.
As Sam noted in his interview, each independent school has its own specific culture and context.
As a result, these recommendations provide a general framework that school leaders can use,
filling in the details and specific actions steps that make sense given the particular context of
their institution.
Recommendation 1: Now Is the Time to Delay Start Time to Meet the AAP
Recommendation
This is the time to make a meaningful change to start time. Not only do the findings of
this study support a change, but the ongoing Covid-19 pandemic has hastened the pace of change
at many institutions. As previously discussed, the findings of this study reflect a significant shift
in perception of a delayed school start time, even in the last few years. Davis et al. (2015) found
that only just over half of the parents they surveyed would support a delay in school start time.
This study found that more than 90% of parents and educators were in favor of the change. This
high rate of support should be encouraging to school leaders who are interested in making a
delay to school start time. The adolescent sleep research has been consistent, and the AAP made
its recommendation seven years ago. Now that stakeholders are in support of the change, the
time is right for school leaders to take advantage of that support.
108

Beyond the support for school start time delay, the Covid-19 pandemic has also shifted
perspectives on student learning and change within independent schools. In the AAAIS, many
schools have adapted in such a way as to continue in-person learning during the pandemic. Like
most organizations, the pandemic has asked schools to rethink the way they have been operating.
In his work on organizational change, W. Warner Burke (2002) cites Kurt Lewin’s model for
change as “unfreeze – movement – refreeze” (pp.106-107). In effect, Covid-19 has forced many
organizations, independent schools included, into the “unfreeze” state of Lewin’s model. Schools
are now in a position to be more open to examine all available options to make the best decision
moving forward. That best decision is not necessarily simply returning to the way things were
before the pandemic. Because most parents and educators are ready for school start time delay
and because many organizations are primed for change, school leaders should take advantage of
this opportunity to make a change.
Recommendation 2: Engage Faculty in Building the New “Daily Schedule”
A change in start time requires the examination of the full academic daily schedule.
Schools looking to make a change will be forced to examine period lengths and the school day
end time. The findings of this study show faculty concerns centered around the potential end
time and the impact that might have on their extracurriculars as well as the potential loss of
instructional minutes. They were also interested in building in time in the schedule for faculty
meetings, collaboration, and student tutorial sessions. Because of the myriad concerns the faculty
have about the schedule in which they will be operating, they should be an active part of the
change process. Leaders at both Berwick School and Dunhill Academy discussed having a
committee of faculty engaged in the process of developing the new schedule.
109

In addition to having a small group of key faculty members actively engaged in
developing schedule options with the administration, those key faculty members should also
engage with the faculty at large to bring them into the discussion. As faculty are confronted with
the benefits and challenges of the schedule options, they will engage with the process and have
an opportunity to express their concerns about the things they shared on the surveys. Feeling as
their opinions are valued and considered in the change process will help bring faculty on board
with the change itself. Additionally, this process allows faculty to identify the particular things
that matter most within the culture of the particular institution and develop a daily schedule that
affirms those priorities. In addition to the academic daily schedule, schools also need to consider
the broader implications of a student’s full schedule of commitments to school.
Recommendation 3: Plan for All of the Time Students Are Committed on Campus
Beyond the roughly 8:30 a.m. to 3:00 p.m. academic daily schedule, schools need to
carefully consider the way students are expected to use their time before and after school. As one
participant warns in their interview, “In a fast-paced community where people are always
wanting to sort of get 1% better every day,” coaches and club sponsors look for opportunities to
use any and all available time. Instead of increasing sleep time, some are apt to see the time as
space that can be filled with other commitments. As a result of their institution not setting strict
limitations on the way time before school is used, the leader felt they were missing an
opportunity to actually improve student sleep outcomes despite the later start. Parents and faculty
also addressed concerns about how early schools are asking kids to arrive at school and how late
they are asking them to stay. When planning to implement a schedule change, schools should
look beyond the academic school day and set limits on the earliest arrival time for an activity and
the latest end time for a practice. While not all event timing is within an individual school’s
110

control, like athletic competitions or evening arts performances, a default start and end time for
campus activities can help students actually achieve the sleep gains intended by a delayed start
time. Berwick School made this choice and implemented a 6:00 p.m. end time for campus
activities. As a result, the delay in start time did not delay all subsequent activities. Beyond
earliest arrival and latest departure times, schools also need to plan to accommodate students
who might need to arrive to campus at their pre-delay time to as a result of parents’ jobs or
siblings start times.
Along with potentially limiting early arrivals and late departures, schools do need to
prepare spaces and supervision for students that still need to arrive on campus at their previous
regular time. Because of the lack of flexibility in parental work schedules or student who need to
drive siblings in other divisions, there will be students on campus early. Allison shared that there
were places specifically set up for those students to go in the mornings at Dunhill Academy, and
that will be an important feature in the planning phase for school leaders making a delay in start
time.
Finally, in addition to the consideration of the time schools are requiring students to be on
campus, leaders should consider the homework expectation of students as a component of the
overall commitment a student has to the school on a daily basis. Though research about the
efficacy and duration of homework is beyond the scope of this study, it is a contributing factor in
student sleep and a concern raised by parents and faculty on their surveys. A “regular” work
week for adults is 40 hours. Typically, students are in school for roughly 35 hours a week in the
academic daily schedule. They go then to their extracurricular commitments, practices, and
rehearsals. After those commitments end, they return home to eat dinner and do homework, often
long into the night. The concern about start time delay pushing students even later on completing
111

their homework is reflected in the parent survey, and the concerns ask that schools consider the
homework expectations as part of a student’s overall daily commitment to school activities. After
deciding to make a change and considering the ramifications of that delay on the academic
schedule and full range of student commitments, school leaders need to meaningfully
communicate their decisions to the necessary stakeholders.
Recommendation 4: Communicate the Intended Changes with a Clear Rationale
Clear communication is key to any meaningful organizational change. When
communicating about a delay in start time, school leaders should lead their communication with
the body of existing research. In addition to clearly written community emails that articulate the
change and the rationale for the change, as Sam discussed in his interview, schools should
provide opportunities for stakeholder education. In line with the AAP’s recommendation for start
time delay implementation, schools should provide educational opportunities for students,
parents, and faculty that is rooted in current research. Stakeholders need to know about
adolescent sleep needs, the impacts of sleep deprivation on teenagers, and the benefits of a
delayed start time. When parents and faculty are confronted with the data about the amount of
sleep their students need compared to the amount of sleep they are getting, those decision-
making stakeholders may be more likely to support a start time delay. Additionally, as Kohl
Malone et al. (2017) recommend, increasing the efficacy of the messaging in those education
sessions is important. Parents are particularly responsive to salient messaging and clear graphics
addressing the real risks associated with adolescent sleep deprivation, including highlighting
those risk factors that have life and death ramifications for people’s children, which can be
mitigated, in part, by school start time delay.
112

While clearly explaining the scientific rationale and providing opportunities for
stakeholder education are important in effectively communicating this change, school leaders
also need to address the concerns of the stakeholders in that communication. The findings of this
study indicate that most parents are ready for schools to make this change to better meet the
needs of their children. After having addressed the daily schedule and the full extent of the time
commitments of their students, school leaders need to communicate to parents how they are
implementing change that will benefit the students. If school leaders address the start and end
times, as well as any changes that impact transportation, extracurriculars, and student
expectations, parents will be more likely to be on board. They are ready for this change; they just
need to know that school leaders have thought through and addressed the myriad potential
challenges. As Allison shared in her interview, once you tell parents that you are making a
change to improve health outcomes for their children, there will ultimately be little resistance.
Limitations and Delimitations
This study is limited in scope by both the location of the study and the particular context
of the research. By situating the study in the context of independent schools, this study
purposefully excluded a number of factors that would contribute to the discussion about school
start time delay in public school systems, including legislative and fiscal constraints unique to
that context. Specific exclusions include systems critical to public school start times like bussing.
Additionally, a number of stakeholder responses in the surveys and interviews indicated a
flexibility in drop off that might not be as wide ranging in the public-school contexts. Further
limitations include the socio-economic status of the participant pool and how that might impact
perceptions. There are independent school families in lower income brackets, who often live
farther away from school campuses, but only a small percentage of lower income participants
113

responded to the survey. This limitation is due in part to the specific sample size of the
participant pool and partly due to the limited number of participating schools. However, because
of the specific independent school context, the researcher has the ability to make certain
assumptions in the protocols about the similarities in the AAAIS member schools and
participants.
Because the study was conducted specifically with AAAIS member schools, it is not
generalizable to all independent schools. Additionally, because only eight of 31 possible schools
opted into the survey, the findings are reflective of the participants in this study, but may not be
fully reflective of the entire AAAIS population. This poses a limitation in the reliability of the
study, as research replicating the study might engage the potential participants in the same way
and generate participants from schools in other areas of metro Atlanta and with other cultures
and contexts. Similarly, because only two of the five schools with delayed start times offered
interviews with their leadership, the ability to triangulate findings was limited to five participants
at the two schools. This limited the researcher’s ability to triangulate the interview findings
across multiple institutions. Additionally, the added step of not being responsible for the
distribution of the survey and relying on individual school leaders to distribute it as they saw fit
meant that the researcher was also not able to send reminders, time the communications, or
impact the specific method of participant recruitment. This was potentially a limiting factor in
the total number of participants. Though the 302 usable surveys generated a robust amount of
data, they represent a small percentage of the total number of parents and faculty at the 31
AAAIS high schools that were invited to participate in the study.
One significant limitation of the study was that it only examined the impact of school
start time on adolescent sleep. Open-ended responses discussed a wide range of other factors
114

impacting adolescent, including homework, technology, and the full range of expectations that
are placed on students daily. This study did not delve into academic rigor as it relates to sleep
with questions about homework duration or Advanced Placement courses, nor did the study as
parents to report the number or types of activities that students are engaged in, which might
inform early start times or late bedtimes. This study was limited in scope as it only examined
impacts and perceptions of start time, and not the myriad other factors impacting adolescent
sleep.
There are also limitations to the accuracy of some of the data collected because the
questions asked for perception estimates rather than direct observation. A number of questions
asked for estimates of normal experience, including sleep duration, commute times, and students
sleeping in class. In particular, the parents’ estimate of student sleep duration is one that is
speculative. Previous studies have demonstrated a discrepancy between adolescent sleep duration
in studies conducted on teens as compared to sleep data that has been reported by their parents.
Because the nature of this study aimed to examine organizational barriers, taking only the
parents’ responses potentially limited the accuracy of that question.
Another limitation of this study was the ongoing Covid-19 pandemic at the time of the
research. The pressure to re-open for in-person or hybrid learning in the fall and the burdens of
executing those plans served as a barrier to participation in this study. In order to execute those
plans, many schools had conducted recent surveys of their parents and faculty and did not want
to ask stakeholders to take a survey that was not specifically pertinent to their organization.
Additionally, some simply shared in their rejection of the invitation to participate that they did
not feel that they could, in good faith, add an additional request of their parents, faculty, or
leadership. Furthermore, some of the specific answers to survey questions were likely impacted
115

by the pandemic. In metro Atlanta, the number of people working remotely has notably reduced
traffic and shortened commute times. Also, for students who are learning remotely, the time
normally needed to leave the house and get to school on time is no longer a contributing factor in
their sleep duration.
Recommendations for Future Research
This study specifically sought to address school start time as a way to improve adolescent
sleep outcomes for students in independent schools. As a result, there are many areas related to
adolescent sleep that were not explored. Given the pace of change of technology and the
pandemic shift to online learning, there is the opportunity to more deeply explore technology as a
barrier to adolescent sleep and virtual schooling as a potential benefit. This could look at the way
students use their personal technology for social media and video gaming, as well as the impact
of long hours of screen time as an expectation of the school day.
Additionally, while there is anecdotal evidence in this study pointing to the impact of
homework on sleep patterns and that is briefly discussed in the findings and recommendations,
further research could more deeply explore the connection between the two. There seemed to be
a feeling that there are students who would go to bed earlier but are up late trying to finish the
expected homework. Particular to the independent school context, perceptions of academic rigor
might be a factor that impacts homework duration. While the scope of this study intended to
explore the barriers to organizational change, the findings highlighted real concerns about other
barriers to adolescent sleep, particularly in the independent school context. Beyond asking
parents about the number of early arrival days, a future study could offer a more robust
examination of the full scope of student commitments on a daily basis.
116

This study focused exclusively on traditional independent high schools within the
AAAIS. However, one of the features of many independent schools is that they consist of
multiple divisions within the same campus and often need to coordinate between those divisions.
Because this study focused solely on high school stakeholders, it omits the perspectives of
middle and elementary school parents, faculty, and leadership that might be impacted by a high
school division change. While this study serves as a guide to high school leaders, if a middle
school division head asked how the change might impact middle school families, this study
would not offer meaningful insight. To that end, further research could more deeply examine the
ways families with children in multiple divisions might be impacted by a change in one.
Further, because this study was conducted during the Covid-19 pandemic, further studies
could seek to replicate the findings after the pandemic. A number of findings, including a
willingness to change and the potential impacts of school start time delay, could be affected by
the current conditions of the pandemic. Additionally, the pandemic had the effect of suppressing
response rate, and after the pandemic, future researchers might find participants more willing to
engage in the research process.
Conclusion
School leaders have little control in the lives of their students. They cannot put them to
bed at the right time or choose their extracurricular activities. School leaders do not dictate how
much time a student might spend on particular homework assignments or keep students from
holding jobs or participating in off-campus organizations. School leaders are not able to make a
student leave their phone outside of their bedroom at night or limit time spent playing video
games. However, school leaders are still held responsible for the health and well-being of the
students that arrive early to their campus each morning and stay late into the evening. School
117

leaders are asked to address the challenges when students engage in risky behavior, use or abuse
drugs and alcohol, or struggle with mental health issues on their campuses. There is so much out
of the control of school leaders that it could be easy to feel like there is nothing they can do to
improve student health, well-being, and academic outcomes. It is easy for leaders to feel that
some issues, particularly in the realm of student physical and mental health, are just too complex
to tackle. This issue is not one of them.
This study demonstrates that there is one clear change school leaders have full control
over making that will be a significant benefit to their students in many ways. Parents and faculty
are ready to make a change in school start time. They anticipate notable benefits in the realm of
student health, safety, stress, happiness and academic performance. They have concerns about
the potential ramifications a change might have on student, family, and faculty schedules as well
as the impact on the amount of instructional time in the academic day. All of these challenges are
manageable and within the control of the school leadership. From an organizational perspective,
the actual negative ramifications of a start time change are minor, but the potential benefits for
the students are tremendous. School leaders are fully in control of bringing along the other key
stakeholders in implementing a school start time delay that will lead to a significant benefit in the
lives of their students. There is a significant body of research that shows the benefits of school
start time delay on adolescent sleep, and the findings of this study affirm that parents, faculty,
and school leaders are also in support of making such a change. The support for school start time
delay is there and the timing is right. School leaders have the power to tangibly improve
outcomes for their students.

118

References
About AAAIS. (2018) AAAIS. https://aaais.org/about-aaais/
Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change.
Psychological Review, 84(2), 191–215.
http://dx.doi.org.libproxy1.usc.edu/10.1037/0033-295X.84.2.191
Bandura, A. (1988). Organisational applications of social cognitive theory. Australian Journal of
Management, 13(2), 275–302. https://doi.org/10.1177/031289628801300210
Boergers, J., Gable, C., & Owens, J. (2014). Later school start time is associated with improved
sleep and daytime functioning in adolescents. Journal of Developmental and Behavioral
Pediatrics : JDBP, 35(1), 11–17. https://doi.org/10.1097/DBP.0000000000000018
Boylan, L., Harvey, M., Riggs, W., Campbell, B. (2015). Changing school start times to combat
adolescent sleep deprivation: Perceptions of school stakeholders. Journal of School
Public Relations, 36(1), 7-42
Burke, W. (2002). Organization change: Theory and practice. Sage Publications.

Carskadon, M., Vieira, C., & Acebo, C. (1993). Association between puberty and delayed phase
preference. Sleep, 16(3), 258–262.
Chen, M., Burley, H., & Gotlib, I. (2012). Reduced sleep quality in healthy girls at risk for
depression. Journal of Sleep Research, 21(1), 68–72. https://doi.org/10.1111/j.1365-
2869.2011.00934.x
Clarke, G., & Harvey, A. (2012). The complex role of sleep in adolescent depression. Child and
Adolescent Psychiatric Clinics of North America, 21(2), 385–400.
https://doi.org/10.1016/j.chc.2012.01.006
119

Creswell, J. (2014). Research design: Qualitative, quantitative, and mixed methods
approaches (4th ed.). SAGE Publications.
Crowley, S., Acebo, C., & Carskadon, M. (2007). Sleep, circadian rhythms, and delayed phase in
adolescence. Sleep Medicine, 8(6), 602–612. https://doi.org/10.1016/j.sleep.2006.12.002
Danner, F., & Phillips, B. (2008). Adolescent sleep, school start times, and teen motor vehicle
crashes. Journal of Clinical Sleep Medicine, 4(6), 533–535.
Davis M.M., Kauffman A.D., Singer D.C., Matos-Moreno A., Clark S.J. (2015). Parents
conflicted about later school start times for teens. C.S. Mott Children's Hospital National
Poll on Children's Health, University of Michigan. 23(1).
Dunietz, G. L., Matos-Moreno, A., Singer, D. C., Davis, M. M., O'Brien, L. M., & Chervin, R.
D. (2017). Later school start times: What informs parent support or opposition?. Journal
of Clinical Sleep Medicine, 13(07), 889-897.
Edwards, F. (2012). Early to rise? The effect of daily start times on academic performance.
Economics of Education Review, 31(6), 970–983.
https://doi.org/10.1016/j.econedurev.2012.07.006
Eight Major Obstacles to Delaying School Start Time (n.d.). Retrieved from:
https://www.sleepfoundation.org/articles/eight-major-obstacles-delaying-school-start-
times
Galland, B., Short, M., Terrill, P., Rigney, G., Haszard, J., Coussens, S., Foster-Owens, M., &
Biggs, S. (2018). Establishing normal values for pediatric nighttime sleep measured by
actigraphy: A systematic review and meta-analysis. Sleep, 41(4).
https://doi.org/10.1093/sleep/zsy017
120

Galván, A. (2020). The need for sleep in the adolescent brain. Trends in Cognitive Sciences,
24(1), 79–89. https://doi.org/10.1016/j.tics.2019.11.002
Hafner, M., Stepanek, M., & Troxel, W. (2017). Later Start Times in the U.S.: An Economic
Analysis. RAND Corporation.
Hirshkowitz, M., Whiton, K., Albert, S., Alessi, C., Bruni, O., DonCarlos, L., Hazen, N.,
Herman, J., Katz, E., Kheirandish-Gozal, L., Neubauer, D., O'Donnell, A., Ohayon, M.,
Peever, J., Rawding, R., Sachdeva, R., Setters, B., Vitiello, M., Ware, J., & Adams
Hillard, P. (2015). National Sleep Foundation’s sleep time duration recommendations:
methodology and results summary. Sleep Health, 1(1), 40–.
https://doi.org/10.1016/j.sleh.2014.12.010
Hummer, D. L., & Lee, T. M. (2016). Daily timing of the adolescent sleep phase: Insights from
a cross-species comparison. Neuroscience and Biobehavioral Reviews, 70, 171–181.
https://doi.org/10.1016/j.neubiorev.2016.07.023
Jacob, B & Rockoff, J. (2011). Organizing schools to improve student achievement: Start times,
grade configurations, and teacher assignments. Discussion Paper Series (Hamilton
Project), 8, 1–28.
Kelley, P., Lockley, S. W., Foster, R. G., & Kelley, J. (2015). Synchronizing education to
adolescent biology: ‘Let teens sleep, start school later.’ Learning, Media and Technology,
40(2), 210–226. https://doi.org/10.1080/17439884.2014.942666
Killgore, W. (2010). Effects of sleep deprivation on cognition. Human Sleep And Cognition, Part
I: Basic Research, 185, 105–129. https://doi.org/10.1016/B978-0-444-53702-7.00007-5
121

Kohl Malone, S., Ziporyn, T., & Buttenheim, A. (2017). Applying behavioral insights to delay
school start times. Sleep Health: Journal of the National Sleep Foundation, 3(6), 483–
485. https://doi.org/10.1016/j.sleh.2017.07.012
Malloy, C. (2011). Moving beyond data: Practitioner-led inquiry fosters change. Phi Delta
Kappa International, 6(4), 1-20.
Maxwell, J. A. (2013). Qualitative research design: An interactive approach. (3rd ed.). Thousand
Oaks: SAGE.
Martiniuk, A., Senserrick, T., Lo, S., Williamson, A., Du, P., Grunstein, R., Woodward, M.,
Glozier, N., Stevenson, M., Norton, R., Ivers, R. (2013). Sleep-deprived young drivers
and the risk for crash: the DRIVE prospective cohort study. (Report). JAMA Pediatrics,
167(7), 1–8. https://doi.org/10.1001/jamapediatrics.2013.1429
Maslowsky, J., & Ozer, E. (2014). Developmental trends in sleep duration in adolescence and
young adulthood: Evidence from a national United States sample. Journal of Adolescent
Health, 54(6), 691–697. https://doi.org/10.1016/j.jadohealth.2013.10.201
McEwan, E. K., & McEwan, P. J. (2003). Making sense of research: What’s good, what’s not,
and how to tell the difference. Corwin Press.
Merriam, S. B., & Tisdell, E. J. (2016). Qualitative research: A guide to design and
implementation (4th ed.). Jossey-Bass.
Ming, X., Koransky, R., Kang, V., Buchman, S., Sarris, C., & Wagner, G. (2011). Sleep
insufficiency, sleep health problems and performance in high school students. Clinical
Medicine Insights: Circulatory, Respiratory and Pulmonary Medicine, 5(2011), 71–719.
https://doi.org/10.4137/CCRPM.S7955
122

Minges, K., & Redeker, N. (2015). Delayed school start times and adolescent sleep: A systematic
review of the experimental evidence. Sleep Medicine Reviews, 28, 82–91.
https://doi.org/10.1016/j.smrv.2015.06.002
National Sleep Foundation. (2006). Sleep in America poll. Washington, DC: National Sleep
Foundation.
National Sleep Foundation. (2014). Sleep in America poll. Washington, DC: National Sleep
Foundation.
Owens, J. (2014). Insufficient sleep in adolescents and young adults: an update on causes and
consequences. (Report). Pediatrics, 134(3), e921–32. https://doi.org/10.1542/peds.2014-
1696
Owens, J., Belon, K., & Moss, P. (2010). Impact of delaying school start time on adolescent
sleep, mood, and behavior. Archives of Pediatrics & Adolescent Medicine, 164(7), 608–
614. https://doi.org/10.1001/archpediatrics.2010.96
Paiva, T., Gaspar, T., & Matos, M. G. (2015). Sleep deprivation in adolescents: Correlations with
health complaints and health-related quality of life. Sleep Medicine, 16(4), 521–527.
https://doi.org/10.1016/j.sleep.2014.10.010
Paksarian, D., Rudolph, K., He, J., & Merikangas, K. (2015). School start time and adolescent
sleep patterns: Results from the U.S. national comorbidity survey--adolescent
supplement. American Journal of Public Health, 105(7), 1351–7.
doi:10.2105/AJPH.2015.302619
Perkinson-Gloor, N., Lemola, S., & Grob, A. (2012). Sleep duration, positive attitude toward
life, and academic achievement: The role of daytime tiredness, behavioral persistence,
123

and school start times. Journal of Adolescence, 36(2), 311–318.
doi:10.1016/j.adolescence.2012.11.008
Portantino, A. (2017) School Start Time Research and Information.
https://sd25.senate.ca.gov/sites/sd25.senate.ca.gov/files/research_booklet_sb_328_pdf.pd
f
Pupil Attendance: School Start Time, Cal. Educ.§46148 (2019).
https://leginfo.legislature.ca.gov/faces/billCompareClient.xhtml?bill_id=201920200SB32
8&showamends=false#
RAND Corporation (2017). Later school start times in the U.S.: An economic analysis. Santa
Monica, CA: Hafner, M., Stepanek, M., & Troxel, W. M.
Robinson, S. B., & Leonard, K. F. (2019). Designing quality survey questions. Sage.
School Search. (2018) AAAIS. https://aaais.org/school-search/?wpv_view_count=648&wpv-
grade%5B%5D=upper-school
School Start Times for Adolescents. (2014). Pediatrics. Retrieved from
http://pediatrics.aappublications.org/content/early/2014/08/19/peds.2014-1697.abstract
Start School Later (2020). Legislation. Retrieved April 28, 2020, from
https://www.startschoollater.net/legislation.html
Taylor, D., Jenni, O., Acebo, C., & Carskadon, M. (2005). Sleep tendency during extended
wakefulness: Insights into adolescent sleep regulation and behavior. Journal of Sleep
Research, 14(3), 239–244. https://doi.org/10.1111/j.1365-2869.2005.00467.x
Troxel, W., & Wolfson, A. (2017). The intersection between sleep science and policy:
introduction to the special issue on school start times. Sleep Health: Journal of the
National Sleep Foundation, 3(6), 419–422. https://doi.org/10.1016/j.sleh.2017.10.001
124

Urrila, A., Artiges, E., Massicotte, J., Miranda, R., Vulser, H., Bézivin-Frere, P., Lapidaire, W.,
Lemaître, H., Penttilä, J., Conrod, P., Garavan, H., Paillère Martinot, M., & Martinot, J.
(2017). Sleep habits, academic performance, and the adolescent brain structure. Scientific
Reports, 7(1), 41678–. https://doi.org/10.1038/srep41678
U.S. Department of Education. (2020). Data point: School start time for public schools.
https://nces.ed.gov/pubs2020/2020006.pdf
Vorona, R., Szklo-Coxe, M., Wu, A., Dubik, M., Zhao, Y., Ware, J. (2011). Dissimilar teen
crash rates in two neighboring southeastern Virginia cities with different high school start
times. Journal of Clinical Sleep Medicine : JCSM : Official Publication of the American
Academy of Sleep Medicine, 7(2), 145–151.
http://search.proquest.com/docview/863423657/
Vorona, R. D., Szklo-Coxe, M., Lamichhane, R., Ware, J. C., McNallen, A., & Leszczyszyn, D.
(2014). Adolescent crash rates and school start times in two central Virginia counties,
2009-2011: A follow-up study to a southeastern Virginia study, 2007-2008. Journal of
clinical sleep medicine, 10(11), 1169-1177.
What Are Independent Private Schools? (n.d) https://parents.nais.org/learn/what-are-
independent-private-schools/
Wheaton, A. G., Ferro, G. A., & Croft, J. B. (2015). School start times for middle school and
high school students—United States, 2011–12 school year. MMWR. Morbidity and
Mortality Weekly Report, 64(30), 809.
Wheaton, A. G., Olsen, E. O., Miller, G. F., Croft, J. B. (2016). Sleep Duration and Injury-
Related Risk Behaviors Among High School Students — United States, 2007–2013.
125

MMWR Morbidity and Mortality Weekly Report 2016;65:337–341. DOI:
http://dx.doi.org/10.15585/mmwr.mm6513a1
Winsler, A., Deutsch, A., Vorona, R. D., Payne, P. A., & Szklo-Coxe, M. (2014). Sleepless in
Fairfax: The difference one more hour of sleep can make for teen hopelessness, suicidal
ideation, and substance use. Journal of Youth and Adolescence, 44(2), 362–378.
https://doi.org/10.1007/s10964-014-0170-3

126

Appendix A: Parent Survey Protocol
Question Open
or
Closed?
Level of
Measure
ment.
(nominal,
ordinal,
interval,
ratio)
Response options (if
close-ended)
RQ Concept being
measured
(from
conceptual
framework)
1. I currently have a high
school aged child.
Closed Ordinal Yes
No
dem

Academe of the Oaks The Heritage School St. Francis Day School
Athens Academy Holy Innocents' Episcopal School
St. Pius X Catholic High
School
Atlanta Girls' School Holy Spirit Preparatory School Stratford Academy
Atlanta International School The Lovett School The Walker School
Blessed Trinity Catholic High School Marist School The Weber School
Brandon Hall School Mount Vernon School Wesleyan School
Christian Heritage School Notre Dame Academy Westminster
The Cottage School Pace Academy Whitefield Academy
Darlington School The Paideia School Woodward Academy
The Galloway School Pinecrest Academy
Greater Atlanta Christian School Savannah Country Day School
2. My high school aged
child currently attends
one of the Atlanta Area
Association of
Independent Schools
(AAAIS) member
schools listed above:
Closed Ordinal Yes
No
dem
3. How many children do
you have in your care?
Closed Ordinal Drop down 1-4+ dem
4. How many children do
you currently have in
high school?
Closed Ordinal Drop down 1-4+ dem
5. How many children of
yours have completed
high school?
Closed Ordinal Drop down 0-4+
6. On average, how long
is the morning
commute to school (in
minutes)? If your child
is fully virtual, please
enter 0.
Open Ordinal dem
7. On average, how long
is the afternoon
Open Ordinal

dem
127

commute from school
(in minutes)? If your
child is fully virtual,
please enter 0.
8. How much sleep do
you think the average
teenager needs per
night?
Open Ordinal 1 Behavior
Please answer the following questions about your children currently in high school. This set of
questions will repeat for each of your children. Please start with your oldest child first. (The
following six questions repeated based on how many high school aged children were reported in
question 4)
9. What is this child’s
current age?
Closed Ordinal 13 or younger
14
15
16
17
18
19 or older

10. What is this child’s
current grade?
Closed Ordinal 9
10
11
12

11. On average, how many
hours of sleep does this
child get on school
nights?
Closed Ordinal Under 5 hours
5 hours
5.5 hours
6 hours
6.5 Hours
7 hours
7.5 hours
8 hours
8.5 hours
9 hours
Over 9 hours
Unsure
1 Behavior
12. On a typical school
day, what time does
your child’s school
begin classes?
Closed Ordinal Before 7:00 am
7:00-7:29 am
7:30-7:59 am
8:00-8:29 am
8:30 am or later
1 Behavior
13. Considering extra-
curriculars, athletics,
academic support, and
other personal factors,
how many days a week,
on average, does your
Closed Ordinal 0
1
2
3
4
5
1 Environment/
personal
128

child arrive to school
more than 30 minutes
before the expected
start time?
14. To what extent do you
agree with the
following statement:
Currently, I believe my
child is getting enough
sleep.
Closed Ordinal Strongly agree
Agree
Disagree
Strongly Disagree
1 Behavior
15. In 2014, The American
Academy of Pediatrics
(AAP) made a
recommendation for
school start time. What
time do you think AAP
recommends as the
earliest start time for
high schools?
Open Nominal 1 Environment
16. The AAP urges high
schools and middle
schools to start at 8:30 am
or later, allowing students
to get 8.5-9.5 hours of
sleep in order to improve
physical and mental
health, safety (by
preventing drowsy
driving crashes),
academic performance
and quality of life.

To what degree do you
agree or disagree with the
AAP guideline.
Closed Ordinal Strongly agree
Agree
Disagree
Strongly Disagree
2 Personal
factors
17. To what extent would
you support a change in
your child’s school’s
start time to 8:30 or
later.
Closed Ordinal Strongly support
Support
Oppose
Strongly oppose
Not Applicable
2 Behavior
18. To what extent do you
agree with the
following statement?
Please consider how
these questions would
apply to a regular
school year that is not
impacted by Covid-19.
Closed Ordinal Strongly agree
Agree
Disagree
Strongly Disagree
RQ3
129

An 8:30 a.m. start time
would:
a. Allow my child to get
more sleep
b. Improve my child’s
school performance
c. Improve my child’s
health
d. Negatively affect my
child’s after-school
activities and athletics
e. Decrease my child’s
stress
f. Reduce my child’s ability
to have a job
g. Interfere with mealtimes
h. Alter my child’s
transportation plans in a
negative way
19. To what extent do you
agree with the
following statement?
Please consider how
these questions would
apply to a regular
school year that is not
impacted by Covid-19.
An 8:30 a.m. start time
would:

a. Interfere with my work
schedule
b. Reduce rates of
adolescent traffic
accidents
c. Give me a healthier
child
d. Give me a happier child
e. Interfere with my
commute
Closed Ordinal Strongly agree
Agree
Disagree
Strongly Disagree

RQ3
20. What concerns do you
have about a start time
after 8:30 am?
Open Nominal 3
21. What potential benefits
do you see in a start
time after 8:30 am?
Open Nominal 3
22. Is there anything else
you would like to say
Open Nominal 3
130

regarding a potential
change in school start
time?
23. What is your gender? Open Ordinal dem
24. What is your age? Closed Ordinal Drop down
continuum 21-65+
dem
25. What is your ethnic
background/race?
Closed Ordinal A. White
B. Black or African
American
C. American Indian
or Alaska Native
D. Native Hawaiian
or Pacific Islander
E. Asian
F. Latinx
G. Other
dem
26. Which best describes
your level of
educational attainment?
Closed Ordinal A. Did not graduate
from High School
B. High School
Diploma or GED
C. Associates Degree
D. Bachelor’s degree
E. Master’s Degree
F. Doctorate
G. Other (with a fill
in the blank)

dem
27. What best describes
your annual household
income
Closed Ordinal Less than $50,000
$50,000 to $99,999
$100,000 to $149,999
$150,000 to $200,000
More than $200,000
dem

Appendix B: Faculty and Leadership Survey Protocol
131

Question Open or
Closed?
Level of
Measurement.
(nominal,
ordinal,
interval,
ratio)
Response
options (if
close-ended)
RQ Concept
being
measured
(from
conceptu
al
framewor
k)

Academe of the Oaks The Heritage School St. Francis Day School
Athens Academy Holy Innocents' Episcopal School
St. Pius X Catholic
High School
Atlanta Girls' School Holy Spirit Preparatory School Stratford Academy
Atlanta International School The Lovett School The Walker School
Blessed Trinity Catholic High
School
Marist School The Weber School
Brandon Hall School Mount Vernon School Wesleyan School
Christian Heritage School Notre Dame Academy Westminster
The Cottage School Pace Academy Whitefield Academy
Darlington School The Paideia School Woodward Academy
The Galloway School Pinecrest Academy
Greater Atlanta Christian School Savannah Country Day School
1. I am currently employed
at one of the Atlanta Area
Association of
Independent Schools
(AAAIS) member schools
listed above:
Closed Ordinal Yes
No
dem
2. How do you define your
current primary position in
the school? Please check
all that apply
Closed Ordinal A. Classroom
teacher
B. Counselor
C.
Administrator
E. Other
dem
3. How many years have you
worked at your current
school
Closed Ordinal Continuum – 1-
20+
dem
4. How many years have you
worked in independent
schools
Closed Ordinal Continuum – 1-
20+
dem
5. On average, how long is
your morning commute?
If you are currently fully
virtual, please enter 0.
Open Ordinal dem
132

6. On average, how long is
your morning commute?
If you are currently fully
virtual, please enter 0.
Open Ordinal dem
7. On a typical school day,
what time does the
academic school day
begin?
Closed Ordinal Before 7:00 am
7:00-7:29 am
7:30-7:59 am
8:00-8:29 am
8:30 am or later
1 Behavior
8. How much sleep do you
think the average teenager
needs per night?
Open Ordinal 1 Behavior
9. Currently, I believe my
students are getting
enough sleep.
Closed Ordinal Strongly agree
Agree
Disagree
Strongly
Disagree
1 Behavior
10. In 2014, The American
Academy of Pediatrics
(AAP) made a
recommendation for
school start time. What
time do you think AAP
recommends as the
earliest start time for high
schools?
Open Nominal 1 Environ-
ment
11. The AAP urges high schools
and middle schools to start at
8:30 am or later, allowing
students to get 8.5-9.5 hours
of sleep in order to improve
physical and mental health,
safety (by preventing drowsy
driving crashes), academic
performance and quality of
life.
12. To what extent do you agree
or disagree with the AAP
guideline?
Closed Ordinal Strongly agree
Agree
Disagree
Strongly
Disagree
2 Personal
factors
13. To what extent would you
support a change in your
school’s start time to 8:30
or later?
Closed Ordinal Strongly
support
Support
Oppose
Strongly oppose
Not applicable
2 Behavior
133

14. In general, how many
times a week do students
fall asleep in your class?

Closed Ordinal Drop down
continuum 0 to
10+
1 Behavior
15. To what extent do you
agree with the following
statements?
a. Typically, students fall
asleep more often in
my morning classes.
b. Typically, students fall
asleep more often in
my afternoon classes.
c. I feel that a school
start time of 8:30 am
or later benefits
students by more
closely matching their
natural sleep schedule.
d. I feel that a school
start time of 8:30 am
or later benefits
students by taking
advantage of the best
times for learning.
e. I am concerned a
school start time of
8:30 am or later would
have a negative impact
on my students'
schedules.
16. I am concerned a school
start time of 8:30 am or
later would have a
negative impact on my
schedule
Closed Ordinal Strongly agree
Agree
Disagree
Strongly
Disagree
3 Environ-
ment
17. To what extent do you
agree with the following
statement? Please consider
how these questions
would apply to a regular
school year that is not
impacted by Covid-19. An
8:30 a.m. start time would:
a. Allow my students to get
more sleep.
Closed Ordinal Strongly agree
Agree
Disagree
Strongly
Disagree
3 Environ-
ment
134

b. Improve my students’
school performance.
c. Improve my students’
health.
d. Limit my students’ time
for after-school activities
and athletics.
e. Decrease my students’
stress.
Reduce my students’ ability to
have an after-school job.
18. To what extent do you
agree with the following
statement? Please consider
how these questions
would apply to a regular
school year that is not
impacted by Covid-19. An
8:30 a.m. start time would:
a. Negatively affect my
childcare plan.
b. Interfere with my
commute.
c. Allow me to get more
sleep
d. Affect my
extracurricular
commitments after
school
Closed Ordinal Strongly agree
Agree
Disagree
Strongly
Disagree
3 Environ-
ment
19. What concerns do you
have about a start time
after 8:30 am
Open Nominal 3
20. What benefits do you see
in a start time after 8:30
am?
Open Nominal 3
21. Is there anything else you
would like to say
regarding school start
time?
Open Nominal 3
22. What is your gender? Open Ordinal dem
23. What is your age? Closed Ordinal Drop down
continuum 21-
65+
dem
24. What is your ethnic
background/race
Closed Ordinal A. White dem
135

B. Black or
African
American
C. American
Indian or
Alaska Native
D. Native
Hawaiian or
Pacific Islander
E. Asian
F. Latinx
F. Other
25. Which best describes your
highest level of
educational attainment?
Closed Ordinal A. Bachelor’s
degree
B. Master’s
degree
C. Doctorate
D. Other
dem
136

Appendix C: Leadership Interview Protocol

1. How would you describe your school? (demographic)
2. What is your position at the school? (demographic)
a. How long have you been here?
b. What other roles have you served in at the school?
3. When did your school move to a high school start time that is later than 8:30 AM?
(Background information)
a. What was the previous start time?
b. What is the current start time?
4. What factors led to the school considering a delay in start time? (RQ2; self-efficacy)
5. Once it was decided this was a change worth exploring, what action steps did the school
take to make the change? (RQ 4; behavioral factors)
6. How did you engage the parents on the issue of school start time? (RQ1; stakeholder
knowledge and beliefs)
a. Was there pushback - how did they respond? (RQ 3; anticipated barriers)
7. In what ways did you engage other school leadership and faculty? (RQ1; stakeholder
knowledge and beliefs)
a. Was there pushback - how did they respond? (RQ 3; anticipated barriers)
8. Prior to making the change, what were some of the anticipated benefits and challenges?
(RQ 3; anticipated barriers)
a. After making the change, what benefits have you seen?
b. To what extent did they match the anticipated benefits?
c. What challenges have you seen as a result of making a change?
d. To what extent did they match the anticipated challenges?
9. After you decided to make the delay, what other systems required examination or
change? (RQ4; how did time impact other systems or stakeholders)
a. Daily schedule?
b. Bussing?
c. After-school sports?
d. Carpool lines
10. Are the systems that needed change the ones you expected? (RQ4; how did time impact
other systems or stakeholders)
11. What has the response from parents been? (RQ4; personal factors)
12. How have the faculty responded? (RQ4; personal factors)
13. What advice would you give to a school looking to make a similar change? (RQ 4;
personal and environmental interactions)
14. If you had it to do over again, what would you do differently? (RQ 4; personal and
environmental interactions)

Abstract (if available)

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Implementing organizational change in a medical school

PDF

First-generation college students and persistence to a degree: an evaluation study

PDF

College preparation coursework opportunities for students of color at the secondary level

PDF

A case Study of gender gaps in the legal profession

PDF

A qualitative examination of the methods church leaders use to increase young adult attendance in Christian churches: an evaluation study

PDF

Corporate innovation labs: exploring the role of university research park innovation lab leaders

PDF

School connectedness and teacher reflective practices

PDF

Better together: teacher attrition, burnout, and efficacy

PDF

Faculty experiences: sense of belonging for sexual and gender minority college students

PDF

School-based interventions for chronically absent students in poverty

PDF

Prior learning assessment portfolios: an evaluation study

PDF

Fostering leadership resilience: examining the influence of social networks on female administrators’ capacity to lead in times of crisis or organizational change

PDF

Challenging stigmas and perceptions in alternative high schools through mentorship: an innovation study

PDF

The role of professional development and certification in technology worker turnover: An evaluation study

PDF

Primary care physicians' experiences working within the patient-portal to improve the quality of patient care

PDF

Advancing equity and inclusion in higher education: the role of the chief diversity officer and the institution in creating more diverse campus climates

PDF

Affluent teens and school stress: an evaluation study

PDF

Understanding teacher leadership styles in relation to student outcomes

PDF

A qualitative analysis on Latino parents' beliefs regarding their middle school child's motivation

PDF

School integration as a reform strategy: the principal’s role, an evaluation study

Asset Metadata

Creator Eisenman, Matthew Stephen (author)

Core Title Organizational barriers to delaying high school start time in independent schools

School Rossier School of Education

Degree Doctor of Education

Degree Program Organizational Change and Leadership (On Line)

Publication Date 03/18/2021

Defense Date 03/08/2021

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag adolescent sleep,High School,independent school,OAI-PMH Harvest,school leadership,school start time,Sleep,start time,start time delay

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Hirabayashi, Kimberly (committee chair), Cash, David (committee member), Muraszewski, Alison (committee member)

Creator Email Eisenman.matt@gmail.com,meisenma@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c89-429054

Unique identifier UC11666623

Identifier etd-EisenmanMa-9332.pdf (filename),usctheses-c89-429054 (legacy record id)

Legacy Identifier etd-EisenmanMa-9332.pdf

Dmrecord 429054

Document Type Dissertation

Rights Eisenman, Matthew Stephen

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA