University of Southern California Dissertations and Theses

Family conflict, negative mood, and adolescents' daily problems in school

(USC Thesis Other)

Family conflict, negative mood, and adolescents' daily problems in school

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content Running head: FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 1

Family Conflict, Negative Mood, and Adolescents’ Daily Problems in School
Adela C. Timmons
University of Southern California

FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 2
Table of Contents
Abstract……………………………..……………………………………….…….…....…………3
Introduction…………………………………………….………………………………..………...4
Method……………………………………………………………………………..…………….10
Results………………………………………………………..……………………..……………15
Discussion…………………………………………………………………………………..........24
References……………………………………………..…………………………………………30
Tables and Figures………………………………………………………….……………………37
Appendix…………………………………………………………………………………………57

FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 3
Abstract
Research indicates that family conflict may interfere with adolescents’ achievement in school.
To understand how family conflict may disrupt academic achievement, several researchers have
examined spillover patterns using daily diary data. The current study expands on extant work by
examining how conflict in specific family dyads is associated with daily negative mood and
problems in school. Participants consisted of 106 adolescents who provided reports of mother-
child conflict, father-child conflict, parent-parent conflict, negative mood, and school problems
daily for 14 days. Both mother-child and father-child conflict were associated with same-day
problems in school but parent-parent conflict was not. Results from a cross-lagged panel model
indicated that the effects were bidirectional such that problems in school predicted next-day
parent-child conflict and that parent-child conflict predicted next-day problems in school. Results
also showed that negative mood (time t) mediated the relationship between parent-child conflict
(time t) and next-day problems in school (time t + 1) and that negative mood (time t + 1)
mediated the relationship between problems in school (time t) and next-day parent-child conflict
(time t + 1). Mediation effects across a 3-day period were not significant. These findings suggest
that parent-child conflict may impact negative mood, which could affect engagement in school
and interfere with adolescents’ academic achievement.
Keywords: family conflict, negative mood, problems in school, daily diary data

FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 4
Family Conflict, Negative Mood, and Adolescents’ Daily Problems in School
Achievement in school is an important developmental task for youth. The school
environment provides a context outside of the home for children to cultivate adult skills and
competencies (Masten & Coatsworth, 1998). During adolescence, the importance of academic
achievement increases. Achievement during high school is often necessary for pursuing
advanced education and entering a specialized profession and may also have implications for
mental and physical health in adulthood. Individuals who have low levels of achievement
typically earn less income, have poorer access to healthcare, and experience more life stress
(Adler, Boyce, Chesney, Folkman, & Syme, 1993; Backlund, Sorlie, & Johnson, 1999).
Moreover, these individuals are at increased risk for a variety of negative health outcomes,
including obesity, depression, and mortality (Backlund et al., 1999; Crimmins & Saito, 2001;
Miech, Caspi, Moffitt, Wright, & Silva, 1999; Taras & Potts-Datema, 2005).
Despite the importance of academic achievement in adolescence, research shows that
achievement and involvement in school may decrease during the teenage years (Anderman &
Midgley, 1997; Barber & Olsen, 2004; Denault & Poulin, 2009; Gottfried, Marcoulides,
Gottfried, Oliver, & Guerin, 2007; Simpson & Oliver, 1990). For example, math and science
achievement decrease in adolescence, and students report feeling less academically competent
across middle school and high school (Anderman & Midgley, 1997; Barber & Olsen, 2004;
Gottfried et al., 2007). Given that underachievement in adolescence is associated with various
academic and occupational outcomes, as well as mental and physical health, it is important to
understand what factors contribute to low achievement at this age. Such work would help
researchers to develop effective interventions for preventing academic declines in adolescence,
which may in turn protect youth from the risks associated with low educational attainment.
FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 5
Family Conflict and Academic Achievement
Research suggests that the family plays an important role in promoting academic
achievement (Amato & Fowler, 2002; Dubois, Eitel, & Felner, 1994; Fan & Chen, 2001;
Majoribanks, 2005; Turner, Chandler, & Heffer 2009). For example, authoritative parenting and
parental involvement are both associated with achievement in adolescence (Fan & Chen, 2001;
Glasgow, Dornbusch, Troyer, Steinberg, & Ritter, 1997; Hill et al., 2004; Turner et al., 2009). In
addition, several studies have documented links between achievement in adolescence and family
conflict (Bahrassa, Syed, Su, & Lee, 2011; Dotterer, Hoffman, Crouter, & McHale, 2008;
Ghazarian & Buehler, 2010; Gordon, Aitken, & Shelton, 2007; King, 1998; Unger, McLeod,
Brown, & Tressell, 2000). Youth from high conflict homes are two to four times more likely to
have low grade point averages than their peers (King, 1998). Moreover, several longitudinal
studies have shown that, after statistically adjusting for initial levels of academic achievement,
family conflict predicts declines in achievement (Bahrassa et al., 2011; Dotterer et al., 2008).
This relationship also appears to be bidirectional: Among families with low education, low
achievement is associated with increases in family conflict, even after statistically adjusting for
the initial quality of the parent-adolescent relationship (Dotterer et al., 2008).
Using Daily Data to Study Problems in School
Although the aforementioned research demonstrates that family conflict is associated
with poor achievement in school, the majority of these studies have used cumulative measures of
achievement, such as GPA and standardized test scores (Bahrassa et al., 2011; Dotterer et al.,
2008; Ghazarian & Buehler, 2010; Gordon et al., 2007; King, 1998; Unger, et al., 2000). An
alternative to these ‘end products’ is to use daily diaries to study the activities of youth on a daily
basis, such as their daily homework and their involvement in school. Assessing adolescents’
FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 6
performance and engagement in school on a daily basis might provide insight into microlevel
processes that are explanatory mechanisms of associations at the macrolevel (Laurenceau &
Bolger, 2005; Repetti, Wang, & Saxbe, 2009). Information on the daily level is important for
developing interventions because interventions are typically targeted at the small, daily steps
toward academic success. In addition to these advantages, daily data assess variables in their
natural context, which increases ecological validity relative to data collected in a laboratory
setting. Finally, daily data can reduce bias due to retrospective reporting because accuracy tends
to decrease as the amount of time between the event and the occasion of reporting elapses
(Bolger, Davis, & Rafaeli, 2003; Laurenceau & Bolger, 2005; Reis, 2012; Trull & Ebner-
Priemer, 2009).
Research on Daily Family Conflict and Problems in School
A handful of studies have begun to use diary studies to link daily family conflict and
problems in school. For example, children’s reports of academic failure are related to their
reports of aversive interactions with parents the same evening (Repetti, 1996), and research
suggests anxiety and self-esteem partially mediate this relationship (Lehman & Repetti, 2007).
Salamon, Johnson, and Sweden (2011) reported similar results in a sample of French adolescents
in which they found reciprocal spillover between family and school events within days. Results
further indicated that the effect remained significant after accounting for immediate mood
responses and that academic difficulty, anxiety, depression, and gender did not moderate the
relationship. In 2008, Flook and Fuligni conducted a similar study with adolescents except that
they examined bidirectional influences across days rather than within days. Results showed that
there were bidirectional influences between family conflict and problems in school up to two
days later. The authors also tested moderating factors, including ethnicity, gender, and level of
FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 7
education. Neither gender nor ethnicity moderated the relationship between family stress and
problems in school. However, spillover across family and school domains was greater in Chinese
females. Furthermore, daily family stress and academic adjustment problems measured at the
beginning of high school predicted lower GPA at the end of high school, statistically adjusting
for previous GPA, suggesting that daily processes may contribute to more cumulative outcomes.
The Present Study
These studies show preliminary connections between how family conflict and school
problems mutually influence each other. However, more information about how spillover
processes unfold is necessary for designing effective interventions. First, it is unclear if different
types of family conflict (mother-child, father-child, and parent-parent) have differential impacts
on adolescents’ daily problems in school and also if the effects are equivalent in both directions
(i.e., family conflict to school problems versus school problems to family conflict). Second, only
one study to our knowledge has examined family-school spillover across days (Flook & Fuligni,
2008). Spillover across days may be an especially important outcome because it suggests that for
some youth, the impacts of family conflict and problems in school are more prolonged. A certain
degree of spillover across different domains is likely normative, but it is unclear how long the
effects last and at what point spillover becomes excessive or destructive to overall functioning.
Thus, it is important to understand the duration and timing of spillover processes in adolescents’
daily lives. Although other work has examined negative mood as a mediator within days (e.g.,
Lehman & Repetti, 2007), no studies to our knowledge have tested negative mood as a mediator
across days.
Furthermore, few moderators of spillover between family conflict and problems in school
have been identified. One likely moderator of this spillover is gender, especially because other
FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 8
work has demonstrated that females exhibit greater emotional reactivity to daily interpersonal
stressors than do males (Flook, 2011). However, diary studies to date have not found gender
differences in this spillover process (e.g., Flook & Fuligni, 2008; Salamon et al., 2011). One
possibility is that gender does not moderate the relationship between family conflict and school
problems, but it does moderate connections between daily family conflict and negative mood.
Therefore, we propose using moderated multilevel mediation (Bauer, Preacher, & Gil, 2006) to
test gender differences in each path in the mediational sequence (i.e., family conflict to negative
mood, negative mood to problems in school, and family conflict to problems in school, as well as
all paths in the opposite direction). We test these moderation effects both within days and across
days because it is possible that females, compared to males, (1) have stronger initial emotional
reactions to family conflict and problems in school and (2) sustain these reactions over longer
periods of time.
In addition to investigating these substantive questions, the present study makes several
methodological contributions. In general, past work using daily data has used single reporters of
daily family conflict, with the exception of one study (Lehman & Repetti, 2007). That study,
which used a sample of fifth graders, did not find a significant association between parents’
reports of aversive parent-child interactions and children’s reports of problems in school,
suggesting either that (1) youth perception of conflict is the most important predictor of problems
in school and/or that (2) using a single reporter creates method bias and therefore contributes to
the association between family conflict and school problems. For example, it is possible that on a
given day children in negative mood states are more likely to report both negative family events
and poor school functioning. Thus, this study uses multiple reporters of different types of family
conflict to determine whether there is within-day agreement across different reporters, as well as
FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 9
to determine whether the relationships remain significant when testing the associations across
different reporters, for example testing the association between fathers’ reports of father-child
conflict and adolescents’ reports of problems in school.
A second methodological contribution of the present study is the use of multilevel
structural equation modeling (MSEM). MSEM is an analysis technique that combines multilevel
modeling (MLM) and structural equation modeling (SEM) (Mehta & Neale, 2005; Muthén &
Asparouhov, 2008) and allows researchers to examine within- and between-level hypotheses
simultaneously (Roesch et al., 2010). For example, one can test whether individuals with high
family conflict are more likely to exhibit problems in school (between level) and simultaneously
test whether on a given day, family conflict is associated with problems in school (within level).
Moreover, using a lagged approach, one can test whether these variables predict outcomes on
later days. A lagged approach provides a stronger test of mediation than cross-sectional tests
because cross-sectional analyses do not account for the temporal sequence of events and may
under- or overestimate a mediation effect (Cole & Maxwell, 2003; Maxwell & Cole, 2007; Selig
& Preacher, 2009). This technique is also useful for testing mediation because unlike traditional
MLM approaches, MSEM provides unconflated estimates of indirect effects (Preacher, Zyphur,
Zhang, 2010) and allows for the testing of alternative mediational sequences in longitudinal data
(Card, 2012). Finally using MSEM, one can test effects in both directions simultaneously (e.g.,
family conflict to school problems and school problems to family conflict), as well as compare
competing models directly by using nested model comparisons.
In summary, the present study tests the bidirectional relationships between all types of
family conflict and problems in school, both within and across days and tests whether negative
mood is a mediator of these within- and across-day links. These analyses will include two
FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 10
possible mediational sequences, first with family conflict and negative mood co-occurring within
days and negative mood linking family conflict and problems in school across days (2-day
mediational sequence) and second with family conflict predicting next-day negative mood and
negative mood predicting next-day problems in school (and effects in the opposite direction; 3-
day mediational sequence). In addition to these main questions, we examine whether gender
moderates any of the paths in the mediational sequence. It is hypothesized that (1) family conflict
and problems in school will be significantly associated and that (2) negative mood will mediate
this relationship. These hypotheses will be tested first within days and then across days. Across-
day analyses will include effects in both directions (i.e., family conflict to problems in school
versus problems in school to family conflict). Finally, it is hypothesized that (3) females will
experience greater negative mood in response to family conflict and problems in school both
within and across days. As exploratory analyses, we will test whether effects are equal across
different types of family conflict and whether effects are equal in both directions.
Method
Overview
The present study uses daily data from the Family Studies Project, a longitudinal research
study examining the relationship between conflict, violence, and developmental outcomes among
adolescents. The original project consisted of 119 families, beginning in 2000. Participants were
recruited through advertisements and flyers posted in the Los Angeles community. All families
underwent an initial phone screening procedure. In order to participate, each family was required
to have of at least one child, age 9-10 at the start of the study, and two parents (defined as either
two parents, parent and step-parent, or parent and significant other). The parents and child were
also required to live together for at least three years prior to participating in the study. Seven
FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 11
years later, a second cohort of 70 families was added to the sample to increase the overall sample
size. The inclusion criteria were the same except that the youth were 13 to 14 years old at the
time of recruitment. This was done so that the new cohort would be the same approximate age as
the original sample.
Participants
A total of 118 adolescents filled out daily questionnaires. However, 12 adolescents filled
out the questionnaires when school was not in session (e.g., summer break). Thus, the current
study uses daily data from 106 youth from both cohorts, which were collected during the fourth
wave of data collection for cohort one and the second wave for cohort two. The adolescents were
between 13 and 17 years of age (M = 15.4, SD = 0.7) and were in grades 8 through 11 (M = 9.9,
SD = 0.7). Of the youth included in the study (54 female, 52 male), 31.1% identified as
Hispanic/Latino. In addition, 50.9% identified as Caucasian, 17.9% as multiple ethnicities,
21.7% as African American, 8.5% as Asian American, and 0.9% as Native American. Mothers (n
= 103) and fathers (n = 100) also provided daily reports of conflict with their spouse and child. In
total, data were provided by both parents in 98 out of 106 cases and by at least one parent in 105
out of 106 cases (one adolescent filled out the home data questionnaires online even though the
parents did not participate in either in-lab or home data procedures). In two cases, all family
members participated in the in-lab procedures, but the fathers elected not to provide home data.
In five other cases, only one of the parents participated in the wave of data collection because the
parents had divorced (resulting in two mothers and three fathers missing). Because families
missing one or both parent reporters could be systematically different than families without
missing reporters (e.g., marital status is likely related to family conflict), all available cases were
FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 12
included in the analyses and multiple imputation was used to reduce bias associated with
nonrandom missingness (see Compliance and Missingness section).
For mothers, age ranged from 35 to 59 (M = 45.4, SD = 6.4) and for fathers, age ranged
from 33 to 72 (M = 48, SD = 6.8). Among the parents, 25% identified as Hispanic/Latino. In
addition, 56.1% identified as Caucasian, 11.3% as multiple ethnicities, 23.6% as African
American, 8.5% as Asian American, and 0.5% as Native American (note that demographic data
were obtained for all parents, whether or not they provided home data). Parents’ years of
education ranged from 7 to 20 (M = 14.9, SD = 2.6) for mothers and 10 to 20 (M = 15, SD = 2.5)
for fathers. Family income varied considerably with: 21.9% ≤ $50,000, 31.3% ≥ $51,000 and ≤
100,000, 26% ≥ $101,000 and ≤ $150,000, and 20.8% ≥ $151,000 (Mdn = $93,500). Parents
had been living together for 14.86 years on average (SD = 5.93), and the number of children in
the family ranged from 1 to 6 (M = 2.6, SD = 1.58). In total, 93.2% of the couples were married,
and 10.7% of the families consisted of at least one non-biological parent.
Procedure
Families took part in a lab session during which they completed a number of
questionnaires and had a family discussion. Prior to leaving the lab, each of the three family
members was instructed in the home data collection procedures and completed the first daily
questionnaire for the immediate preceding day—ending at bedtime the prior night (except for the
two fathers who chose not to provide home data). The family members were instructed to fill out
a daily questionnaire at the end of each day for 13 more days with day 2 to be completed on the
night of the lab meeting. All families were given the choice to use an online system (for
adolescents n = 49; for mothers n = 37; for fathers n = 50) or to complete the questionnaires in
paper format (for adolescents n = 57; for mothers n = 66; for fathers n = 50). To encourage the
FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 13
participants to complete the questionnaires at the end of each day, the online questionnaires were
emailed to each family member (the youth and both parents) at 5:00 in the evening on the day of
reporting. When the online questionnaires were completed, they were automatically uploaded to
a secure online database with a timestamp. For questionnaires completed on paper, participants
recorded the date and time directly on the questionnaires. Participants were instructed to mail the
questionnaires within 24 hours and postmarks were used to assess compliance. Families received
$10 for each day of data provided.
Measures
Parent-Parent Conflict. Adolescents, mothers, and fathers provided daily reports of
parent-parent conflict. On the youth questionnaire, adolescents completed four items assessing
parent-parent conflict, two for mother as the actor and two for father as the actor, i.e., “My mom
yelled at or criticized my dad” and “My dad seemed annoyed at my mom.” The parent
questionnaire contained eight items total—in four, the parent reported on her or his own
behavior, e.g., “I yelled at or criticized my partner/spouse” and in an identical four, the parent
reported on the partner’s behavior, e.g., “My partner/spouse yelled at or criticized me.” All items
on the youth and parent questionnaires ranged from 0 to 3, with 0 being not at all to 3 being a
lot. To compute scores, items were averaged so that higher scores indicate higher amounts of
parent-parent conflict. See the Appendix for a list of the items used in the study. Multilevel
estimates of the internal consistency for all measures were conducted using the equations
outlined by Shrout and Lane (2011). Table 1 presents a complete list of multilevel reliability
estimates for all constructs and reporters. All multilevel reliability estimates for parent-parent
conflict were ≥ .87.
FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 14
Mother-Child and Father-Child Conflict. On the youth questionnaire, adolescents
completed four items pertaining to mother-child conflict and four items pertaining to father-child
conflict. Examples of items assessing mother-child conflict include “My mom seemed irritated
with me today” and “My mom and I argued.” The items assessing father-child conflict were the
same except that “dad” was substituted for “mom.” On the parent questionnaire, mothers and
fathers each completed eight items assessing direct conflict with their child. Examples of items
on the parent questionnaire include “My child was irritated with me” and “I said something mean
to my child.” For all questionnaires, items ranged from 0 to 3 with 0 being not at all and 3 being
a lot. To compute the scores, items were averaged so that higher values indicate higher levels of
mother-child or father-child conflict. For both mother-child and father child conflict, all
multilevel reliability estimates were ≥ to .77.
Negative Mood. Youth completed seven items assessing negative mood. Participants
rated the items on a scale from 0 to 3, with 0 being not at all to 3 being a lot, and the values were
averaged so that higher scores indicate greater negative mood. Examples include “Today I felt
sad” and “Today I felt nervous.” Within-level reliability was estimated to be .80, and between-
level reliability was estimated to be .97.
Problems in School. On the youth questionnaire, seven items assessed problems in
school. These items include behaviors that occur while at school, for example, “I was late for
school or late to a class at school” and “I got a bad grade or did poorly on homework” and other
behaviors that could occur either at school or at home, for example, “I didn’t finish my
homework” and “I can’t understand or can’t do some of my schoolwork.” Participants rated the
items on a scale from 0 to 3, with 0 being not at all to 3 being a lot. Within-level reliability was
estimated to be .70, and between-level reliability was estimated to be .95. The within-level
FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 15
reliability for problems in school was adequate but not as high as the other measures, likely
because youth typically do not exhibit all possible problematic behaviors within a given day
(e.g., cutting class, not finishing homework, getting a bad grade, etc.).
Results
Overview of Analyses. Models were estimated with maximum likelihood estimation in
Mplus Version 7. As previously discussed, MSEM is a flexible statistical method that
decomposes variability into within and between person effects, allowing models to be tested at
both levels of analysis simultaneously. MSEM also allows for testing lagged models at the
within-person level (Card, 2012). All other analyses were conducted in either R Version 2.15 or
SPSS Version 20. Because pooling methods for model fit indices with imputed data have not yet
been developed (Enders, 2010), we report the means and standard deviations. In addition, we
examined the empirical distribution of values for fit indices and found no values outside
acceptable ranges (i.e., greater than .10 for RMSEA, SRMR
within,
and SRMR
between,
and less than
.90 for CFI). For the same reason, nested model comparisons are conducted using Wald tests
rather than chi-square difference tests. Across all analyses, model fit was generally good.
Because of the large number of analyses conducted, fit indices for the preliminary analyses
(correlations, regressions, etc.) are not reported. However, fit indices for all hypothesized path
analysis models are included with the figures. Because of the large number of paths estimated in
the following models, as well as the fact that there is no a priori reason for expecting specific
paths to be zero, non-significant paths were not trimmed in the following analyses.
Sample Selection. A series of tests were conducted to compare the socioeconomic and
demographic breakdown of the current sample to that of Los Angeles County (United States
Census, 2010). Results from a one-sample Wilcoxon signed-rank test showed that the income of
FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 16
the sample (Mdn = 93,500) was greater than that of Los Angeles County (Mdn = 56,266), Z =
6.15, p < .001. In addition, results of exact binomial tests indicated that the gender of the sample
did not differ from Los Angeles County but that the percentage of parents with high school and
college degrees was greater. Moreover, the ethnic/racial composition of the sample varied from
the composition of Los Angeles County such that the current sample had a greater number of
adolescents who identified as African American and multiple ethnicities and had a lesser number
of adolescents who identified as Caucasian and Hispanic/Latino. Results of the exact binomial
tests are included in Table 2. Thus, the racial/ethnic composition of the sample was different than
Los Angeles County but was not necessarily less diverse in general. The composition of the
current sample should be kept in mind when interpreting the results. Because the youth in the
present sample are from more educated, wealthier families, the adolescents may have exhibited
less variability in terms of problems in school, as compared to a more representative sample. It is
possible that this decreased variability could have attenuated the relationship between family
conflict and problems in school in this study, although in general this relationship was found to
be significant, as is discussed in more detail in both the Concurrent and Lagged Analyses
sections.
Compliance and Missingness. The participants were generally compliant with the
protocol. On average, adolescents provided 12.26 days of data, mothers provided 12.08 days, and
fathers provided 11.64 days (Mdn = 14 for all reporters). Of those questionnaires that were
completed, adolescents finished 89.96% within 24 hours of the day of reporting (96.54% within
48 hours), mothers finished 95.95% within 24 hours of the day of reporting (97.64% within 48
hours), and fathers completed 93.91% within 24 hours of the day of reporting (98.04% within 48
hours). Days that were filled out over 5 days after the day of reporting were treated as missing
FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 17
days, resulting in the deletion of 0.86% of days for adolescents, 1.85% for mothers, and 0.97%
for fathers. In total, adolescents provided 1,295 out of 1,484 possible days of data (12.74%
missing) and provided data on 649 school days out of the 1,066 weekdays reported. School days
were likely lost due to sick days and holidays. Out of the possible days that adolescents reported,
mothers provided data on 86.73% of those same days and fathers provided data on 83.02% of
those same days. Some participants failed to provide consecutive daily reports (e.g., completed 1
day, skipped 1 day, and then resumed reporting for another 13 days), in which case a placeholder
was added, and the skipped day was counted as missing. Thus, the measurement domain ranged
from 14 to 21 days, with 96.22% reporting across 14 days, 1.89% reporting across 15 days, and
.94% reporting across 16 days and 21 days. As a result, a total of 14 placeholder days were
added to the dataset, which resulted in 1,498 possible days to be included in the imputation
model (see below for details on the imputation model).
Because missingness is likely systematically related to variables of interest in this study,
an analysis of missingness was conducted using a series of multilevel logistic regressions.
Predictors of day-level missingness that were tested include 13 possible variables for each
reporter: ethnic/racial status, age, report number, day of the week, weekend, income, gender,
years living together, marital status, number of children, grade in school, whether the parents are
married, and the parents’ years of education. Table 3 presents the results of the multilevel
logistic regressions that were significant. In general, missingness increased over time and was
related to several demographic variables, such as youth age and number of kids in the household.
In addition to examining day-level missingness, predictors of missingness for specific variables
(i.e., all types of family conflict, negative mood, and problems in school) were tested for all
reporters. For specific variables, no additional predictors were significant and thus are not
FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 18
shown. All variables that were significantly associated with missingness were included in the
imputation model, as were all of the main variables of interest in the study (i.e., family conflict,
negative mood, problems in school, and gender). Multiple imputation is a method of dealing with
missing data that uses variables that are related missingness to improve the estimation of model
parameters and to reduce bias associated with nonrandom missingness (Enders, 2010).
Specifically, 20 datasets were imputed using a fully saturated random effects multilevel model in
order to estimate all possible relationships, including cross-level interactions. Analyses were then
conducted on all 20 datasets and pooled using guidelines described by Rubin (1987). To
maximize power, the data were imputed at the item level. Values for problems in school were not
imputed on weekend days.
Descriptive Analyses. The number of days in which different types of family conflict
were endorsed is included in Table 4. In addition, means and standard deviations of all variables
by reporter and adolescent gender are included in Table 5 (within level) and Table 6 (between
level). Mean levels of daily family conflict were generally consistent across different reporters,
with the exception that within days, adolescents endorsed parent-parent conflict less often as did
parents (𝛽 = -.14, p < .001; all other p-values > .23), likely because adolescents do not witness all
of the conflict that parents engage in. We also examined whether there are significant differences
in terms of daily levels of different types of family conflict (after averaging across reporters).
Results showed that within-day father-child conflict was significantly lower than within-day
mother-child conflict (𝛽 = -.12 p < .001). Moreover, gender was not associated with parent-
parent conflict, mother-child conflict, father-child conflict, negative mood, or problems in school
(all p-values > .12).
FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 19
Next, we examined agreement across reporters (e.g., youth report of mother-child conflict
with mother report of mother-child conflict). Results are included in Table 7. In summary, all
correlations of the same construct across different reporters were significant both on the within
and between levels. Finally, correlations among different types of family conflict were examined
both within the same reporter (e.g., youth report of mother-child conflict with youth report of
father-child conflict) and across different reporters (e.g., youth report of mother-child conflict
with father report of father-child conflict; refer to Table 8 for specific results). Within the same
reporter, the associations between different types of conflict were all significant. In contrast,
associations across different reporters were inconsistent. However, it was generally found that
correlations between mother-child and father-child conflict were significant on the within level
and that correlations between mother-child and parent-parent conflict were significant on the
between level. Finally, after averaging scores across reporters, all correlations among different
types of family conflict were significant at both the within and between levels of analysis.
After examining correlations among different types of family conflict both within and
across reporters, we tested whether any demographic variables were related to daily problems in
school and found no significant associations (all p-values > .20). Because previous research with
daily data has found that adolescents’ endorsement of negative events decreases over time
(Nishina, 2012), we examined whether the day of reporting was associated with each type of
family conflict, as well as negative mood and problems in school. Results indicated that report
number was negatively associated with negative mood, problems in school, child-reported
parent-parent conflict, mother reported mother-child conflict, and mother reported parent-parent
conflict. Because the degree of family conflict and negative mood may change depending on the
day of the week (e.g., mood may drop on Mondays, more conflict may occur on Sundays when
FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 20
families spend more time at home), the association between these variables was examined.
Results showed that day of the week was not significantly associated with any of these variables
but that several types of family conflict were more likely to occur on weekend days (See Table 9
for these results). Because both report number and weekend were significantly associated with
variables of interest in this study, they are included as covariates on the within level in all
subsequent analyses. Because no school data were reported on weekends, weekend was not
included as a covariate for pathways in which the outcome variable was problems in school.
Within-Day Analyses. Because the relationship between family conflict, negative mood,
and problems in school may vary by reporter, we first conducted analyses separately for youth,
mother, and father report (see Table 10 for within-level analyses and Table 11 for between-level
analyses). Results showed that for youth report, both mother-child and father-child conflict were
significantly associated with negative mood and problems in school on both levels of analysis.
Parent-parent conflict was significantly associated with negative mood on the within but not the
between level and was not associated with problems in school at either level. Mother report of
mother-child conflict was associated with youth report of negative mood on both levels of
analysis, and father report of father-child conflict was significantly associated with youth report
of negative mood on the within level and problems in school on both the within and between
levels. Neither mother report nor father report of parent-parent conflict was significantly
associated with child report of negative mood or problems in school. Because the correlations
between different reporters were not high enough to create well-fitting latent factors, we then
proceeded by averaging scores across reporters. After averaging scores, all associations between
family conflict, negative mood, and problems in school were significant, with the exception of
FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 21
paths that included parent-parent conflict, which were not significant (see Table 12). Thus,
parent-parent conflict was dropped from subsequent analyses.
After conducting these analyses, we added mother-child and father-child conflict to the
same regression model. Both mother-child (𝛽 = .13, p = .03) and father-child conflict (𝛽 = .18, p
= .007) were associated with same-day problems in school (hypothesis 1, within day). However,
effects on the between level were no longer significant (both p-values > .14), except for the
correlation between mother-child and father-child conflict (r = .66, p < .001). After these initial
analyses, we proceeded to construct models and conduct nested model comparisons. First, we
tested whether negative mood mediates the same-day association between parent-child conflict
and problems in school (hypothesis 2, within day). Thus, we tested separate mediation paths for
both predictors (mother-child and father-child conflict) using Monte Carlo simulated confidence
intervals (Selig & Preacher, 2008). Results showed that on the within level, negative mood fully
mediated the relationship between mother-child conflict and problems in school (𝛽 = .04, 95%
CI [.02-.06]) and partially mediated the relationship between father-child conflict and problems
in school (𝛽 = .04, 95% CI [.01-.07]). In contrast, all paths on the between level were not
significant, with the exceptions that negative mood was associated with problems in school (𝛽 =
.57, p < .001) and that mother-child and father-child conflict were significantly correlated (r =
.74, p < .001). Because few paths on the between level were significant, only the within-level
results are depicted in Figure 1 (although model fit indices reflect the entire model).
Next, we conducted Wald tests to determine whether mother-child and father-child
conflict had differential effects on daily negative mood and problems in school. Results indicated
that there was not a significant loss of model fit associated with constraining the paths for
mother-child and father-child conflict to be equal (∆χ² [1, N = 1,498] = 0.04, p = .84 for negative
FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 22
mood and ∆χ² [1, N = 1,498] = 1.04, p = .31 for problems in school). Finally, we tested whether
paths across levels were equivalent (among significant paths only). Constraining the correlations
between mother-child and father-child conflict across the within and between levels did not
result in a significant loss of model fit (∆χ² [1, N = 1,498] = 0.52, p = .47). However,
constraining the within- and between-level paths for the association between negative mood and
problems in school resulted in a significant loss of model fit (∆χ² [1, N = 1,498] = 11.35, p <
.001), indicating that the effect on the between level was greater than on the within level.
Negative mood mediated the within-day relationships between mother-child conflict and
problems in school and father-child conflict and problems in school (𝛽 = .04, 95% CI [.02-.06]
for both predictors). Thus, the results were generally the same in the constrained model, except
negative mood partially mediated the relationship between mother-child conflict and problems in
school, rather than fully mediated. The final model with equality constraints is depicted in Figure
2.
Lagged Analyses. Our next objective was to examine the bidirectional relationships
between parent-child conflict and problems in school using cross-lagged panel models. To
conduct the lagged analyses, an extension of the method described in Bolger et al. (2003) and
Hawkley, Preacher, and Cacioppa (2007) was used. Specifically, we created lagged versions of
the variables and used the lagged variables (at time t) to predict outcomes at later time points (t +
1 and t + 2). For all across-day analyses, models were tested on the within level only. To reduce
the complexity of subsequent analyses, we collapsed mother-child and father-child conflict into a
single variable of parent-child conflict. Results from the cross-lagged panel model indicated that
the effects were bidirectional such that problems in school predicted next-day parent-child
conflict, and parent-child conflict predicted next-day problems in school (hypothesis 1, across
FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 23
day). Moreover, the association across days remained significant up to two days later. We also
tested whether the bidirectional effects were equivalent (i.e., parent-child conflict on problems in
school versus problems in school on parent-child conflict). Results of Wald tests indicated no
significant differences (all p–values > .36). The constrained models for lagging 1 and 2 days are
included in Figures 3 and 4, respectively.
To determine whether negative mood mediates the across-day associations between
parent-child conflict and problems in school, we examined a series of models using different
temporal sequences. In the first model, we tested whether negative mood (t) mediates the
relationship between parent-child conflict (t) and next-day problems in school (t + 1) and
whether negative mood (t + 1) mediates the relationship between problems in school (t) and
next-day parent-child conflict (t + 1). Results indicated that negative mood fully mediated the
relationship between parent-child conflict and next-day problems in school (𝛽 = .04, 95% CI
[.01-.08]) and partially mediated the relationship between problems in school and next-day
parent-child conflict (𝛽 = .03, 95% CI [.01-.07]; Figure 5; hypothesis 2, across day, 2-day). In
the final mediation model, we tested whether negative mood (t + 1) mediates the relationship
between parent-child conflict (t) and problems in school (t + 2) and whether negative mood (t +
1) mediates the relationship between problems in school (time t) and parent-child conflict (t + 2).
The results showed that negative mood did not mediate the relationship between parent-child
conflict and problems in school two days later (𝛽 = .002, 95% CI [-.01-.01]) and did not mediate
the relationship between problems in school and parent-child conflict two days later (𝛽 = .002,
95% CI [-.001-.01]; hypothesis 2, across day, 3-day). Because of the large number of estimated
paths, the complete model is depicted in Figure 6 (without coefficients) and the significant paths
(with coefficients) are included in Figure 7.
FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 24
Moderated Multilevel Mediation Analyses. The final goal of the study was to
determine whether gender moderates any of the paths in the mediation models (i.e., parent-child
conflict to negative mood, negative mood to problems in school, parent-child conflict to
problems in school, and these effects in the opposite direction for the lagged analyses; hypothesis
3). We limited our analyses to the two statistically significant mediation models: the within-day
and 2-day mediational sequences. For the within-day mediation model, which included effects
for mother-child and father-child conflict, gender was not a significant moderator of any of the
mediation paths (all p-values > .19). In the 2-day mediation model, which included the collapsed
variable parent-child conflict, gender significantly moderated the within-day relationship
between parent-child conflict and negative mood at time t (𝛽 = .13, p = .04) such that within-day
spillover between parent-child conflict and negative mood was greater among females (𝛽
females
=
.31, p < .001; 𝛽
males
= .18, p < .001). None of the other paths in the mediational sequence were
significant (all p-values > .08). Figure 8 depicts this interaction effect.
Discussion
Results of the current study suggest that there are reciprocal influences between family
and school domains in adolescents’ daily lives. The first hypothesis that family conflict and
problems in school are linked within days and across days was supported, except for the case of
parent-parent conflict. Because the present study used a community-based sample, it likely did
not include the severest levels of marital conflict, as might have been seen in a more high-risk
sample. Also, adolescents likely spend less time at home than do younger children and thus may
witness less marital conflict than younger age groups. Thus, the impact of parent-parent conflict
on youth’s daily problems in school should be examined in children of different ages. The effects
of mother-child conflict versus father-child conflict were not significantly different from each
FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 25
other, and there was no evidence that across-days effects were stronger in one direction versus
the other. The second hypothesis that negative mood mediates spillover across family and school
was supported in both the within-day and 2-day models but not in the 3-day model. Finally, the
third hypothesis that gender moderates the relationship between family conflict and negative
mood was supported within days but not across days.
In total, this study replicates findings from a nascent literature on daily family conflict
and problems in school and builds on this early literature by exploring in more detail how
spillover processes unfold. Specifically, the present study examined conflict between specific
dyads, tested the strength of effects in both directions simultaneously, examined negative mood
as a mediator within and across days, and explored the timing and duration of these effects.
Results suggest that parent-child conflict and school problems may have reciprocal influences
and that problems in one domain may exacerbate problems in other domains, creating a cycle of
events that could become engrained over time. These findings also highlight the role of negative
mood in linking negative events across different domains. When adolescents are in negative
mood states, they may be more likely to engage in conflict with others. Similarly, negative
moods may cause adolescents to experience lowered concentration or motivation in school. If an
adolescent receives a poor grade in school, he or she may feel angry or irritated, which could
increase the likelihood of fighting with parents. Engaging in conflict with parents may then
exacerbate negative mood and thereby increase the likelihood that adolescents will experience
difficulty in school.
If adolescents cannot recover from negative moods that originate in one life domain, they
may be more susceptible to experience problems in other domains, which could perpetuate stress
over time. Most adolescents likely experience some spillover across different domains, but high
FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 26
levels of emotional reactivity could negatively impact functioning by creating a feedback loop
that causes functioning to spiral downward. Although research indicates that spillover occurs at
multiple developmental stages (e.g., Lehman & Repetti, 2007; Schulz, Cowan, Pape Cowan, &
Brennan, 2004), adolescence may be a particularly important time to study spillover processes.
As youth enter adolescence, they experience a greater number of stressful life events in family,
school, and peer domains and have stronger emotional reactions to negative events (Larson &
Ham, 1993). Moreover, levels of daily positive mood decrease in the teenage years (Weinstein,
Mermelstein, Hankin, Hedeker, & Flay, 2007). As a result, it is possible that spillover in
adolescence could impact the developmental trajectory of youth and thus contribute to negative
outcomes in adulthood.
A corollary to the above arguments is the concept of stress generation, which has been
explored more extensively in the context of depression (Hammen, 2006). In this model, people
who experience depression are not passively shaped by environmental factors. Rather, there is an
interaction between personal characteristics and the environment, such that stressors contribute
to distress but the individual reacts in ways that increase the likelihood of future stressors. Over
time, this interaction is thought to contribute to the occurrence and maintenance of depression.
Similarly, when negative events occur in one life domain, such as school, adolescents may carry
that stress to other life domains. Thus, the adolescent may become caught in a self-perpetuating
negative cycle that could contribute to the development of psychopathology and other outcomes,
such as academic achievement. More longitudinal work examining how spillover contributes to
later, more global outcomes is an important area of future work.
Another important area of future research is identifying who is at-risk for spillover in
daily life. Currently, few moderators of spillover between family conflict and problems in school
FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 27
have been identified (e.g., Flook & Fuligni, 2008; Salamon et al., 2011), even though previous
work indicates that females are more emotionally reactive to daily interpersonal stress than are
males (Flook, 2011). The current study used moderated multilevel mediation to demonstrate that
gender moderates the same-day relationship between parent-child conflict and negative mood,
rather than the direct effect between parent-child conflict and school problems. Similarly, it is
important to identify buffers of spillover, both in terms of the initial response to the negative
event and the ability to recover over time. In the current analyses, spillover occurred both within
days and across days. Across-day spillover may be especially important because it suggests that
some adolescents have difficulty recovering from negative events. Prolonged reactions to daily
negative events may increase the likelihood of spillover across domains and thus may have a
greater impact on overall functioning. Future work should continue to examine moderators of
spillover (at all points in mediation sequence) to target those youth who are most at-risk, as well
as to identify those who are most protected from spillover processes.
The present study also has implications for intervention with families and adolescents.
Daily diary studies are well-suited to informing intervention because they provide information
about the daily behaviors, i.e., microlevel processes, such as studying or paying attention in
class, that might contribute to macrolevel outcomes, such as standardized test scores or
graduation rates (Repetti et al., 2009). This information is especially useful because interventions
are typically targeted at daily behaviors, rather than cumulative outcomes. It is possible that
preventing negative events in one domain of functioning could precipitate changes in other
domains. Moreover, interrupting the mechanisms that transmit problems across domains may be
a particularly effective point of intervention for preventing spillover. For example, helping youth
recover from negative events that occur at school might decrease the likelihood that problems in
FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 28
school will translate into problems at home. If clinicians can interrupt the feedback loop between
negative events at home and school, this may prevent some adolescents from spiraling downward
and perhaps set them on an upward trajectory to achieve in school.
In addition to examining these substantive questions, the current study made several
methodological contributions. First, this study used multiple reporters of family conflict.
Although the results were generally consistent across different reporters (with some exceptions),
it is still unclear to what degree reporter bias impacts daily data results. One option is to create
latent variables, although the correlations across reporters were not high enough to achieve good
model fit in this sample. Another interesting possibility is that discrepancies in reporting are
related to youth functioning, as has been demonstrated in other studies (e.g., Borelli, Luthar, &
Suchman, 2010; De Los Reyes & Kazdin, 2006) but not, to our knowledge, with daily data. If
parents fail to recognize, acknowledge, or discuss conflict in the home, adolescents may feel
invalidated or disconnected from others in the family, which could contribute to the development
of internalizing or externalizing symptoms. The second methodological contribution was the use
of MSEM to examine alternative temporal sequences and to parse out how these effects play out
on a daily basis. Using cross-lagged panel models, we examined bidirectional influences, which
provided a stronger test of directional hypotheses than has yet been used in previous work. In
addition, this analytic framework allowed us to conduct model comparisons to test the equality
of different paths in the model.
Although the present study has several strengths, a number of limitations must be noted.
First, causal relationships between family conflict, negative mood, and problems in school
cannot be proven. It is possible that other variables better account for the observed relationships.
Moreover, the mediation effects across the 3-day period were not significant. In the 2-day
FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 29
mediation model, parent-child conflict and negative mood were linked within days, and negative
mood and problems in school were linked across days. Thus, the temporal sequence between
parent-child conflict and negative mood could not be established in these data. Future work
should include multiple measurements of parent-child conflict and negative mood within days, as
well as across days, because some variables may influence one another within a day, while
others might have more lasting effects across days.
Another limitation this present study is that we only examined two domains (family and
school) in adolescents’ daily lives, even though other domains are likely relevant as well. For
example, a number of studies have begun to examine daily peer processes and its relation to
other domains in adolescents daily lives (e.g., Chung, Flook, & Fuligni, 2011). Adolescence is a
time typically marked by greater emphasis on peer groups and a lesser emphasis on the family
(Larson, & Richards, 1991). In the current study, it appears that the family is still an important
predictor of adolescent functioning. However, the full influence of peers in daily lives, either in
terms of protective or risk factors is not fully understood and is likely an important component of
daily functioning. Other important considerations include parents’ responses to youth’s negative
moods, the role of sibling conflict in these processes, and the possibility of conflict spreading
from one dyad to other dyads in the family system, as has been shown in other work (e.g.,
Almeida, Wethington, & Chandler, 1999). Despite these limitations, the present study provides
important information in terms of how family conflict may affect adolescent functioning,
especially in terms of daily negative mood and functioning in school. Achievement in school is a
critical task in adolescence and may have implications for future life success. Understanding
what factors promote and inhibit school performance is crucial for promoting academic
achievement in America’s youth.
FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 30
References
Adler, N. E., Boyce, T., Chesney, M. A., Folkman, S., & Syme, L. (1993). Socioeconomic
inequalities in health: No easy solution. The Journal of the American Medical
Association, 269(24), 3140-3145.
Almeida, D. M., Wethington, E., & Chandler, A. L. (1999). Daily transmission of tensions
between marital dyads and parent-child dyads. Journal of Marriage and the Family.
61(1), 49-61.
Amato, P. R., & Fowler, F. (2002). Parenting practices, child adjustment, and family diversity.
Journal of Marriage and Family, 64(3), 703-716.
Anderman, E. M., & Midgley, C. (1997). Changes in achievement goal orientations, perceived
academic competence, and grades across the transition to middle-level schools.
Contemporary Educational Psychology, 22, 269–298.
Backlund, E., Sorlie, P. D., & Johnson, N. J. (1999). A comparison of the relationships of
education and income with mortality: The national longitudinal mortality study. Social
Science and Medicine, 49, 1373-1384.
Bahrassa, N. F., Syed M., Su, J., & Lee, R. M. (2011). Family conflict and academic
performance of first-year Asian American undergraduates. Cultural Diversity and Ethnic
Minority Psychology, 17(4), 415-426.
Barber, B. K., & Olsen, J. A. (2004). Assessing the transitions to middle and high school.
Journal of Adolescent Research, 19(1), 3-30.
Bauer, D. J., Preacher, K. J., & Gil, K. M. (2006). Conceptualizing and testing random indirect
effects and moderated mediation in multilevel models: New procedures and
recommendations. Psychological Methods, 11, 142-163.
FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 31
Bolger, N., Davis, A., & Rafaeli, E. (2003). Diary methods: Capturing life as it is lived.
Annual Review of Psychology, 54, 579-616.
Borelli, J. L., Luthar, S. S., & Suchman, N. E. (2010). Discrepancies in perceptions of
maternal aggression: Implications for children of methadone-maintained mothers.
American Journal of Orthopsychiatry, 80(3), 412-421.
Card, N. A. (2012). Multi-level mediation analysis in the study of daily lives. In M. R. Mehl &
T. S. Conner (Eds.), Handbook of research methods for studying daily life (pp. 479-494).
New York, NY: The Guilford Press.
Chung, G. H., Flook, L., & Fuligni, A. J. Reciprocal associations between family and peer
conflict in adolescents’ daily lives. Child Development, 82(5), 1390-1396.
Cole, D. A., & Maxwell, S. E. (2003). Testing meditational models with longitudinal data:
Questions and tips in the use of structural equation modeling. Journal of Abnormal
Psychology, 112, 558-577.
Crimmins, E. M., & Saito, Y. (2001) Trends in healthy life expectancy in the United States,
1970-1990: Gender, racial, and educational differences. Social Science and Medicine, 52,
1629–1641.
De Los Reyes, A., & Kazdin, A. E. (2006). Informant discrepancies in assessing child
dysfunction within mother-child interactions. Journal of Child and Family Studies, 15,
645-663.
Denault, A., & Poulin, F. (2009). Intensity and breadth of participation and organized activities
during the adolescent years: Multiple associations with youth outcomes. Journal of Youth
and Adolescence, 38, 1129-1213.
Dotterer, A. M., Hoffman, L., Crouter, A. C., & McHale, S. M. (2008). A longitudinal
FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 32
examination of the bidirectional links between academic achievement and parent-
adolescent conflict. Journal of Family Issues, 29(6), 762-779.
Dubois, D. L., Eitel, S. K., & Felner, R. D. (1994). Effects of family environment and parent-
child relationships on school adjustment during the transition to early adolescence.
Journal of Marriage and Family, 56, 405-414.
Enders, C. (2010). Applied Missing Data Analysis. New York, NY: Guilford Press.
Fan, X., & Chen, M. (2001). Parental involvement and students’ academic achievement: A meta-
analysis. Educational Psychology Review, 13(1), 1-22.
Flook, L. (2011). Gender differences in adolescents’ daily interpersonal events and well-being.
Child Development, 82(2), 454-461.
Flook, L., & Fuligni, A. J. (2008). Family and school spillover in adolescent’ daily lives. Child
Development, 79(3), 776-787.
Ghazarian, S. R., & Buehler, C. (2010). Interparental conflict and academic achievement: An
examination of mediating and moderating factors. Journal of Youth and Adolescence, 39,
23-35.
Glasgow, K. L., Dornbusch, S. M., Troyer, L., Steinberg, L., & Ritter P. L. (1997). Parenting
styles, adolescents’ attributions, and educational outcomes in nine heterogeneous high
schools. Child Development, 68(3), 507-529.
Gottfried, A. E., Marcoulides, G. A., Gottfried, A. W., Oliver, P. H., & Guerin, D. W. (2007).
Multivariate latent change modeling of developmental decline in intrinsic math
motivation and achievement: Childhood through adolescence. International Journal of
Behavioral Development, 31(4), 317-327.
Gordon, G. T., Aitken, J. J., & Shelton, K. H. (2007). Inter-parental conflict and children’s
FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 33
academic attainment: A longitudinal analysis. Journal of Child Psychology and
Psychiatry, 48(12), 1223-1232.
Hammen, C. (2006). Stress generation in depression: Reflections on origins, research, and future
directions. Journal of Clinical Psychology, 62(9), 1065-1082.
Hawkley, L. C., Preacher, K. J., & Cacioppo, J. T. (2007). Multilevel modeling of social
interactions and mood in lonely and socially connected individuals: The MacArthur
social neuroscience studies. In A. D. Ong & M. & H. M. van Dulmen (Eds.), Oxford
handbook of methods in positive psychology (pp. 559-575). New York, New York:
Oxford University Press.
Hill, N. E., Castellino, D. R., Lansford, J. E., Nowlin, P., Dodge, K. A., Bates, J. E., & Pettit, G.
S. (2004). Parent academic involvement as related to school behavior, achievement, and
aspirations: Demographic variations across adolescence. Child Development, 75(5),
1491-1509.
King, A. R. (1998). Family environment and scale predictors of academic performance.
Psychological Reports, 83, 1319-1327.
Larson, R., & Ham, M. (1993). Stress and “storm and stress” in early adolescence: The
relationship of negative events with dysphoric affect. Developmental Psychology, 29(1),
130-140.
Larson, R., & Richards, M. (1991). Daily companionship in late childhood and early
adolescence: Changing developmental contexts. Child Development, 62(2), 284-3000
Laurenceau, J. P., & Bolger, N. (2005). Using diary methods to study marital and family
processes. Journal of Family Psychology, 19(1), 86-97.
Lehman, B. J., & Repetti, R. L. (2007). Bad days don’t end when the school bell rings: The
FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 34
lingering effects of negative school events on children’s mood, self-esteem, and
perceptions of parent-child interaction. Social Development, 16(3), 596-618.
Majoribanks, K. (2005). Correlations among family environment, academic achievement, and
academic attainment in a large sample of young Australian adults. Psychological Reports,
97, 639-644.
Masten, A. S., & Coatsworth, J. D. (1998). The development of competence in favorable and
unfavorable environments. American Psychologist, 53(2), 205-220.
Maxwell, S. E., & Cole, D. A. (2007). Bias in cross-sectional analyses of longitudinal mediation.
Psychological Methods, 12(1), 23-44.
Mehta, P. D., & Neale, M. C. (2005). People are variables too: Multilevel structural equation
modeling. Psychological Methods, 10(3), 259-284.
Miech, R. A. Caspi, A. Moffitt, T. E., Wright, B. R. E., & Silva, P. A. (1999). Low
socioeconomic status and mental disorders: A longitudinal study of selection and
causation during young adulthood. American Journal of Sociology, 104, 1096-1131.
Muthén, B. O., & Asparouhov, T. (2008). Growth mixture modeling: Analysis with non-
Gaussian random effects. In G. Fitzmaurice, M. Davidian, G. Verbeke, & G.
Molenberghs (Eds.), Longitudinal data analysis (pp. 143–165). Boca Raton, FL:
Chapman & Hall/CRC.
Nishina, A. (2012). Adolescent daily and general maladjustment: Is there reactivity daily
repeated measures methodologies? Child Maladjustment, 83(2), 405-412.
Preacher, K. J., Zyphur, M. J., & Zhang, Z. (2010). A general MSEM framework for assessing
multilevel mediation. Psychological Methods, 15(3), 209-233.
Reis, H. T. (2012). Why researchers should think “real-world”: A conceptual rationale. In M. R.
FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 35
Mehl & T. S. Conner (Eds.), Handbook of research methods for studying daily life (pp.
479-494). New York, NY: The Guilford Press.
Repetti, R. (1996). The effects of perceived daily social and academic failure experiences on
school-age children’s subsequent interactions with parents. Child Development, 67(4),
1467-1482.
Repetti, R., Wang, S., & Saxbe, D. (2009). How outside stressors shape families’ everyday lives.
Current Direction in Psychological Science, 18(2), 106-111.
Roesch, S. C., Aldridge, A. A., Stocking, S. N. Villodas, F., Leung, Q., Bartley, C.E, & Black, L.
J. (2010). Multilevel factor analysis and structural equation modeling of daily diary
coping data: Modeling trait and state variation. Multivariate Behavioral Research, 45,
767-789.
Rubin, D.B. (1987) Multiple Imputation for Nonresponse in Surveys. New York: John Wiley &
Sons, Inc.
Salamon, R., Johnson, E. I., & Swendsen, J. (2011). Daily life mechanisms of stress spillover
among early adolescents experiencing academic difficulty. European Journal of
Psychological Education. 453-463.
Schulz, M. S., Cowan, P. A., Pape Cowan, C., & Brennan, R. T. (2004). Coming home
upset: Gender, marital satisfaction, and the daily spillover of workday experience into
couple interactions. Journal of Family Psychology, 18(1), 250-263.
Selig, J. P., & Preacher, K. J. (2008, June). Monte Carlo method for assessing mediation: An
interactive tool for creating confidence intervals for indirect effects [Computer software].
Available from http://quantpsyc.org/.
Selig, J. P., & Preacher, K. J. (2009). Mediation models for longitudinal data in developmental
FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 36
research. Research in Human Development, 6, 144-164.
Shrout, P. E., & Lane, S. P. (2012). Psychometrics. In M. R. Mehl & T. S. Conner (Eds.),
Handbook of research methods for studying daily life (pp. 479-494). New York, NY: The
Guilford Press.
Simpson, R. D., & Oliver, J. S. (1990). A summary of the major influences on attitude towards
achievement in science among adolescent students. Science Education, 74(1), 1-18.
Taras, H., & Potts-Datema, W. (2005). Obesity and student performance at school. The
Journal of School Health, 75(8), 291-295.
Trull, T. J., & Ebner-Priemer, U. W. (2009). Using experience sampling methods/ecological
momentary assessment (ESM/EMA) in clinical assessment and clinical research:
Introduction to the special section. Psychological Assessment, 21(4), 457-462.
Turner, E. A., Chandler, M., & Heffer, R. W. (2009). The influence of parenting styles,
achievement motivation, and self-efficacy on academic performance in college students.
Journal of College Student Development, 50(3), 337-346.
Unger, D. G., McLeod, L. E., Brown, M. B., & Tressell, M. S. (2000). The role of family support
in interparental conflict and adolescent academic achievement. Journal of Child and
Family Studies, 9(2), 191-202.
U.S. Census Bureau. (2010). State and county quickfacts. Retrieved from
http://quickfacts.census.gov/qfd/states/06/06037.html.
Weinstein, S. M., Mermelstein, R. J., Hankin, B. L., Hedeker, D., & Flay, B. R. (2007).
Longitudinal patterns of daily affect and global mood during adolescence. Journal of
Research on Adolescence, 17(3), 587-600.

FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 37
Tables and Figures
Table 1
Within- and Between-Level Alpha Coefficients

Construct Number of Items Within-Level 𝛼 Between-Level 𝛼
Youth Report
Mother-Child Conflict 4 .90 .96
Father-Child Conflict 4 .77 .92
Parent-Parent Conflict 4 .88 .96
Negative Mood 7 .80 .97
Problems in school 7 .70 .95

Mother Report

Mother-Child Conflict 8 .87 .98
Parent-Parent Conflict 8 .91 .99

Father Report

Father-Child Conflict 8 .87 .98
Parent-Parent Conflict 8 .87 .98

FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 38
Table 2

Exact Binomial Tests Comparing Sample Characteristics to Los Angeles County
Sample Characteristic Expected % Expected n Observed n 95% CI p
Parents Education

High School Degree 76.1 161 203 [.92-.98] < .001
College Degree 29.2 62 110 [.45-.59] < .001

Youth Gender

Female 50.7 54 54 [.41-.60] 1.0

Youth Ethnicity/Race

Native American 1.5 2 1 [.01-05] 1.0
Asian American 14.2 15 14 [.04-16] .10
African America 9.3 10 23 [.14-.31] < .001
Multiple Ethnicities 2.8 3 19 [.11-.27] < .001
Caucasian 71.8 76 54 [.41-.61] < .001
Hispanic/Latino 48.1 51 34 [.23-.42] < .001

Note. Expected values are based on the 2010 United States Census. n = 212 for parents and 106
for youth. Demographic data were obtained for all parents, whether or not they provided home
data

FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 39
Table 3
Multilevel Logistic Regressions to Predict Day-Level Missingness
Predictor 𝑏 SE 𝑏/SE p
Youth Missingness
Father Hispanic/Latino -2.54 .77 -3.28 < .01
Mother Hispanic/Latino -3.50 .77 -4.53 < .001
Report Number .29 .01 49.33 < .001
Weekend -.82 .24 -3.36 < .01
Youth Age 1.70 .41 4.19 < .001
Youth Ethnicity/Race .74 .34 2.19 .03
Youth Hispanic/Latino 4.54 .99 4.57 < .001

Mother Missingness
Father Years Education .51 .25 2.04 .04
Mother Age -.10 .05 -2.25 .03
Number of Kids .55 .24 2.28 .02
Report Number .37 .10 3.67 < .001
Youth Age 1.90 .72 2.63 < .01
Youth Hispanic/Latino 2.20 .74 3.00 < .01

Father Missingness
Mother Hispanic/Latino -.66 .34 -1.97 .05
Number of Kids .02 .21 4.46 < .001
Parents Married 3.84 1.94 1.99 .05
Report Number .22 .07 3.01 < .01

Note. N = 106. Because these analyses were not of substantive interest but rather meant for
selecting auxiliary variables to reduce bias from nonrandom missingness, we do not include
complete information (i.e., all 13 predictors for all three reporters and all χ² statistics, etc.).
Tested predictors include ethnic/racial status, age, report number, day of the week, weekend,
income, gender, years living together, marital status, number of children, grade in school,
whether the parents are married, and the parents’ years of education.

FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 40
Table 4
Daily Endorsement of Family Conflict, Negative Mood, and Problems in School

Construct Reported Days Endorsed Days %
Youth Report
Mother-Child Conflict 1295 431 33.28
Father-Child Conflict 1295 377 29.12
Parent-Parent Conflict 1295 237 18.30
Negative Mood 1295 736 56.83
Problems in school 1066 547 51.31

Mother Report

Mother-Child Conflict 1297 426 32.85
Parent-Parent Conflict 1297 363 27.99

Father Report

Father-Child Conflict 1242 224 18.04
Parent-Parent Conflict 1242 347 27.94

FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 41
Table 5
Within-Level Means (Standard Deviations) by Reporter and Adolescent Gender

Construct Entire Sample Males Females
Mother-Child Conflict
Youth Report .23 (.46) .27 (.47) .20 (.44)
Mother Report .20 (.40) .16 (.33) .24 (.46)
Average Report .22 (.35) .23 (.33) .21 (.37)

Father-Child Conflict

Youth Report .14 (.34) .19 (.39) .11 (.31)
Father Report .14 (.34) .12 (.29) .15 (.38)
Average Report .14 (.26) .16 (.26) .12 (.25)

Parent-Parent Conflict

Youth Report .09 (.31) .09 (.29) .09 (.33)
Mother Report .18 (.44) .13 (.36) .23 (.50)
Father Report .21 (.41) .19 (.38) .23 (.43)
Average Report .16 (.28) .14 (.25) .18 (.31)

Negative Mood .29 (.40) .31 (.43) .28 (.38)
Problems in School .44 (.38) .44 (.42) .43 (.35)

Note. Reports of negative mood and problems in school are only provided by the youth.

FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 42
Table 6
Between-Level Means (Standard Deviations) by Reporter and Adolescent Gender

Construct Entire Sample Males Females
Mother-Child Conflict
Youth Report .24 (.22) .29 (.27) .20 (.14)
Mother Report .20 (.25) .16 (.17) .23 (.30)
Average Report .23 (.20) .23 (.19) .22 (.21)

Father-Child Conflict

Youth Report .15 (.16) .19 (.19) .11 (.11)
Father Report .15 (.20) .14 (.15) .17 (.23)
Average Report .15 (.12) .17 (.13) .12 (.11)

Parent-Parent Conflict

Youth Report .09 (.13) .09 (.12) .09 (.13)
Mother Report .20 (.29) .15 (.22) .24 (.32)
Father Report .21 (.24) .19 (.21) .24 (.26)
Average Report .16 (.16) .14 (.12) .19 (.19)

Negative Mood .28 (.24) .29 (.28) .27 (.20)
Problems in School .42 (.25) .45 (.29) .40 (.21)

Note. Reports of negative mood and problems in school are only provided by the youth.

FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 43
Table 7
Multilevel Correlations between Different Reporters (of the Same Construct)

Construct Within Level Between Level
Mother-Child Conflict
Child Report with Mother Report .30* .54*
Father-Child Conflict

Child Report with Father Report .23* .34*

Parent-Parent Conflict

Child Report with Mother Report .25* .70*
Child Report with Father Report .27* .55*
Mother Report with Father Report .50* .52*

Note. * = significant at p < .05.

FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 44
Table 8
Multilevel Correlations Between Different Types of Family Conflict

Construct Within Level Between Level
Correlations within Reporters

(Y) Mother-Child Conflict with (Y) Father-Child Conflict .46* .70*
(Y) Mother-Child Conflict with (Y) Parent-Parent Conflict .12* .37*
(Y) Father-Child Conflict with (Y) Parent-Parent Conflict .20* .35*
(M) Mother-Child Conflict with (M) Parent-Parent Conflict .21* .76*
(F) Father-child Conflict with (F) Parent-Parent Conflict .35* .69*

Correlations across Reporters

(Y) Mother-Child Conflict with (F) Father-Child Conflict .11* .35
(M) Mother-Child Conflict with (Y) Father-Child Conflict .12* .23
(M) Mother-Child Conflict with (F) Father-Child Conflict .13* .67*

(Y) Mother-Child Conflict with (M) Parent-Parent Conflict .04 .39*
(M) Mother-Child Conflict with (Y) Parent-Parent Conflict .02 .50*

(Y) Father-Child Conflict with (F) Parent-Parent Conflict .07 .13
(D) Father-Child Conflict with (Y) Parent-Parent Conflict .12* .34

Correlations across Averaged Reports

Mother-Child Conflict with Father-Child Conflict .34* .68*
Mother-Child Conflict with Parent-Parent Conflict .16* .62*
Father-Child Conflict with Parent-Parent Conflict .26* .62*

Note. Reporters are included in parentheses (Y = youth report; M = mother report; F = father
report). * = significant at p < .05.

FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 45
Table 9
Daily Predictors of Family Conflict, Negative Mood, and Problems in School
Construct 𝛽 SE 𝛽/SE p
Report Number

(C) Mother-Child Conflict -.01 .04 -.37 .71
(C) Father-Child Conflict -.04 .04 -.99 .32
(C) Parent-Parent Conflict -.07 .03 -2.10 .04
(C) Negative Mood -.12 .04 -3.03 < .01
(C) Problems in School -.12 .05 -2.28 .02
(M) Mother-Child Conflict -09 .04 -2.46 .01
(M) Parent-Parent Conflict -.09 .04 -2.44 .02
(F) Father-Child Conflict -.01 .05 .19 .85
(F) Parent-Parent Conflict -.08 .04 -1.87 .06

Weekend

(C) Mother-Child Conflict .10 .03 2.97 < .01
(C) Father-Child Conflict .07 .03 2.11 .04
(C) Parent-Parent Conflict .05 .03 1.43 .15
(C) Negative Mood .02 .03 .49 .62
(M) Mother-Child Conflict .07 .03 2.26 .02
(M) Parent-Parent Conflict .07 .03 2.24 .03
(F) Father-Child Conflict .07 .03 2.11 .03
(F) Parent-Parent Conflict .07 .03 2.28 .02

Note. Reporters are included in parentheses (Y = youth report; M = mother report; F = father
report). Analyses were run with separate regression models. Negative values indicated decreased
endorsement and positive values indicate increased endorsement.

FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 46
Table 10
Within-Level Associations between Family Conflict, Negative Mood, and Problems in School
𝛽 SE 𝛽/SE p R
2

Child Report

Negative Mood on Mother-Child Conflict .26 .04 6.71 < .001 .09
Negative Mood on Parent-Parent Conflict .07 .03 2.23 .03 .02
Problems in School on Negative Mood .22 .04 5.10 < .001 .06
Problems in School on Mother-Child Conflict .24 .06 3.87 < .001 .07
Problems in School on Parent-Parent Conflict -.02 .06 -.33 .74 .02

Mother Report

Negative Mood on Mother-Child Conflict .12 .03 4.18 < .001 .03
Negative Mood on Parent-Parent Conflict .05 .04 1.40 .16 .02
Problems in School on Mother-Child Conflict -.02 .05 -.37 .71 .02
Problems in School on Parent-Parent Conflict -.08 .07 -1.38 .17 .02

Father Report

Negative Mood on Father-Child Conflict .13 .05 2.77 .01 .03
Negative Mood on Parent-Parent Conflict .04 .04 .05 .34 .02
Problems in School on Father-Child Conflict .18 .07 2.40 .02 .04
Problems in School on Parent-Parent Conflict .06 .06 .89 .38 .01

Note. Analyses were initially run in separate regression models. Weekend and report number
were included as covariates where appropriate (i.e., weekend number was not included as a
covariate when the outcome was problems in school).

FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 47
Table 11
Between-Level Associations between Family Conflict, Negative Mood, and Problems in School
𝛽 SE 𝛽/SE p R
2

Child Report

Negative Mood on Mother-Child Conflict .54 .12 4.62 < .001 .29
Negative Mood on Parent-Parent Conflict .29 .13 1.73 .08 .05
Problems in School on Negative Mood .70 .08 8.59 < .001 .49
Problems in School on Mother-Child Conflict .69 .09 7.79 < .001 .49
Problems in School on Parent-Parent Conflict .27 .15 1.83 .07 .07

Mother Report

Negative Mood on Mother-Child Conflict .23 .09 2.49 .01 .05
Negative Mood on Parent-Parent Conflict .12 .12 1.06 .29 .02
Problems in School on Mother-Child Conflict .26 .15 1.75 .08 .07
Problems in School on Parent-Parent Conflict .17 .16 1.05 .29 .03

Father Report

Negative Mood on Father-Child Conflict .18 .13 1.38 .17 .04
Negative Mood on Parent-Parent Conflict .14 .13 1.11 .27 .02
Problems in School on Father-Child Conflict .28 .14 1.96 .05 .08
Problems in School on Parent-Parent Conflict .12 .17 .71 .48 .02

Note. Analyses were initially run in separate regression models. Weekend and report number
were included as covariates where appropriate (i.e., weekend number was not included as a
covariate when the outcome was problems in school).

FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 48
Table 12
Associations between Family Conflict, Negative Mood, and Problems in School after Averaging
across Reporters

Within Level

𝛽 SE 𝛽/SE p R
2

Negative Mood on Mother-Child Conflict .25 .03 7.40 < .001 .08
Negative Mood on Father-Child Conflict .23 .04 6.09 < .001 .07
Negative Mood on Parent-Parent Conflict .07 .04 1.76 .08 .02
Problems in School on Negative Mood .22 .04 5.10 < .001 .06
Problems in School on Mother-Child Conflict .15 .06 2.63 .01 .03
Problems in School on Father-Child Conflict .20 .09 2.87 .004 .05
Problems in School on Parent-Parent Conflict -.02 .06 -.35 .73 .02

Between Level

Negative Mood on Mother-Child Conflict .42 .105 4.06 < .001 .18
Negative Mood on Father-Child Conflict .40 .10 4.07 < .001 .16
Negative Mood on Parent-Parent Conflict .18 .117 1.51 .13 .03
Problems in School on Negative Mood .70 .08 8.49 < .001 .50
Problems in School on Mother-Child Conflict .50 .10 5.05 < .001 .25
Problems in School on Father-Child Conflict .49 .10 4.77 < .001 .24
Problems in School on Parent-Parent Conflict .20 .16 1.26 .21 .04

Note. Analyses were initially run in separate regression models. Weekend and report number
were included as covariates where appropriate (i.e., weekend number was not included as a
covariate when the outcome was problems in school).

FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 49

Figure 1. Within-level mediation model with negative mood mediating the relationship between
both mother-child and father-child conflict and problems in school. For simplicity, the between-
level estimates are not pictured. Dashed lines are non-significant. Standardized coefficients are
shown. For within-level negative mood, R
2
= .10, and for within-level problems in school, R
2
=
.11. The means (standard deviations) for the model fit indices were: χ² (1, N = 1,498) = 0.00
(0.00); RMSEA = 0.00 (0.00); CFI = 1.00 (0.00); SRMR
within
= 0.01 (0.002); SRMR
between
=
0.00 (0.00). For simplicity, the covariates weekend and report number are not shown.

Mother-
Child
Conflict

Father-
Child
Conflict

Problems in
School

Negative
Mood

.34*
.20*
.16*
.19*
.07
.13*
FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 50

Figure 2. Constrained within-level mediation model with negative mood mediating the
relationship between both mother-child and father-child conflict and problems in school. For
simplicity, the between-level estimates are not pictured. Standardized coefficients are shown. For
within-level negative mood, R
2
= .10, and for within-level problems in school, R
2
= .10. The
means (standard deviations) for the model fit indices were: χ² (4, N = 1,498) = 8.01 (5.29), p =
.09; RMSEA = 0.02 (0.02); CFI = 0.99 (0.01); SRMR
within
= 0.02 (0.004); SRMR
between
= 0.02
(0.01). For simplicity, the covariates weekend and report number are not shown.

Mother-
Child
Conflict

Father-
Child
Conflict

Problems in
School

Negative
Mood

.33*
.22*
.22*
.19*
.13*
.13*
FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 51

Figure 3. Within-level constrained model of the across-day relationship between parent-child
conflict and problems in school at time t and time t + 1. Standardized coefficients are shown. For
parent-child conflict (t + 1), R
2
= .21, and for problems in school (t + 1), R
2
= .27. The means
(standard deviations) for the model fit indices were: χ² (1, N = 1,498) = 2.09 (1.96), p = .15;
RMSEA = 0.02 (0.02); CFI = 1.00 (0.001); SRMR
within
= 0.01 (0.003). For simplicity, the
covariate report number is not shown.

Parent-Child
Conflict
(t)

Problems in
School
(t)

Parent-Child
Conflict
(t + 1)

Problems in
School
(t + 1)

.35*
.46*
.44*
.09*
.24*
.09*
FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 52

Figure 4. Within-level constrained model of the across-day relationship between parent-child
conflict and problems in school at time t and time t + 2. Standardized coefficients are shown.
For parent-child conflict (t + 2), R
2
= .15, and for problems in school (t + 2), R
2
= .21. The means
(standard deviations) for the model fit indices were: χ² (1, N = 1,498) = 1.18 (1.15), p = .28;
RMSEA = 0.01 (0.02); CFI = 0.99 (0.001); SRMR
within
= 0.01 (0.003). For simplicity, the
covariate report number is not shown.

Parent-Child
Conflict
(t)

Problems in
School
(t)

Parent-Child
Conflict
(t + 2)

Problems in
School
(t + 2)

.34*
.46*
.38*
.10*
.26*
.10*
FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 53

Figure 5. Within-level mediation model of (1) negative mood (t) mediating the relationship
between parent-child conflict (t) and next-day problems in school (t + 1) and (2) of negative
mood (t + 1) mediating the relationship between problems in school (t) and next-day parent-child
conflict (t + 1). Standardized coefficients are shown. Dashed lines represent non-significant
paths. For parent-child conflict (t + 1), R
2
= .27, for negative mood (t + 1), R
2
= .27, and for
problems in school (t + 1), R
2
= .26. The means (standard deviations) for the model fit indices
were: χ² (2, N = 1,498) = 3.61 (4.08), p = .16; RMSEA = 0.02 (0.02); CFI = 1.00 (0.001); SRMR
within
= 0.03 (0.04). For simplicity, the covariates weekend and report number are not shown.

Parent-Child
Conflict
(t)

Problems in
School
(t)

.35*

Negative
Mood
(t)

Parent-Child
Conflict
(t + 1)

Negative
Mood
(t + 1)

Problems in
School
(t + 1)

.30*
.27*
.32*
.31*
.39*
.41*
.45*
.03
.10*
.03
.08*
.30*
.13*
.06
FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 54

Figure 6. Within-level mediation model of (1) negative mood (t + 1) mediating the relationship
between parent-child conflict (t) and problems in school (t + 2) and (2) of negative mood (t + 1)
mediating the relationship between problems in school (t) and parent-child conflict (t + 2). For
simplicity, this model depicts all estimated paths but no coefficients. See Figure 7 for the
coefficients of the significant paths. Dashed lines represent non-significant paths. For parent-
child conflict (t + 1), R
2
= .17, for parent-child conflict (t + 2), R
2
= .22, for negative mood (t +
1), R
2
= .26, for negative mood (t + 2), R
2
= .29, for problems in school (t + 1), R
2
= .21, and for
problems in school (t + 2), R
2
= .28. The means (standard deviations) for the model fit indices
were: χ² (13, N = 1,498) = 46.59 (13.12), p < .001; RMSEA = 0.04 (0.01); CFI = 0.99 (0.004);
SRMR
within
= 0.03 (0.004). For simplicity, the covariates weekend and report number are not
shown.

Parent-Child
Conflict
(t)

Problems in
School
(t)

Negative
Mood
(t)

Parent-Child
Conflict
(t + 2)

Negative
Mood
(t + 2)

Problems in
School
(t + 2)

Parent-Child
Conflict
(t + 1)

Negative
Mood
(t + 1)

Problems in
School
(t + 1)

FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 55

Figure 7. Within-level mediation model of (1) negative mood (t + 1) mediating the relationship
between parent-child conflict (t) and problems in school (t + 2) and (2) of negative mood (t + 1)
mediating the relationship between problems in school (t) and parent-child conflict (t + 2). For
simplicity, this model depicts only the significant paths. Standardized coefficients are shown. For
parent-child conflict (t + 1), R
2
= .17, for parent-child conflict (t + 2), R
2
= .22, for negative
mood (t + 1), R
2
= .26, for negative mood (t + 2), R
2
= .29, for problems in school (t + 1), R
2
=
.21, and for problems in school (t + 2), R
2
= .28. The means (standard deviations) for the model
fit indices were: χ² (13, N = 1,498) = 46.59 (13.12), p < .001; RMSEA = 0.04 (0.01); CFI = 0.99
(0.004); SRMR
within
= 0.03 (0.004). For simplicity, the covariates weekend and report number
are not shown.

Problems in
School
(t)

Parent-Child
Conflict
(t)

Negative
Mood
(t)

Parent-Child
Conflict
(t + 2)

Negative
Mood
(t + 2)

Problems in
School
(t + 2)

Parent-Child
Conflict
(t + 1)

Negative
Mood
(t + 1)

Problems in
School
(t + 1)

.33*
.33*
.42*
.22*
.39* .32*
.46*
.30*
.37*
.37* .32*
.24*
.29*
.27*
.19*
.15*
.08*
.30*
.28*
.12*
.06*
.20*
FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 56

Figure 8. A cross-level interaction in which the within-day relationship between parent-child
conflict and negative mood is moderated by gender. Females (n = 54) exhibited greater spillover
between parent-child conflict and negative mood than did males (n = 52).

0
0.2
0.4
0.6
0.8
1
1.2
1.4
1 2 3
Negative Mood
Parent-Child Conflict
Gender as a Moderator of Spillover between
Parent-Child Conflict and Negative Mood
Females
Males
FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 57
Appendix
Child Report of Negative mood
1. Felt restless or jumpy
2. Felt lonely
3. Was angry at myself
4. Was angry at someone else
5. Felt sad
6. Felt nervous
7. Felt foolish or embarrassed in front of others

Child Report of Parent-Parent Conflict
1. My mom yelled at or criticized my dad
2. My dad seemed annoyed at my mom
3. My mom seemed annoyed at my dad
4. My dad yelled at or criticized my mom

Child Report of Mother-Child Conflict
1. My mom seemed irritated with me today
2. My mom said something mean to me today
3. My mom and I argued
4. My mom made me feel stupid

Child Report of Father-Child Conflict
1. My dad seemed irritated with me today
2. My dad said something mean to me today
3. My dad and I argued
4. My dad made me feel stupid

Child Report of Problems in School
1. I didn’t finish my homework
2. I felt bored at school
3. I don’t understand or can’t do some of my schoolwork
4. I got a bad grade or did poorly on homework, a quiz or test
5. I had to do something that I didn’t like
6. I was late for school or late for a class school
7. I cut class/classes

Parent Report of Parent-Parent Conflict
1. My partner/spouse was angry with me
2. I was angry with my partner/spouse
3. My partner/spouse was annoyed with me
4. I was annoyed with my partner/spouse
5. My partner/spouse yelled at or criticized me
6. I yelled at or criticized my partner/spouse
7. My partner/spouse flew off the handle or exploded at me.
FAMILY CONFLICT, MOOD, AND PROBLEMS IN SCHOOL 58
8. I flew off the handle or exploded at my spouse

Parent Report of Parent-Child Conflict
1. I said something mean to my child
2. My child said something mean to me
3. I argued with my child
4. My child argued with me
5. I was irritated with me child
6. My child was irritated with me
7. I made my child feel stupid
8. My child made me feel stupid

Abstract (if available)

Abstract Research indicates that family conflict may interfere with adolescents’ achievement in school. To understand how family conflict may disrupt academic achievement, several researchers have examined spillover patterns using daily diary data. The current study expands on extant work by examining how conflict in specific family dyads is associated with daily negative mood and problems in school. Participants consisted of 106 adolescents who provided reports of mother-child conflict, father-child conflict, parent-parent conflict, negative mood, and school problems daily for 14 days. Both mother-child and father-child conflict were associated with same-day problems in school but parent-parent conflict was not. Results from a cross-lagged panel model indicated that the effects were bidirectional such that problems in school predicted next-day parent-child conflict and that parent-child conflict predicted next-day problems in school. Results also showed that negative mood (time t) mediated the relationship between parent-child conflict (time t) and next-day problems in school (time t + 1) and that negative mood (time t + 1) mediated the relationship between problems in school (time t) and next-day parent-child conflict (time t + 1). Mediation effects across a 3-day period were not significant. These findings suggest that parent-child conflict may impact negative mood, which could affect engagement in school and interfere with adolescents’ academic achievement.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Couple conflict during pregnancy: Do early family adversity and oxytocin play a role?

PDF

Risky behaviors, interpersonal conflict, and their relation to fluctuations in adolescents’ diurnal HPA rhythms

PDF

Marital conflict and parenting: moderating effects of ethnicity and parent sex

PDF

Family aggression, prosocial friends, and the risk of dating and friend victimization in late adolescence

PDF

Parents' attunement: relation to adolescent problem behavior and the moderating role of marital conflict

PDF

Evaluating a cultural process model of depression and suicidality in Latino adolescents

PDF

Direct and indirect predictors of traumatic stress and distress in orphaned survivors of the 1994 Rwandan Tutsi genocide

PDF

Prenatal predictors of parental sensitivity in first-time mothers and fathers

PDF

Couples’ neuroendocrine activity in response to family conflict discussions: the role of self-reported anger and previous marital aggression

PDF

Emergent literacy skills and home literacy environment of Latino children who live in poverty

PDF

Family aggression exposure and community violence exposure associated with brain volume in late adolescence: a comparison of automated versus manual segmentation

PDF

Adolescent conduct problems and substance use: an examination of the risk pathway across the transition to high school

PDF

Emotion regulation as a mechanism linking parents’ marital aggression to adolescent behavioral problems: a longitudinal analysis

PDF

The relationship between dating status and academic and social functioning among Latinx and Asian-American middle adolescents

PDF

The link between maternal depression and adolescent daughters' risk behavior: the mediating and moderating role of family

PDF

Examining the associations of respiratory problems with psychological functioning and the moderating role of engagement in pleasurable activities during late adolescence

PDF

Social support, constraints, and protective buffering in prostate cancer patients and their partners

PDF

Sources of stability and change in the trajectory of openness to experience across the lifespan

PDF

An examination of in-school and online protective factors for adolescent trajectories of online victimization and harassment over time

PDF

Antecedents of marijuana initiation

Asset Metadata

Creator Timmons, Adela C. (author)

Core Title Family conflict, negative mood, and adolescents' daily problems in school

School College of Letters, Arts and Sciences

Degree Master of Arts

Degree Program Psychology

Publication Date 06/28/2013

Defense Date 05/15/2013

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

Tag daily diary data,family conflict,negative mood,OAI-PMH Harvest,problems in school

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Margolin, Gayla (committee chair), Gatz, Margaret (committee member), McArdle, John J. (committee member)

Creator Email adelatim@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-282175

Unique identifier UC11293605

Identifier etd-TimmonsAde-1721.pdf (filename),usctheses-c3-282175 (legacy record id)

Legacy Identifier etd-TimmonsAde-1721.pdf

Dmrecord 282175

Document Type Thesis

Format application/pdf (imt)

Rights Timmons, Adela C.

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