Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
A model of student performance in principles of macroeconomics
(USC Thesis Other)
A model of student performance in principles of macroeconomics
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
INFORMATION TO USERS
This manuscript has been reproduced from the microfilm master. UMI
films the text directly from the original or copy submitted. Thus, some
thesis and dissertation copies are in typewriter face, while others may be
from any type o f computer printer.
The quality of this reproduction is dependent upon the quality of the
copy submitted. Broken or indistinct print, colored or poor quality
illustrations and photographs, print bleedthrough, substandard margins,
and improper alignment can adversely affect reproduction.
In the unlikely event that the author did not send UMI a complete
manuscript and there are missing pages, these will be noted. Also, if
unauthorized copyright material had to be removed, a note will indicate
the deletion.
Oversize materials (e.g., maps, drawings, charts) are reproduced by
sectioning the original, beginning at the upper left-hand comer and
continuing from left to right in equal sections with small overlaps. Each
original is also photographed in one exposure and is included in reduced
form at the back of the book.
Photographs included in the original manuscript have been reproduced
xerographically in this copy. Higher quality 6” x 9” black and white
photographic prints are available for any photographs or illustrations
appearing in this copy for an additional charge. Contact UMI directly to
order.
UMI
A Bell & Howell Information Company
300 North Zed) Road, Ann Arbor MI 48106-1346 USA
313/761-4700 800/521-0600
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Copyright 1996
A MODEL OF STUDENT PERFORMANCE
IN PRINCIPLES OF MACROECONOMICS
by
Jill Sundie
A Thesis Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
MASTER OF ARTS
(Economics)
August 1996
Jill M. Sundie
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UMI Number: 1381609
Copyright 1996 by
Sundie, Jill Marie
All rights reserved.
UMI Microform 1381609
Copyright 1996, by UMI Company. All rights reserved.
This microform edition is protected against unauthorized
copying under Title 17, United States Code.
UMI
300 North Zeeb Road
Ann Arbor, MI 48103
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
UNIVERSITY O F SOUTHERN CALIFORNIA
T H E GRADUATE SCHOOL
UNIVERSITY PARK
L O S ANOELES. CALIFORNIA S0007
This thesis, •written by
under the direction of h^~..JThesis Committee,
and approved by all its members, has been pre
sented to and accepted by the Dean of The
Graduate School, in partial fulfillment of the
requirements fo r the degree of
Jill Sundie
Dal* 1 - & - P
D im
THESIS COMMITTEE / y
Chairman
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
TABLE OF CONTENTS
Chapter Page
1 INTRODUCTION..................................................................................1
Statement of Purpose.............................................................................1
Objective of Study.................................................................................1
Method of Approach............................................................................. 2
Review of Literature............................................................................. 3
2 THE PRODUCTION FUNCTION...................................................... 10
The Attendance Data...........................................................................13
The Attendance Record.......................................................................14
The Relationship Between Attendance and Performance.................. 15
Regression............................................................................................25
Regression Analysis............................................................................ 42
3 CONCLUSION.................................................................................... 49
BIBLIOGRAPHY.............................................................................................53
APPENDIX......................................................................................................55
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER 1: INTRODUCTION
Statement of Purpose
For several decades, researchers have attempted to uncover the determinants of
success in college level principles of economics courses. The search for a model of
student performance in “economics 101” continues, with a recent focus on an important
factor for academic success - student attendance. Does attendance matter for
performance even when variables representing the student’s academic abilities are
controlled for? This study attempts to address the issue of student performance in
principles of macroeconomics at Arizona State University.
Objective of the Study
When considering the determinants of performance in principles of economics
courses, it is useful to classify the variables expected to determine performance into three
categories: ability, experience, and effort. Student ability to succeed in university level
principles of economics courses may be measured by ACT and SAT scores, high school
grade point average or class rank, grade point average in other college courses, college
choice, and possibly gender. Student experience can be subdivided into experience with
college courses in general, and experience with "real life" situations which may improve
their economic understanding.
Experience with college courses may make the learning environment and testing
methods in the principles courses more familiar to students. They may also be more
aware of what is required to succeed in the college arena. Real-world experiences
1
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
accumulate with age. Older students may have gathered life experiences which will help
them relate to and understand economic concepts and theories. The effort extended by
students in these principles courses may be measured in part by their attendance
throughout the course. The objective of this study is to investigate which of the effort,
ability, and experience variables identified above impact student performance in
principles of macroeconomics courses.
Method of Approach
Students who attend class more often are likely to perform better in the course.
However, attendance may simply reflect things like ability or experience which may be
the true determinants of performance. First, a series of calculations will be performed to
examine whether attendance appears to have a positive relationship with performance.
Once this simple relationship between attendance and performance is established, the
model of student performance can be constructed. By building a model for student
performance in the macroeconomics courses, and controlling for student ability, effort
and experience, the effect of attendance on performance can be more accurately
examined. Previous research on the determinants of student performance in economics
courses, and college courses in general, directs us as to how we would expect each of
these variables to impact student performance. The following is a review of research
related to each of the variables chosen to explain student performance.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Review of Literature
Does the amount of experience students have had with college courses influence
their performance in principles o f macroeconomics? Bonello et al (1984) found that
sophomores outperformed freshman in an introductory economics course, even when
they controlled for SAT scores, college and high school grade point averages, and
previous economic understanding. Students were split into quartiles by their total SAT
scores, and the variable for class membership (freshman or sophomore) was significant at
the 5% level for the top three SAT quartiles. Williams et al (1992) found that freshman
performed worse overall than upperclassmen. These studies suggest that we should
expect students with a greater number of cumulative college credit hours to perform
better in a principles of economics course than those students with fewer cumulative
credit hours.
Does the student's age, which reflects a bigger stock of real-world knowledge,
contribute to success in economics courses? In a study of the performance of returning
students enrolled in an introductory economics and business statistics course, Leppel
(1984) found that these older returning students outperformed the younger continuing
students. Leppel explained this difference in performance in terms of differential ability
and time invested in studying for the courses. The greater ability of the older students is
attributed in part to their greater stock of real-life examples, which they can apply to the
content presented to them in the course to enhance their understanding of the material.
In another study testing student understanding of economics, after a course in principles
3
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
of economics, Manahan (1983) found student age to be significant (at the 5% level).
These studies would lead us to expect older students to perform better than younger
students in principles of economics courses.
Student ability may be reflected in any number of variables available for this
study, including high school grade point average and class rank, college placement exam
scores, college choice, and college grade point average. These variables may all serve as
indicators of a students ability to achieve in future college coursework.
For economics courses, and principles courses in particular, we may find that
some of these ability variables have a stronger impact on performance than others.
Performance in high school as represented by high school grade point average and/or
high school class rank may provide a good indicator of a student's ability to achieve in
college coursework. A number of studies of performance in principles of economics
courses have found that one or both of these variables contribute to student achievement
(Anderson, et al 1994, Bonello et al 1984, Borg et al 1989, Schmidt 1983). For freshman
principles students, the high school record may represent their ability more accurately
than their very limited university record. We would expect that high school grade point
average would contribute positively to performance in college courses, and that class
rank would affect college performance negatively (a high percentage class rank reflects
lower performance relative to the other high school students).
We would expect that better performance on college entrance exams would
indicate an ability to perform well in college courses. Leppel (1984) found that the
4
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
composite SAT score positively affected performance in an introductory economics and
business statistics course. Borg et al (1989) and Schmidt (1983) found that a composite
representation of SAT and ACT scores explained performance in economics courses.
Park and Kerr (1990) found that ACT scores explained course grades in a money and
banking course. Manahan (1983) found that total ACT scores affected performance on a
standardized test of economic understanding that was given after the completion of a
principles of macroeconomics course.
It is possible that the verbal and quantitative portions of the SAT and the ACT
affect performance in economics courses differently. Bonello et al (1984) examined the
effect of the verbal and quantitative portions of the SAT separately, and found that each
section of the exam influenced performance in principles of macroeconomics and
microeconomics positively and significantly. Williams et al (1992) found that the math
portion of the SAT strongly determined performance on multiple choice and numerical
and spatial questions on course exams. The verbal portion of the SAT had a negative
impact on performance on the spatial and numerical exam questions, and a positive
impact on performance on the multiple choice questions. Durden and Ellis (1995) found
the quantitative and verbal portions of the SAT both had a large impact on course
average in principles of macroeconomics and microeconomics. Ferber et al (1983) found
that verbal and quantitative SCAT scores positively influenced performance on both
essay and multiple choice final exam questions. Given the previous research in which
standardized test scores were included as independent variables in explaining
5
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
performance, we would expect both verbal and quantitative portions of the SAT and
ACT to positively affect performance in the principles of economics courses at Arizona
State.
College grade point averages reflect the students’ ability to succeed in other
courses. Romer (1993) and Bonello et al (1984) found that college grade point average
had a positive effect on performance on course exams. Borg et al (1989) and Park and
Kerr (1990) found that college grade point average positively affected course grades in
principles of economics. College grade point average had a large impact on course
average in another study by Durden and Ellis (1995). Ferber et al (1983) found that grade
point average positively affected final exam performance in principles of economics.
Given these studies, we would expect that the college grade point averages of students
enrolled in principles of macroeconomics at Arizona State would positively affect their
performance in that course.
The effect of student gender on performance in economics courses has been
examined extensively in studies of student performance. In a study of first year students
enrolled in an economics course in Singapore, Tay (1994) found that men performed
better than women on essay exams. Anderson et al (1994) found that males
outperformed females by 3.3% in introductory economics courses. In a study of
returning students enrolled in economics courses, Leppel (1984) found that returning
student women received higher grades than returning men. Manahan (1983) found that
males performed better on average than women on a post-course test of economic
6
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
understanding, but not significantly so. In a large study of students in the United
Kingdom, Lumsden and Scott (1987) found that males did better than females on
multiple choice exam questions, while females performed better on essay exam
questions. Females were also found to have lower learning rates in their economics
courses than their male counterparts. MacDowell et al (1977) reported that high school
learning of economics is not gender biased, and provided further evidence that gender
differences in economic performance tend to appear first at the university level. Ferber
et al (1983) found that men outperformed women on both essay and multiple choice final
exam questions in principles of economics. When SCAT scores were controlled for,
gender was only significant in explaining performance on the multiple choice final exam
questions, with males performing better. Males still outperformed females on the essay
questions when SCAT scores were included as explanatory variables, but not
significantly so.
Siegfried (1979) reviewed much of the relevant literature on gender differences in
economics. He found insufficient evidence for a sex difference in learning economics
(the increase in economic understanding gained over a economics course), but concluded
that the evidence for sex differences in the absolute level of understanding of economics
was strong. From this extensive and somewhat conflicting evidence for gender
differences in economic performance, the prediction for gender in explaining
performance is somewhat unclear. Given that the present study only examines measures
of the understanding of economics, and that the exams in the principles course at Arizona
7
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
State are multiple choice, it is expected that females will perform slightly worse than
males on exams.
Different from the ability and experience students bring to their economics
courses, effort is the active participation of the student in the learning process. One way
to measure this effort is to record student attendance throughout the semester.
Several studies have been conducted to explore the effects of attendance on
performance in principles of economics courses. Romer (1993) attempts to address two
quantitative questions. First, he addresses the extent of absenteeism in various
economics courses, and finds absence rates ranging from 25-40% for a given day in
economics courses at three universities. The universities chosen are described as highly
competitive, and have 2,500,6,000 and 20,000 undergraduates respectively. Second, he
attempts to discover the effect of attendance on performance in an intermediate level
economics course. Attendance is measured six times in his intermediate
macroeconomics course in the Fall of 1990. In a series of regressions, Romer controls
for prior grade point average, and/or effort by using the proxy of number of homework
sets completed. He finds the attendance coefficient is positive and significant in all
regressions. Romer finds a significant relationship between attendance and performance
as measured by the total score on three course exams.
In a similar study Schmidt (1983) attempts to explain student performance by
attendance among other relevant variables. Time spent in class and discussion sections
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
are used as explanatory variables. The coefficients on hours spent in lectures and
discussions are both positive and significant.
A third study on the effects of absences on performance was conducted by Park
and Kerr (1990). Although the coefficients on absences were negative, they were only
significant in some of the regressions.
A fourth study by Durden and Ellis (1995) provides additional evidence that
attendance is important for achievement in principles of economics courses. They also
find that the effect of increased attendance on performance is non-linear. Missing less
than four classes does not hurt performance, but excessive absenteeism can be very
detrimental. Given the previous research on attendance in economics courses, we would
expect attendance to positively impact performance in principles of economics.
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER TWO: THE PRODUCTION FUNCTION
This model of student performance in principles of macroeconomics takes into
account the student's ability, effort, and experience as determinants of performance on
exams and in the course as a whole. Following the suggestions of previous research into
performance in economics courses, and the data available for the students at Arizona
State, the following model for performance was generated.
Performance =_/(capital, labor)
capital = c(ability, experience)
labor = /(effort)
ability « SATv, SATq, ACTe, ACTm, AdjCGPA, HSGPA, HSRank, gender, EngAr. Other
experience « adj hours, age
effort ~ attendance
Variable Name Meaning
SATv verbal portion of the SAT
SATq quantitative portion of the SAT
ACTe English portion of the ACT
ACTm math portion of the ACT
AdjCGPA college grade point average, adjusted to exclude the principles
of macroeconomics grade
HSGPA high school grade point average
HSRank percentage of high school students in their class above their
high school grade point average
Gender gender of the student
Adjhours number of college credit hours adjusted to exclude the 3
credits earned in principles of macroeconomics
Age age of the student at the end of the semester in which they
were enrolled in the principles of macroeconomics course.
Attendance the number of attendance quizzes the student was present for
EngAr students who are in the colleges of architecture and
engineering
Other students who are in colleges other than business, engineering,
or architecture
10
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
In this model of student performance, capital reflects all of the skills, knowledge
and experiences the students bring to the macroeconomics course. It is this stock of
human capital which is available for the students to draw upon in learning the material
presented in the course. This stock of capital will influence many decisions the students
make throughout the semester. It will influence their judgments about what is required
on their part to fulfill their goals for the course, including when, what, and how much to
study for exams. It will influence how easily the material presented in the course is
absorbed and retained, and how much of what they have learned is accurately
demonstrated on the course exams. This model makes use of several variables which are
believed to represent the human capital the students have acquired. This capital is
characterized by both the ability and the experience of the students in these principles
courses.
Ability can be represented by some or all of the variables listed above. High
school grade point average and high school class rank represent the ability the students
have demonstrated in previous coursework. Whether these high school statistics are
important or not in determining performance in principles of macroeconomics depends
on what skills and knowledge the high school variables reflect and how much of them
can be applied to perform successfully in this economics course. College grade point
average reflects the students’ ability to succeed in other types of college coursework.
The college grade point average is adjusted in the model so that the economics course
grade is not included in the grade point average. Standardized test scores are included to
11
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
capture the effect of the general intelligence or ability of the students. Including these
variables will help to explain performance above that which can be explained by their
achievement in coursework. It is possible that performance on the verbal and
quantitative portions of those tests influences performance in the course differently. For
this reason, these sections of the ACT and SAT are considered as different types of
capital.
Gender has explained variation in performance in a number of studies of student
performance in economics courses. This sex difference in performance may be
explained by some unaccounted for difference in human capital that exists on average
between men and women in these courses. This sex difference in human capital may be
due to (but not limited to) differential spatial abilities, cognitive abilities, problem
solving strategies, testing methods, preferences, or attitudes. These additional
components of human capital have been proposed to account for the sex differences in
performance on the quantitative portions of the ACT and SAT and in economics courses.
The students’ choice of college reflects both their interests and their perceived
ability to succeed in the coursework offered by the college. Students who enroll in the
colleges of business or architecture are likely to perform well in quantitative courses.
This quantitative background may help them excel in macroeconomics. Students who
choose to enroll in non-quantitative and non-business majors may have difficulty
performing well in the principles course. College choice may reflect the quantitative or
non-quantitative nature of the students’ stock of human capital.
12
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The experience of the students in the college environment and the real world can
contribute to their performance in principles of macroeconomics. The number of college
credit hours the students have obtained reflects their level of experience with college
coursework, and the university system as a whole. This stock of experiences makes up a
portion of the human capital they bring to the course. Real world experiences
accumulate with age. Many of the concepts in the principles of economics course relate
to current economic and political issues. The life experiences obtained by the students
become part of the human capital they bring to the economics course as well.
The effort put forth by the students in this course is largely unobserved. The one
variable that has been recorded, and which represents student effort, is student
attendance. It is important to include attendance as an explanatory variable, because it
has been shown to capture an effect of effort on performance not already captured by the
ability variables. Attendance partially represents the flow of labor put forth by the
students, and reflects their effort throughout the course.
The Attendance Data
The attendance data from Arizona State were collected for three semesters of
principles of macroeconomics; the fall semester 1993 and the spring semesters of 1994
and 1995. Attendance is recorded in the form of unannounced attendance quizzes which
are given out nine times (Spring 1994) or ten times (Fall 1993 and Spring 1995)
throughout the semester. For the 1993 and 1995 courses, there were three quizzes given
before the first and third exams, and four before the second exam. For the 1994 course,
13
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
three quizzes were given before each exam. In each of the semesters, the quizzes are
given on random days before the exam. The quizzes consist of one multiple choice
question, and students receive one point towards their final grade if they turn in the quiz,
whether or not their answer is correct. Records were kept for each quiz day, to measure
whether each student attended class that quiz day. These quiz records will serve as the
measure for student attendance throughout the three principles courses.
The Attendance Record
The attendance record for these three courses varies greatly, depending at what
time in the semester the attendance is taken. The following table displays the percentage
of students that attended each quiz throughout the semester.
QUIZ 1 2 3 4 5 6 7 8 9 10
1993 94% 87% 81% 76% 78% 76% 81% 79% 72% 68%
1994 87% 84% 77% 76% 80% 81%
*
69% 78% 67%
1995 85% 84% 78% 79% 73% 77% 69% 81% 73% 68%
* this quiz was not given in 1994
In Romer’s (1993) study of attendance in economics courses at three major
universities, attendance is taken only once, and is recorded a few weeks before the end of
the semester. In reference to the Arizona State attendance data, the timing of Romer’s
study would correspond to the ninth or tenth attendance quiz. The attendance rate ranges
from 67-78% (22-33% absenteeism) for the time period closest to that of Romer’s
survey. These rates are comparable to the 25-40% absentee rates found by Romer.
14
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The Relationship Between Attendance And Performance
The relationship between attendance before an exam and the exam score received
will be examined. A student’s attendance before an exam was counted as long as they
received a grade for that exam. For example, if a student took the first exam, but not the
second or third, that student’s attendance record was only counted before the first exam.
For this first set of figures, the students were grouped by the number of quizzes attended
before an exam, and the average exam score was calculated for each group.
15
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
In each of the three
sections of the principles of
macroeconomics course at
Arizona State, there were three
attendance quizzes given
before the first exam. The
following graphs display the
relationship between the
number of quizzes attended
before the first exam and the
first exam score. The average
first exam score either
increased or remained constant
as attendance in the course
improved. The maximum
available points on exams were
100.
Average
First Exam
Score
(1993)
3 0 2 1
Number of Quizzes Attended
Before First Exam
75
Average
First Exam
Score
(1994)
40
3 2 0 1
Number of Quizzes Attended
Before First Exam
Average
First Exam
Score
(1995)
45
40
3 2 0 1
Number of Quizzes Attended
Before First Exam
16
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Before the second
80
course exam, the 1993 and 75
Average
1995 students were given four second 65
Exam Score60
, (1993) 55
attendance quizzes, and the ^
45
1994 students three quizzes. 40
The performance results are
similar to those for the first
exam. With the exception of
those who attended once
before the second exam in
1994, and twice before the
third exam in 1995, increases
in attendance were
accompanied by higher
average scores on the second
exam. There appears to be a
positive relationship between
the attendance before the
second exam and the average
second exam score for each
group.
1 2 3
Number of Quizzes Attended
Before Second Exam
Average
Second
Exam Score 60
(1994) 55
2 3 1 0
Number of Quizzes Attended
Before Second Exam
Average
Second 65
Exam Score60
(1995) 55
2 4 0 1 3
Number of Quizzes Attended
Before Second Exam
17
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Increases in attendance
before the third exam, in all
three courses, appear to be
associated with increased
average performances on the
third exam.
Average
Third Exam65
Score 60
(1993) 55
1 2
Number of Quizzes Attended
Before Third Exam
Average
Third Exam
Score
(1994)
80
75
70
60
55
50
45
40
3 2 0 1
Number of Quizzes Attended
Before Third Exam
Average
rhird Exan
Score
(1995)
80
75
70
65
60
55
50
45
40
2 3 0 1
Number of Quizzes Attended
Before Third Exam
18
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
The relationship
between attendance and exam A verage
Quizzes
Attended
performance can be illustrated (3 possible)
in another way. Students were
sorted by the letter grades they
earned on each exam, and the
average number of quizzes
attended was calculated for
each letter grade group on each
Average
Quizzes
of the exams. As expected, in Attended
(3 possible)
almost all cases, decreases in
the letter grade earned on an
exam were accompanied by
decreases in the average
number of quizzes attended
before that exam. As before,
every student’s attendance Quines
Attended
before an exam was counted (3 possible)
as long as they took that exam.
3
2.8
2.8
2.6
2 .4
2.2
1.8
F B D C A
Grade on First Exam (1993)
2.6
2.4
2.2
1.8
F B C D A
Grade on First Exam (1994)
2.8
2.6
2.4
2.2
1.8
F D B C A
Grade on First Exam (1995)
19
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Average
Quizzes
Attended
(4 possible)
Average
Quizzes
Attended
(3 possible)
Average
Quizzes
Attended
(4 possible)
3.8
3.6
3.4
2.8
2.6
2.4
2.2
B D A C F
Grade on Second Exam (1993)
2.8
2.6
2.4
2.2
1.8
B C D F A
Grade on Second Exam (1994)
3.8
3.6
3.4
3.2
2.8
2.6
2.4
2.2
B A C D F
Grade on Second Exam (1995)
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
2.75
Average
Quizzes
Attended
(3 possible)
Average
Quizzes
Attended
(3 possible)
Average
Quizzes
Attended
(3 possible)
2.5
2.25
2
1.75
1.5
1.25
A B C 0 F
Grade on Third Exam (1993)
2.75
2.5
2.25
2
1.75
1.5
1.25
0 F B A C
Grade on Third Exam (1994)
2.75
2.5
2.25
2
1.75
1.5
1.25
0 B F A C
Grade on Third Exam (1995)
21
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
For the aggregate
attendance measures
(measured over the entire
course) only those students
who took the final exam were
counted. First, the
relationship between course
attendance and final exam
performance was examined.
Students were grouped by the
number of quizzes they
attended during the semester.
The average final exam score
was calculated for each group.
There is a rough upward trend
in final exam score
performance as the number of
total attendance points
increases. The maximum
points possible on the final
exam were 150.
Average 1gg
Final Exam
Score
1993
2 3 4 5 6 7 8 9 10
Number of Quizzes Attended
Throughout the Course
Average 100
Final Exam go
2 3 4 5 6 7 8 9
Number of Quizzes Attended
Throughout the Course
Average 10q
Final Exam
0 1 2 3 4 5 6 7 8 9 10
Number of Quizzes Attended
Throughout the Course
* Table A in the appendix shows the number o f
students in each total attendance group, for each course.
22
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Average 250
Adjusted
Total Points225
The relationship
between attendance and
adjusted total points in the
course is also examined. The
total points are adjusted so that
only exam scores are
considered, so as not to
confound the relationship
between total points and
attendance by including the
attendance points earned
throughout the semester. The
same upward trend found for
the final exam is observed for
total points in the course, a
300
275
proxy for final course grade. Average
Adjusted ^
The maximum available points Tot^gg| intS200
175
in the course were 350. 150
125
I ■ 1
0 1 2 3 4 5 6 7 8 9 10
Number of Quizzes Attended
Throughout the Course
Total Points
1994
300
275
250
225
200
175
150
125
0 1 2 3 4 5 6 7 8 9
Number of Quizzes Attended
Throughout the Semester
2 3 4 5 6 7 8 9
Number of Quizzes Attended
Throughout the Course
10
23
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
To detect whether a
discouraged or encouraged
student effect on attendance rcramffan
Attendance
after the first exam exists, the
1993
following calculations were
performed. Students were
sorted by the grade they
received on the first exam, and
their average percent
Average
attendance was calculated Percentage
Attendence
Before Exams
before both the first and one and two
1994
second exams. Average
attendance declined between
the first and second exams in
each of the three sections. It is
possible that the first exam
Average
score received by the students
Before Exams
affects their attendance one ami Two
1995
decisions before the second
exam, as they reassess the
value of attendance given their
E x a m
•" E x a m
A B C D F
Grade on First Exam
E x a m
j “ E x a m
A B C D F
Grade on First Exam
E x a m
E x a m
A B C D F
Grade on First Exam
24
100
95
90
85
80
75
70
65
60
55
50
100
95
90
85
80
75
70
65
60
55
50
100
95
90
85
80
75
70
65
60
55
50
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
first exam score. Students may feel discouraged in response to a low first exam score,
and choose not to attend class regularly before the second exam. These students may
decrease their attendance because they feel previous attendance did not contribute to
their success on the first exam, or that their time would be better spent on other courses
or activities. Students who feel confident in response to a high first exam score may
decide to reallocate their time to other endeavors. The second exam is considered the
most difficult of all three exams for the students, as it traditionally requires the largest
curve. The fall in average attendance and between the first and second exams, and
performance on the second exam might be explained by the discouraged or the
overconfident student effects.
Regression
The variables chosen to explain performance were not available for every student
in the course. Therefore, different sets of regressions were run for a given dependent
variable of interest. The same set of regressions were run for each dependent variable
chosen, and the sample size varied as different explanatory variables were included.
Age, gender, and college choice were available for virtually every student in the courses.
Other variables such as SAT scores or high school grade point average were only
available for a portion of the students.
High school grade point average and high school rank were found to be highly
correlated in all samples. The R squared for the regression of high school rank on high
school grade point average was higher than any R squared from a regression where both
25
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
variables were included to explain performance. Therefore, these two variables were
included separately, in different samples.
When examining performance on the second, third and final exams, previous
exam scores were included as explanatory variables. Students were able to drop one
exam score in calculating their final grade in the course, resulting in some students
systematically missing one of the three course exams. In the regression of final exam on
the groups of independent variables, the two highest exam scores were used as
explanatory variables to more accurately represent student performance on course exams.
This also increased the sample size for these regressions. In the following charts
displaying the regression results, the coefficients are given for each explanatory variable
used. The significance levels are indicated by the number of stars below the coefficients.
26
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
DEPENDENT VARIABLE = EXAM 1 1993
EQN
N=>
A
433
B
402
c
280
D
214
F
200
G
191
H
186
I
136
J
159
K
163
L
155
M
109
Exltot 8.43
*
4.72
*
4.60
*
6.03
*
8.5
*
6.21
*
5.1
*
4.73
♦♦
5.68
*
7.33
*
5.30
*
8.05
*
Age -.27 .25 1.19 1.30 1.00 .5 .6 -1.8 -.44 .1 .43 .7
Sex -1.24 -4.69
*
-6.09
*
-3.92 2.45 -2.6 -3.0 -2.22 -3.62 -.5 -4.59
***
-2.19
EngA r 3.87 .01 2.71 5.65
***
3.88 3.8 3.1 4.25 .47 .6 -.04 2.0
Other -4.05
**
-2.26 -.71 -.11 -1.77 -.6 -.8 -.18 .19 -1.00 .79 .6
CGPA 12.1
*
9.00
*
8.7
*
7.7
*
7.72
*
7.68
*
6.78
*
5.5
**
H ours .08
***
.0 .08 .10 .21
**
.04 -.00 .04
SATv -.01
*
-.01
*
-.01
**
-.00
**
SATq .07 .04 .03 .04
ACTe .04 .35 -.01 -.08
ACTm .89
**
1.40
*
.69
***
1.02
***
H sgpa 8.32
*
6.48
**
6.88
**
H srnk -.30
*
-.10 -.05
(Sex: female = I, male = 0) significance levels: * = 1%, ** = 5%, *** = 10%
Attendance positively impacted exam performance, while gender negatively
affected performance. College grade point average positively impacted exam
performance. The quantitative portions of the SAT and ACT also positively impacted
exam one scores. High school grade point average was positive and significant in all
three regressions.
27
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
DEPENDENT VARIABLE = EXAM 2 1993
EQN
N=>
A
397
B
383
c
267
D
203
F
187
G
182
H
177
I
130
J
151
K
153
L
147
M
102
Exl .51
*
.37
*
.28
*
.33
*
.42
*
.30
*
.28
*
.29
*
.35
*
.43
* *
.37
*
Ex2tot 3.84
*
3.14
*
2.74
*
3.31
*
3.93
*
3.26
*
3.22
*
3.63
*
2.69
**
4.04
*
2.11
***
3.17
**
Age -.18 -.06 1.37 -1.41 .34 .03 .46 -1.10 .81 .53 1.89 -1.23
Sex .82 -1.18 -3.35 -1.77 1.93 -1.09 -1.61 -2.54 -1.24 .97 -3.14 -.46
EngAr 2.89 2.33 .10 1.91 -.55 -.57 -1.49 .28 6.29 6.93 4.86 4.32
Other -1.69 -1.24 -1.39 .96 -3.26 -3.01 -4.17 -2.24 .70 .57 .23 1.37
CGPA 6.81
*
6.59
*
6.85
*
6.49
*
5.23
*
6.15
*
7.64
*
7.39
*
8.92
*
Hours -.03 .00 .10 .04 .05 .11 -.05 -.08 .10
SATv .02 .02 .02 .03
SATq .03
**
.03
***
.03 .00
ACTe .58 .79
**
.44 .16
ACTm -.39 -.15 -.66 .24
Hsgpa 6.07
**
4.64 7.07
**
Hsmk -.08 -.13 .13
(Sex: female = 1, male = 0) significance levels: * = !% ,** = 5%, *** = 10%
Attendance had a positive impact on performance on exam two in all regressions.
Exam one score coefficients were all positive and significant. College grade point
average was significant in each regression in which it was included as an explanatory
variable. Coefficients for the quantitative portions of the SAT and ACT were all
positive, but only in three cases were these variables significant.
28
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
DEPENDENT VARIABLE = EXAM 3 1993
EQN
N=>
A
343
B
341
c
247
D
187
F
166
G
166
H
161
I
118
J
139
K
139
L
136
M
94
Exl .43
*
.36
*
.25
*
.35
*
.42
*
.34
*
.28
*
.37
*
.28
*
.33
*
.23
**
.32
*
Ex2 .18
*
.13
**
.10 .07 .19
*
.14 .13 .07 .07 .17 .06 .09
Ex3tot 4.59
*
4.40
*
4.29
*
3.31
*
4.05
*
3.98
*
3.72
*
3.84
*♦
6.45
*
6.59
*
5.76
*
4.30
**
Age .13 .25 -.37 .55 1.02 .98 1.27 1.39 -1.10 -1.53 -.67 -1.31
Sex -2.26 -3.37
♦ **
-6.51
*
-5.04
**
-.37 -2.63 -3.58 -4.56 -10.1
*
-7.52
*
-10.8
*
-11.3
*
EngAr 1.13 .54 2.26 1.05 3.05 2.79 1.98 4.53 2.31 2.42 1.21 2.00
Other -.29 -.24 1.42 -.76 2.60 2.99 3.16 1.44 1.70 1.24 2.24 .09
CGPA 4.59
*
4.03
**
3.43
***
5.57
*
5.17
*
4.90
**
7.08
*
6.47
*
7.12
*
Hours .02 .05 -.04 .03 .07 -.01 -.06 -.06 - 19
SATv .01 .01 .02 .03
SATq .03
***
03 .02 .01
ACTe .21 .28 .19 -.16
ACTm .02 .28 -.20 -.35
Hsgpa 7.91
*
7.21
**
8.04
***
Hsrnk -.13
***
-.08 -.09
(Sex: female = I, male = 0) significance levels: * = 1%, ** = 5%, *** = 10%
Exam one score and attendance had positive impacts in each regression, and
exam two had a positive impact in five regressions. Gender had a negative impact on
performance, even when most explanatory variables were included. Both college and
high school grade point averages were positively related to exam three performance.
29
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
DEPENDENT VARIABLE = FINAL EXAM SCORE 1993
EQN A B c D F G H I J K L M
N=> 393 390 273 207 187 187 182 134 153 153 150 103
L ow .65 .63 .47 .44 .44 .36 .30 .39 .42 .43 .43 .15
* * * * * ** ** ** ** ** **
H igh .51 .49 .57 .60 .52 .54 .56 .39 .70 .66 .70 .94
* * * * * * * ** * * * *
AttTot 1.32 1.28 .67 1.39 1.69 1.73 1.35 2.01 .93 .94 .83 1.44
* * * * * ** * ***
Age -.20 -.23 -2.03 -2.19 -1.15 -2.52 -1.47 -2.66 -1.42 -.33 -2.15 -2.83
**
Sex 1.15 .69 -.80 .90 2.58 .87 -1.03 -1.37 4.87 4.00 3.92 8.91
**
O ther -1.58 -1.68 -1.53 .40 -.15 -.70 -.25 .14 -3.59 -2.52 -3.59 .54
EngA r 3.87 3.67 4.91 4.97 5.48 4.25 3.31 3.11 8.77 10.1 6.27 12.0
*** ** ***
C G PA 1.84 -.39 -2.14 2.77 1.81 .63 -1.62 -2.98 -4.61
H ours .02 .13 .11 .17 .16 .16 .16 .18 .23
***
** ** *** ***
SA Tv -.01 -.01 -.02 -.02
SA T q .04 .04 .03 .03
** * ***
ACTe .13 .26 .38 .24
A C Tm .87 .69 .56 63
**
Hsgpa 8.16 9.95 5.71
* *
Hsrnk -.06
.
-.13 .00
(Sex: female = I, male = 0) significance levels: * = 1%, ** = 5%, *** = 10%
Attendance, high and low exam scores, and college credit hours were positively
related to performance on the final. The quantitative portion of the SAT and high school
grade point average were also positively related to final exam scores.
30
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
DEPENDENT VARIABLE = ADJUSTED TOTAL POINTS 1993
EQN A B c D F G H I J K L M
n=>
393 390 273 207 187 187 182 134 153 153 150 103
AttTot 9.38 5.59 4.39 5.31 10.8 8.14 6.76 7.22 6.38 9.13 5.07 5.41
* * * * * * * * * * * *
Age -.01 .53 3.25 -.32 3.18 1.88 3.81 -2.37 1.40 1.57 3.82 1.86
Sex -5.71 -11.9 -18.7 -14.1 5.26 -4.58 -7.37 -8.68 -8.74 -.98 -12.5 4.93
* * **
EngAr 16.0 7.90 16.0 21.6 15.4 14.3 10.5 16.2 25.6 28.5 23.1 40.3
** ** * *** *** ** ** ** *
Other -9.30 -7.54 -1.54 2.33 -1.83 .98 -.42 1.91 -3.73 -4.20 -2.62 -1.16
CGPA 35.6 25.2 22.2 25.5 21.1 20.7 24.9 21.5 17.4
* * * * * * * * **
Hours .16 .03 .19 .14 .16 .30 -.06 -.15 .03
SATv -.00 -.03 -.03 -.03
SATq .23 .17 .14 .13
* * * *
ACTe 1.00 1.87 .64 -.43
***
ACTm 2.53 3.73 1.54 3.65
** * **
Hsgpa 34.5 28.1 28.9
* * *
Hsrnk -.56 -.46 -.14
______
* **
(Sex: female = 1, male = 0) significance levels: * = 1%, ** = 5%, *** = 10%
Attendance positively impacted performance in the course in all regressions.
Females performed significantly worse than males in three samples. Engineering and
architecture students outperformed students in other colleges. College grade point
averages and quantitative SAT and ACT scores were positively related to course
performance. High school rank was negatively related to course performance.
31
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
DEPENDENT VARIABLE = EXAM 1 1994
EQN A B c D F G H I J K L M
n=> 484 452 289 228 228 215 206 155 173 186 164 118
Exltot 7.25 4.42 3.89 4.72 5.81 4.15 4.16 3.0 7.24 9.30 6.88 7.77
* * * * * * * *** * * * ♦
Age -.36 -.11 -.24 -.48 .05 -1.32 -.79 -2.62 .72 .96 1.67 1.36
* * *
Sex -1.5 -4.65 -4.8 -6.79 1.10 -2.03 -2.04 -2.50 -5.61 -3.67 -5.85 -5.81
* * * *# * **
EngAr 2.37 -.32 .14 -1.97 2.35 1.04 .67 -.13 -2.9 -2.51 -3.00 -6.25
Other -3.08 -2.65 -2.2 -2.01 -1.20 .10 -.41 1.13 -4.30 -5.59 -4.27 -3.86
♦ ♦♦ *** * * * ** ***
CGPA 10.7 8.43 8.11 6.39 5.03 5.9 6.92 6.41 6.65
* * * * * * * * *
Hours .02 .05 .04 .14 .13 .15 .04 .03 .02
* ** **
SATv .02 .02 .01 .01
***
SATq .07 .05 .05 .04
* * * *
ACTe .01 .25 -.22 -.18
ACTm 1.29 1.79 1.34 1.33
* * * *
Hsgpa 5.98 2.80 5.15
* ***
Hsmk -.11 -.06 -.10
**
(Sex: female = 1, male = 0) significance levels: * = 1%, ** = 5%, *** = 10%
Attendance had a positive and significant impact on exam one performance.
When gender was significant, men outperformed women by four to seven points.
Students who were enrolled in colleges other than business, engineering, or architecture
performed worse on exam one. College grade point averages and math SAT and ACT
scores were positively related to performance.
32
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
DEPENDENT VARIABLE = EXAM 2 1994
EQN
n=>
A
440
B
427
c
277
D
216
F
215
G
208
H
199
I
149
J
164
K
170
L
155
iM
111
Exl .50
*
.25
*
.09 .10 .33
*
.14 .12 .01 .13 .28
*
.12 .14
Ex2tot 4.53
*
2.13
*♦
.40 .80 3.57
**
.96 .53 1.18 1.27 4.36
**
.40 .92
Age -.04 .14 .25 .19 .60 .53 2.05 1.17 1.3 1.10 .76 .76
Sex 3.43
***
.70 -.52 -2.25 4.91
***
.95 1.48 1.32 1.25 3.86 1.06 .93
EngAr -.64 -2.49 -.77 -2.73 -1.49 -2.66 -3.04 -3.69 1.84 1.37 2.95 .48
Other -2.27 -3.94
***
-2.95 -.31 -2.74 -4.03 -3.58 .01 -2.91 -2.02 -3.30 -3.02
CGPA 11.9
*
12.8
*
13.4
*
14.7
*
12.2
*
12.5
*
11.7
*
11.0
*
10.9
*
Hours -.04 .00 .03 .03 -.01 .04 -.07 -.05 -.01
SATv .01 -.01 -.02 -.02
SATq .03
***
.01 .01 .03
ACTe .34 .63 .29 .59
ACTm .45 .74
***
.67 .39
Hsgpa 5.75
**
8.70
*
.12
Hsrnk -.07 -.08 .01
(Sex: Female = 1, male = 0) significance levels: * = 1%, ** = 5%, *** = 10%
Attendance before the exam, and the previous exam score, were only significant
four times. College grade point average was positively related to performance on the
second exam in this course. High school grade point average also had a positive impact
on performance on exam two.
33
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
DEPENDENT VARIABLE = EXAM 3 1994
EQN
n=>
A
388
B
386
c
255
D
195
F
192
G
191
H
183
I
136
J
151
K
151
L
142
M
101
Ext .19
*
.07 .09 .14 .17
**
.12 .14
* * *
.08 -.03 .06 -.00 .12
Ex2 .26
*
.17
*
.17
*
.07 .22
*
.13
**
.13
* * *
.05 .14
***
.24
*
.10 .00
Ex3tot 3.6
*
2.84
*
2.70
**
1.29 3.76
*
2.49
***
2.06 .87 3.45
**
5.06
*
3.29
**
2.13
Age -.16 -.11 -.24 -.69 -.61 .30 .95 3.14 .87 .60 .62 1.17
Sex .32 -1.17 -2.52 -.88 -.31 -.88 .13 1.28 -2.90 -1.80 -2.93 -2.24
EngAr .00 -1.01 .52 .85 1.19 1.66 2.48 2.91 -3.16 -3.02 -1.52 .60
Other -2.93 -3.74
**
-1.46 -1.15 1.34 .98 1.29 2.28 -2.57 -1.45 -3.11 .28
CGPA 7.94
*
8.93
*
9.04
♦
8.95
*
9.78
*
10.5
*
10.5
*
10.7
*
10.7
*
Hours -.03 -.04 -.09 -.12
***
-.15
**
-.25
**
-.06 -.06 -.11
SATv -.02 -.03
***
-.02 -.03
SATq .02 .01 .02 .02
ACTe -.66 -.25 -.37 -.50
ACTm .80
***
.84
***
.94
***
.60
Hsgpa -2.80 -3.60 -4.92
Hsmk .04 .03 .02
(Sex: female = 1, male = 0) significance levels: * = 1%, ** = 5%, *** = 10%
The most important explanatory variables for these samples were exam two score,
attendance before exam three, and college grade point average. The math portion of the
ACT was positively related to performance, and the number of college credits the
students had earned was negatively related to exam scores in three regressions.
34
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
DEPENDENT VARIABLE = FINAL EXAM SCORE 1994
EQN
n=>
A
442
B
439
c
278
D
219
F
211
G
210
H
201
I
ISO
J
165
K
165
L
155
iM
110
Low .88
*
.70
*
.70
*
.71
*
.80
*
.69
*
.70
*
.57
*
.48
*
.56
*
.55
*
.75
*
High .43
*
.36
*
.52
*
.34
**
.48
*
.39
♦*
.41
**
.41
* * ♦
.70
*
.76
*
.71
*
.37
*
Atttot 1.03
**
.46 .17 .28 .96 .57 .60 .97 .67 1.25 .50 1.00
Age .29 .28 .86 -.48 1.22 .55 .49 2.02 1.04 .81 1.30 1.71
Sex .06 -1.70 -2.76 -3.64 .47 -1.29 -1.18 -2.11 -2.69 -1.78 -2.09 -3.74
Other -1.73 -2.66 -3.06 1.18 1.21 .56 .16 4.49 -4.25 -3.70 -5.11 -3.09
EngAr -1.93 -2.69 -6.35
***
-3.58 -5.60 -5.64 -5.06 -3.14 -8.73
**
-8.90
**
-9.43
**
-5.61
CGPA 9.17
*
8.01
*
13.1
*
7.74
*
7.78
**
11.6
*
6.97
**
6.78
**
6.87
***
Hours -.04 .00 .02 .05 .06 .00 -.03 -.04 .03
SATv .01 -.00 .01 .03
SATq .01 .01 .01 .01
ACTe .65 .90
**
.78
***
.94
ACTm .35 .37 .46 .74
Hsgpa -1.13 -3.92 -4.05
Hsrnk .13
***
.24
**
.18
(Sex: female = 1, male = 0) significance levels: * = !% ,** = 5%, *** = 10%
Exam scores and college grade point average were the most important
explanatory variables for the final exam in 1994. Engineering and architecture students
performed worse in four samples. The English portion of the ACT and high school rank
were positively related to performance.
35
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
DEPENDENT VARIABLE = ADJUSTED TOTAL POINTS 1994
EQN A B c D F G H I J K L M
n=>
442 439 278 219 211 210 201 ISO 165 165 155 n o
AttTot 9.38 3.78 3.39 3.47 8.67 3.98 3.93 2.70 6.64 12.4 6.07 8.70
* * * * * * * * ♦ * * * *
Age -.60 .10 -2.02 -2.65 .92 -1.24 .55 -.33 3.83 3.50 1.81 4.12
*** ***
Sex 1.89 -8.38 -11.4 -15.1 6.33 -4.32 -2.05 -3.52 -12.1 -8.84 -11.6 -14.8
** ** * *** *** * * *
EngAr 1.68 -5.97 -1.56 -3.02 3.64 1.01 2.85 5.67 -10.5 -11.7 -8.99 -11.1
Other -8.78 -11.9 -8.68 -2.38 1.76 -1.59 -1.52 7.18 -13.3 -12.2 -15.9 -11.8
* **
CGPA 45.0 44.9 48.9 42.0 39.6 45.9 36.2 37.8 34.9
* * * * * * * * *
Hours -.07 .08 .05 .13 .07 .01 -.09 -.02 .09
SATv .03
©
r
-.03 -.02
SATq .17 .09 .09 .10
* * * *
ACTe .49 2.08 .61 1.62
**
ACTm 3.08 4.07 3.78 3.50
* * * *
Hsgpa 3.78 .57 -9.61
Hsrnk .17 .30 .31
_____
* * *
(Sex: female = 1, male = 0) significance levels: * = 1%, ** = 5%, *** = 10%
Attendance was positively related to performance in the course in 1994. Gender
had more of a negative impact on performance than in the previous semester. Grade
point averages and the quantitative test scores were positively related to performance.
Only one of the high school variables, high school rank, was significant once at the 10%
level (regression I).
36
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
DEPENDENT VARIABLE = EXAM 11995
EQN
n=>
A
328
B
317
c
187
D
147
F
123
G
118
H
115
I
85
J
117
K
121
L
116
M
83
Exltot 4.83
*
2.09
**
2.32
* * *
2.49 5.33 3.49
**
3.61 4.09 3.07
***
5.55
*
3.61
**
3.19
Age .18 .00 .48 .15 2.27
**
1.39 1.39 -.83 2.45
*♦
2.84
*
2.03
* * *
1.97
Sex -.87 -2.63 -2.29 .69 2.83 1.24 1.21 5.39 -1.75 -.17 -.35 3.09
EngAr 4.30
***
3.67
*♦*
5.01
* * *
2.46 5.41 5.61
***
5.54 3.79 1.67 1.44 1.54 -1.89
Other -3.54 -3.46
***
-1.47 -3.03 1.65 .18 .09 -.38 -1.95 -1.42 -1.30 -3.24
CGPA 10.4
*
11.1
*
11.2
*
6.90
*
6.74
*
4.04 8.67
*
10.6
*
10.2
*
Hours -.02 -.05 -.04 .01 .00 .17 -.01 .02 .08
SATv .01 .02 .02 .02
SATq .07 .03 .03 .06
**
ACTe .13 .39 .17 .15
ACTm 1.2
*
1.63
*
1.45
*
1.46
*
Hsgpa -1.32 6.48 -9.15
*
Hsrnk .04 -.04
_____
.27
*
(Sex: female = I, male = 0) significance levels: * = 1%, ** = 5%, *** = 10%
Attendance was positively related to performance on exam one. Student age and
enrollment in the colleges of engineering and architecture were positively related to
performance. College grade point average had a positive impact on performance for this
group of students. Quantitative test scores also had a positive impact. Evidence for the
effect of high school statistics on performance was conflicting.
37
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
DEPENDENT VARIABLE = EXAM 2 1995
EQN
n=>
A
298
B
294
c
176
D
138
F
n o
G
108
H
107
I
79
J
111
K
113
L
110
M
78
Ext .56
*
.39
*
.37
*
.37
*
.44
*
.30
♦*
.31
**
.17 .27
**
.41
*
.31
**
.32
Ex2tot 2.73
*
1.73
***
1.18 2.71
*♦
4.38
*
3.13
***
2.91
♦♦♦
4.75
**
2.03 2.56 1.76 2.46
Age .27 .17 1.59
♦*
1.76
**
1.70 1.31 1.28 2.93 -.15 -.17 .22 4.04
Sex -.29 -1.95 -4.26 -1.73 2.35 .21 -.54 4.83 -1.43 -06 -2.62 3.08
EngAr 1.94 1.97 4.52 4.13 4.90 5.95 5.40 6.46 5.27 4.96 4.70 4.41
Other -.99 -1.75 4.61 6.75
* * *
2.57 1.89 1.91 3.95 3.51 4.01 2.93 6.82
CGPA 7.73
*
5.81
♦ *
6.01
**
7.04
**
6.15
***
4.47 7.05
**
5.54 2.65
Hours .00 -.04 -.02 -.00 .00 .06 .04 .02 -.07
SATv .02 .03 .03 .05
***
SATq .02 -.01 -.01 .00
ACTe .29 .39 .28 .82
ACTm .32 .43 .12 .50
Hsgpa 6.06
***
2.31 5.53
Hsmk -.05 -.15 .05
(Sex: female = 1, male = 0) significance levels: * = !% ,** = 5%, * * * = 10%
Exam one score and college grade point average were most important in
explaining performance on exam two in the spring 1995 semester. Attendance
contributed positively (and significantly) to performance in seven of the regressions
above. Student age was positively related to performance in two of the regressions.
38
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
DEPENDENT VARIABLE = EXAM 3 1995
EQN
n=>
A
245
B
244
c
153
D
119
F
93
G
93
H
93
I
69
J
99
K
99
L
98
M
69
Exl .32
*
.26
*
.22
**
.31
*
.09 .04 .02 .08 .33
*
.38
*
.29
**
.40
**
Ex2 .35
*
.30
*
.25
*
.21
**
.25
*
.22
**
.23
**
.25
**
.27
*
.30
*
.29
*
.17
Ex3tot 3.44
*
2.72
*♦
.77 -.27 .61 -.05 .12 -1.19 .91 1.77 1.18 -1.21
Age -.09 -.08 .81 .72 1.00 1.88 1.66 -.96 2.75
**
1.63 2.54
***
1.84
Sex 2.32 1.86 3.37 1.12 11.6
*
10.0
*
10.5
*
9.57
**
3.61 4.22 4.77 3.81
EngAr 2.49 3.09 8.58
♦ ♦
8.63
**
8.33
***
9.49
♦ ♦
9.49
**
10.1
♦ **
7.21 6.03 7.15 8.27
Other 3.91 3.73 8.14
*
8.24
**
12.5
*
13.0
*
12.8
♦
13.0
*
6.73 5.78 6.88
***
9.18
***
CGPA 4.91
*
9.02
*
5.88
***
5.56
* * *
6.68
**
4.48 5.49 6.89
***
3.89
Hours -.06 -.11 -.07 -.13 -.13 -.01 -.15 -.13 -.07
SATv -.01 -.02 -.02 -.00
SATq .09
*
.08
*
.08
*
.08
**
ACTe .27 .31 .27 .31
ACTm .12 .25 .37 .17
Hsgpa -2.86 -3.21 -6.71
Hsrnk -.00 .16 -.05
(Sex: female = 1, male = 0) significance levels: * = !% ,** = 5%, *** = 10%
Both previous exams scores and the quantitative portion of the SAT had positive
impacts on performance. Attendance was only significant in two samples, and females
outperformed males in the four regressions where gender was significant. Non-business
students performed better on exam three in most samples.
39
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
DEPENDENT VARIABLE = FINAL EXAM SCORE 1995
EQN
n=>
A
299
B
297
c
179
D
143
F
113
G
113
H
111
I
84
J
112
K
112
L
111
M
80
Low .77
*
.67
*
.75
*
.70
*
.78
*
.77
*
.78
*
.79
*
.54
♦
.61
*
.59
*
.62
*
High .62
*
.53
*
.36
*♦
.25 .60
*
.56
*
.53
*
.34 .33
***
.45
*♦
.30 .21
Atttot 1.37
*
.94
**
1.56
*
1.31
***
1.96
♦*
1.50 1.42 1.00 1.29
***
1.98
*
1.13 1.04
Age .23 .21 I 68
**
1.78
**
2.30 2.68 3.23
***
3.99 2.77
***
1.51 3.07
**
4.03
Sex -2.16 -3.50 -7.29
**
-8.98
*
-6.00 -6.96
***
-7.30
***
-7.09 -3.87 -2.93 -5.13 -8.50
***
Other 6.31
**
6.63
*
8.81
*
8.61
*♦
8.74
***
9.03
**
8.18
***
9.66
***
13.4
*
12.0
*
12.7
*
16.2
*
EngAr 4.98
♦ ♦♦
5.77
**
7.79
**
7.52
***
7.93 8.82
***
8.82
***
8.29 10.0
**
8.18
***
9.72
**
11.0
***
CGPA 7.34
*
2.62 6.59
**
4.19 1.81 5.99 10.3
*
8.86
**
9.39
**
Hours -.06 -.12 -.13 -.13 -.18 -.22 -.15 -.17 -.20
SATv .00 -.00 -.01 -.02
SATq .02 .01 .00 .01
ACTe -.16 -.00 -.18 -.34
ACTm .57 .71 .34 .25
Hsgpa 6.65
***
5.68 6.71
Hsrnk
. ...
-.14
***
-.14 -.16
(Sex: female = 1, male = 0) significance levels: * = 1%, ** = 5%, *** = 10%
Low and high exam scores, college grade point averages, age, and college choice
were positively related to performance. Attendance was significant in eight regressions,
and age had a positive impact on performance. Gender had a negative impact on
performance in five regressions.
40
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
DEPENDENT VARIABLE = ADJUSTED TOTAL POINTS 1995
EQN A B c D F G H I J K L M
n=> 301 299 180 144 115 115 112 85 112 112 111 80
AttTot 8.14 3.11 3.48 2.78 10.6 6.77 6.16 6.55 1.70 5.39 1.89 .95
* * * * * * * *
Age .72 .31 5.51 5.00 12.4 9.62 11.6 8.95 10.1 8.39 10.1 14.4
* * * * * *** * * * *
Sex -5.99 -14.7 -15.5 -10.8 4.55 -4.80 -.77 7.05 -4.21 1.69 -3.29 7.48
* **
EngAr 18.9 14.2 28.4 20.6 29.2 28.1 33.7 32.3 22.9 19.9 22.3 15.9
** ** *
♦*
* * * * ** * * * **
Other .75 2.08 18.8 17.0 23.2 19.9 18.8 24.8 19.0 15.3 19.3 22.6
** ** ** ** ** ** ** ** **
CGPA 48.7 42.3 46.1 28.5 29.5 27.4 44.5 45.9 39.3
* * * * * * ♦ * *
Hours -.09 -.29 -.23 .11 -.08 .25 -.37 -.36 -.30
SATv .08 .10 .09 .11
*** ***
SATq .20 .09 .08 .12
*
ACTe .13 .98 .15 .37
ACTm 2.83 4.65 3.09 3.46
* * * *
Hsgpa 4.77 .90 -8.17
Hsrnk -.14 -.12 .24
(Sex: female = 1, male = 0) significance levels: * = !% ,** = 5%, *** = 10%
Attendance and student age had a positive impact on performance. Gender had a
negative impact on performance in two regressions. The engineering and architecture
students, as well as the students in colleges other than the former and business,
performed consistently better. College grade point average was positively related to
course performance. SAT scores did not have a consistent impact on total points, but the
math portion of the ACT positively impacted course performance.
41
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Regression Analysis
The regression results for exam one in the three courses stress that attendance,
college grade point average and math SAT and ACT scores are most often positively
related to performance. Females performed significantly worse than males on exam one
in three 1993 and six 1994 samples. Students in “other” colleges also appear to do
worse, especially in 1994.
Exam one score was consistently and positively related to exam two performance
in both 1993 and 1995. Attendance appears as a significant variable in most regressions
for 1993 and 1995. Exam one score and attendance only impact performance
significantly in four 1994 regressions. College grade point average positively impacted
performance in the majority of exam two regressions. Quantitative SAT scores and high
school grade point average positively affected performance in about half of the
regressions in which those variables were included. The variables which most
consistently had a positive impact on exam two performance were exam one score,
attendance, and college grade point average.
The variables which were most consistently related to exam three performance
were exam one scores, exam two scores and college grade point average. Each of those
three variables had a positive impact on performance. Attendance had a positive impact
on performance in 1993 and 1994, but only twice significantly in 1995. Gender
negatively impacted performance in seven regressions 1993, but positively in four 1995
samples. Quantitative SAT scores were positive and significant in all four regressions in
42
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
1995, but none in 1993 or 1994. The math portion of the ACT positively impacted
performance in 1994, and high school grade point average was always positive and
significant in 1993.
Final exam performance was positively impacted by low and high exam scores
and total attendance in the course. Females performed significantly worse on the final
exam five 1995 samples, and significantly better in one 1993 sample. Students in “other”
colleges performed significantly better on the final in 1995. College grade point average
positively impacted performance in 1994 and 1995, but not at all in 1993.
Course performance, as measured by total points earned in the three courses, was
primarily impacted by total attendance, college grade point average, and quantitative
SAT and ACT scores. Females performed significantly worse in the course than males in
three 1993 samples, six 1994 samples, and two 1995 samples. Engineering and
architecture students performed consistently better in 1993 and 1995, and students in
“other” colleges performed significantly better in 1995. In 1993, high school grade point
average had a positive impact on performance, while class rank had a negative impact on
course performance. The high school variables were largely insignificant in 1994 and
1995.
The variables which appeared consistently and significantly in most regressions
run were college grade point average, attendance, exam scores, and quantitative test
scores. Engineering and architecture students often performed significantly better on
43
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
exams and performance in the course. Gender entered often as a significant variable,
with the majority of coefficients strongly negative.
The evidence from this study suggests that attendance has an important positive
impact on performance in these principles of macroeconomics courses. Is the effect of
attendance linear, or are there differences in the marginal impact of attendance on
performance? To examine this possible non-linear impact of attendance, the number of
times a student attended the course before an exam was broken into dummy variables.
Students who attended zero times were excluded. The following table shows the
coefficients and t-statistics on each dummy variable for attendance before exam one. D1
represents students who attended once before the exam, D2 students who attended twice,
and D3 students who were present for all quizzes given before the exam.
A B c D F G H I J K L M
Dl
‘93
7.4 7.5 10.7
**#
9.2 24.7
*
26.7
*
26.6
*
19.1
***
3.3 6.1 2.9 l.l
D2 19.1 14.0 16.6 16.9 29.6 28.4 27.4 22.7 14.3 15.7 14.5 13.5
‘93
* * * * # * * ** ** * ** **#
D3 25.7 17.0 18.4 19.0 36.4 33.5 30.9 25.6 17.1 21.9 16.1 21.3
‘93
* * * * * * * # * * *
Dl
‘94
1.9 3.2 2.8 -.5 1.5 3.6 3.6 -.85 .54 1.0 -1.0 -3.3
D2
‘94
10.7
**
8.2
**
8.2
***
7.8 6.6 5.2 6.2 5.97 14.7
*
17.8
*
15.9
*
14.1
**
D3 18.2 12.6 11.3 11.2 13.8 11.3 11.6 6.96 19.5 24.3 18.6 17.9
‘94
* * * **# ** ** ** # * * *
Dl
‘95
-.4 .0 -2.6 -1.9 2.6 -.4 -.8 -.1 -3.8 -2.7 -2.6 -27.1
***
D2
‘95
3.1 .1 2.0 .65 8.0 3.4 2.4 7.6 -2.0 6.2 .7 -28.3
***
D3
‘95
10.0
***
4.2 4.0 3.8 14.0
**
7.7 7.4 10.0 3.6 11.3 6.0 -20.1
significance levels => *** = 10%, ** = 5%, * = 1%
44
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
There is evidence for the proposition that attendance affects performance non-
linearly. The coefficients tend to increase in size as the number of quizzes attended rises.
Significance levels often rise as well. This pattern of increasing attendance coefficient
size appears to hold for regressions for the other exams and total points in the course.
Attending two or more quizzes before exams boosted exam scores considerably in most
samples. In most regressions, increases in total course attendance led to greater increases
in total points.
Attendance and college grade point average had the most consistent impact on
performance on exams and in the courses. Both variables were consistently positive and
highly significant, and often at the same level. Which of those two variables had a
greater impact on performance? Using the data obtained from the regressions for total
points earned, the effect on mean total points of increasing each variable by one standard
deviation was examined. These calculations were performed for each regression in
which both grade point average and total attendance in the course were entered as
explanatory variables.
1993 B C D G H I J L M
Art 13.42 10.98 13.81 20.35 16.90 18.77 15.95 12.68 14.61
GPA 27.02 19.40 17.32 19.41 15.82 15.35 19.93 17.22 13.78
In 1993, the impact on total points of increasing attendance and grade point
average by one standard deviation varied. In four of the regressions (G, H, I, M), the
impact on total points of increasing attendance by one standard deviation was greater. In
45
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
the remaining five samples, increasing grade point average by one standard deviation had
a greater impact on total points.
1994 B C D G H I J L M
Att 7.56 6.78 6.94 7.96 7.84 5.40 13.28 12.14 17.38
GPA 32.87 30.52 33.75 28.12 26.13 30.27 25.31 24.95 25.14
The impact on 1994 total points of increasing grade point average by one standard
deviation was greater than the impact of attendance on total points in each regression.
1995 B C D G H I J L M
Att 9.33 10.44 8.34 20.31 18.48 19.65 5.1 5.67 1.9
GPA 34.08 28.77 29.99 21.12 21.54 18.87 28.00 29.40 26.69
With the exception of sample I, the impact on total points of increasing grade
point average by one standard deviation was greater than the impact of increasing
attendance by one standard deviation. The return in total points from increasing grade
point average by one standard deviation is greater than the return from increasing
attendance by one standard deviation. In twenty two out of twenty seven cases, the
impact on total points of increasing grade point average by one standard deviation was
greater than the impact of increasing attendance by one standard deviation.
It is possible that students who are “at risk” for performing poorly in these
economics courses may benefit differently from attending class. Students with college
grade point averages less than or equal to 2.5 were considered as a subgroup. The impact
46
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
on average total points of increasing attendance and college grade point average by one
standard deviation was calculated for this group of “at risk” students.
1993 B C D G H I J L M
D n=179 n=128 n=89 o=92 o=89 0=64 o=70 0=69 n=46
Att 17.04 14.74 15.66 20.82 16.63 18.74 20.90 17.55 11.96
GPA 15.79 17.86 11.41 20.36 19.20 14.68 14.32 10.89 1.81
Increasing attendance by one standard deviation has a greater impact on total
points for the “at risk” students in seven 1993 samples.
1994 B C D G H I J L M
n n=l78 n=H2 o=83 o=73 o=7l o=53 o=73 o=69 o=50
Att 6.96 7.36 7.42 6.80 6.64 3.50 16.98 15.54 20.54
GPA 13.93 16.49 23.85 19.11 19.13 29.06 7.80 9.71 15.40
In 1994, increasing college grade point average by one standard deviation has a
greater impact on total points for the “at risk” students more often than increasing
attendance by one standard deviation.
1995 B C D G H I J L M
0 n=115 o=81 o=51 o=45 n=45 o=29 n=51 n=51 o=33
Att 11.76 12.15 12.99 16.29 15.81 30.78 8.1 8.01 2.49
GPA 19.21 10.07 10.35 8.93 8.47 -6.33 -.38 .25 1.31
In 1995, the “at risk” students would benefit more in total points from increasing
attendance one standard deviation. The exception is sample B, in which increasing grade
point average by one standard deviation contributes more to total points. Overall,
increasing attendance by one standard deviation has a larger impact on total points in
eighteen of the twenty seven samples of “at risk” students. These results differ greatly
47
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
from the results found when all students were considered together. It appears that
attendance is more important for the “at risk” students than for all principles students.
48
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
CHAPTER THREE: CONCLUSION
The calculations performed to examine the relationship between attendance and
performance provided evidence for a positive relationship between the two variables.
Attendance appeared to be associated with student performance on the majority of
average exam scores, final exam scores and total points. Three previous studies of
attendance as a determinant of performance have found evidence that going to class
improves exam scores (Romer 1993, Schmidt 1983, Park and Kerr 1990, Durden and
Ellis 1995). Because of the evidence for a positive relationship between attendance and
performance in the three principles courses at Arizona State, attendance was included as
an explanatory variable in the model for performance in the three principles courses.
By controlling for other variables that were expected to be related to
performance, the effect of attendance on performance was more accurately examined.
Attendance had a positive and significant impact on performance in the principles
courses. Even when the students’ experience and ability to perform on course exams was
controlled for, attendance was still positive and significant in most regressions. These
findings were consistent with previous research on attendance and performance in
economics courses.
College grade point averages generally had positive and significant impacts on
performance in the principles of macroeconomics courses. This finding is consistent
with previous studies that have found that college grade point average impacted
performance in economics courses. When previous exam scores were entered as
49
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
explanatory variables, they were often significant Performances on previous economics
exams is positively associated with performance on future exams. There was some
evidence in this study that high school performance impacted performance in the
economics courses. However, high school grade point average and high school rank
were not consistently significant, and the signs of the coefficients were not consistently
positive or negative. It is unclear from the regression analysis what impact the high
school variables had on performance in the macroeconomics courses. This is in contrast
to previous research which found that high school performance consistently contributed
to performance in economics courses.
Student age did not consistently impact performance. This is in contrast to the
findings of Leppel (1984) and Manahan (1983) that age positively impacted performance
in an economics course and on a test of economic understanding. College credit hours
did not have a consistent impact on performance. College credit hours were, however,
positively related to performance on the final exam in five 1993 samples. This fall
semester course was predominantly enrolled in by first semester freshman, with no
experience taking final exams in a college environment. This may explain why the
number of college credit hours is significant in the final exam regressions for this fall
1993 semester, and not the two spring semesters, in which even freshman have one
semester under their belt.
As expected, quantitative test scores were positively related to performance on
exams, and in the course. Scores on the verbal portions of the SAT and ACT exams had
50
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
almost no significant impact on student performance. This finding is in direct contrast to
four previous studies of student performance, which found that verbal test scores were
positively related to performance (Bonello et al 1984, Williams et al 1992, Durden and
Ellis 1995, Ferber et al 1983). The insignificance of the verbal test scores in most
samples suggests that verbal ability matters little for success in these principles of
macroeconomics courses.
Engineering and architecture students performed better in some samples. This
result was not surprising, due to the students’ expressed interest in those quantitative
fields. The positive and significant coefficients on “Other” were surprising because of
the non-quantitative nature of majors within these colleges. Students in these colleges
were expected to perform worse on exams and in the course, and actually performed
better in some samples.
Females performed significantly worse than males in thirty-three samples on
exams and total points. Females significantly outperformed males in only six samples
(two samples on exam two in 1994, and four samples on exam 3 in 1995). It is not
possible to determine the source of this sex difference in performance from the data
available for this study, but we can speculate given the findings from previous research
on this topic. Sex differences in performance in economics courses may be due to
differences in spatial abilities, cognitive abilities, problem solving strategies,
performance on multiple choice exams, preferences for learning economics, or attitudes
in general. The gender difference in performance may also be due to a sex difference, on
51
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
average, in the stock of previous economic knowledge brought to the course. Analysis of
the impact of these variables on student achievement in economics courses may shed
light on the persisting gender differences in performance.
The most important variables for explaining performance on principles of
macroeconomics exams, and in the three courses overall were previous exam scores,
college grade point average, quantitative test scores, college enrollment, gender, and
attendance. Even when variables representing student ability and experience are
controlled for, attendance consistently appears as an important explanatory variable.
Attendance captures effort put forth by the students in these courses. This effort is
strongly linked to their performance in the three sections of principles of
macroeconomics at Arizona State University.
52
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
BIBLIOGRAPHY
Anderson, Gordon, Benjamin, Dwayne, Fuss, Melvyn A. "The Determinants of
Success in University Introductory Economics Courses." Journal of
Economic Education. Spring 1994:99-119.
Bonello, Frank J., Swartz, Thomas R., Davisson, William I. "Freshman-Sophomore
Learning Differentials: A Comment." Journal of Economic Education, Summer
1984:205-210.
Borg, Mary O., Mason, Paul M., Shapiro, Stephen L. "The Case of Effort Variables in
Student Performance." Journal of Economic Education. Summer 1989: 308-313.
Bymes, James P., Takahira, Sayuri. "Explaining Gender Differences on SAT-Math
Items." Developmental Psychology. 1993,29(5): 805-810.
Casey, M. Beth, Nuttall, Ronald, Pezaris, Elizabeth, Benbow, Camilla Persson. "The
Influence of Spatial Ability on Gender Differences in Mathematics College
Entrance Test Scores Across Diverse Samples." Developmental Psychology.
1995,3/(4): 697-705.
Durden, Garey C., Ellis, Larry V. "The Effects of Attendance on Student Learning
in Principles of Economics." American Economic Review. May 1995: 343-346.
Ferber, Marianne A. "Suggestions for Improving the Classroom Climate for Women in
the Introductory Economics Course: A Review Article." Journal of Economic
Education. Spring 1984: 160-168.
Ferber, Marianne A., Bimbaum, Bonnie G., Green, Carole A. "Gender Differences in
Economic Knowledge: A Reevaluation of the Evidence." Journal of Economic
Education. Spring 1983:24-37.
Friedman, Lynn. "Mathematics and the Gender Gap: A Meta-Analysis of Recent Studies
on Sex Differences in Mathematical Tasks." Review of Educational Research.
Summer 1989: 185-213.
Gallagher, Ann M., De Lisi, Richard. "Gender Differences in Scholastic Aptitude Test-
Mathematics Problem Solving Among High-Ability Students." Journal of
Educational Psychology. 1994,86(2): 204-211.
53
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Horvath, Jane, Beaudin, Barbara Q., Wright, Shiela P. "Persisting in the Introductory
Economics Course: An Exploration of Gender Differences." Journal of Economic
Education. Spring 1992:101-108.
Leppel, Karen "The Academic Performance of Returning and Continuing College
Students: An Economic Analysis." Journal of Economic Education. Winter 1984:
47-54.
Lumsden, Keith G. and Scott, Alex. "The Economics Student Reexamined: Male-Female
Differences in Comprehension." Journal of Economic Education. Fall 1987: 365-
375.
MacDowell, Michael A. and Senn, Peter R. and Soper, John C. "Does Sex Really
Matter?" Journal of Economic Education. Fall 1977: 28-33.
Manahan, Jerry. "An Educational Production Function for Principles of Economics."
Journal of Economic Education. Spring 1983: 11-16.
Park, Kang H., Kerr, Peter M. "Determinants of Academic Performance: a Multinomial
Logit Approach." Journal of Economic Education. Spring 1990: 101-111.
Romer, David. "Do Students go to Class? Should They?" Journal of Economic
Perspectives. Summer 1993: 167-174.
Schmidt, Robert M. "Who Maximizes What? A Study in Student Time Allocation."
American Economic Review. May 1983:23-28.
Siegfried, John J. "Male-Female Differences In Economic Education: a Survey." Journal
of Economic Education. Spring 1979: 1-11.
Strieker, Lawrence J., Rock, Donald A., Burton, Nancy W. "Sex Differences in
Predications of College Grades From Scholastic Aptitude Test Scores." Journal of
Educational Psychology. 1993,55(4): 710-718.
Tay, Richard S. "Students' Performance in Economics: Does the Norm Hold Across
Cultural and Institutional Settings?" Journal of Economic Education. Fall 1994:
291-301.
Williams, Mary L., Waldauer, Charles, Duggal, Vijaya G. "Gender Differences in
Economic Knowledge: An Extension of the Analysis." Journal of Economic
Education. Summer 1992:219-231.
54
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
APPENDIX
Table A: the number of students attending a total number of 0,1,2, etc. quizzes
throughout the semester, that also took the final exam (completed the course).
0 1 2 3 4 5 6 7 8 9 10
1993 3 10 6 6 16 22 32 33 46 84 135
1994 4 4 1 1 19 25 34 39 77 98 130
1995 3 7 5 13 13 15 19 29 48 58 91
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Growth, human capital, and technological change in a centrally planned economy: Evidence from Cuba
PDF
Education in Athens Co., Ohio: A clash of cultures
PDF
A study concerning the relevance of using quality measures of education in economic growth models of sub-Saharan African nations
PDF
Explanations, the availability heuristic, and biases of subjective likelihood in economics
PDF
Belize pegged exchange rate: Heading towards devaluation or maintaining credibility
PDF
Big business concentration and its effect on labor market reform policies in Egypt and Mexico
PDF
Cost analysis of three pharmacy counseling programs for diabetics in a health maintenance organization
PDF
Globalization and the decline of the welfare state in less developed countries
PDF
Debt reduction by way of inflation: The case of Lebanon
PDF
Effects of threat and self-focus on consensual bias in majorities
PDF
Dimensions of the locational conflict around siting service facilities for the homeless
PDF
Intragroup evaluations, attitude source, and in-group member derogation
PDF
Bayesian estimation using Markov chain Monte Carlo methods in pharmacokinetic system analysis
PDF
Decentralizing of public finance: Centralizing forces in developing countries
PDF
Course-taking patterns of Latina community college students: On the transfer path
PDF
Effects of personal resource sufficiency on perceived difficulty and desirability of earthquake preparedness
PDF
The roles of China, Hong-Kong, and Taiwan in the Chinese economic region
PDF
Does intense probation monitoring of truants work? An empirical econometric analysis of its effect on school and individual attendance, grades, and behavior
PDF
A feasibility study of support for certification in gerontology
PDF
An enquiry into the effectiveness of export assistance organizations serving southern California SMEs
Asset Metadata
Creator
Sundie, Jill Marie (author)
Core Title
A model of student performance in principles of macroeconomics
School
Graduate School
Degree
Master of Arts
Degree Program
Economics
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
Economics, General,Education, higher,Education, Social Sciences,OAI-PMH Harvest
Language
English
Contributor
Digitized by ProQuest
(provenance)
Advisor
[illegible] (
committee chair
), [illegible] (
committee member
), Kalaba, Robert E. (
committee member
)
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c16-6575
Unique identifier
UC11337005
Identifier
1381609.pdf (filename),usctheses-c16-6575 (legacy record id)
Legacy Identifier
1381609.pdf
Dmrecord
6575
Document Type
Thesis
Rights
Sundie, Jill Marie
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 au...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA
Tags
Education, Social Sciences