Student Perspectives on Grade Changes from Test to Test
Baron Perlman and Lee I. McCann
University of Wisconsin-Oshkosh

Abstract

We surveyed undergraduate psychology students whose exam grades had gone up or down a minimum of one full letter grade from one exam to the next. We asked about their perceptions of how their studying and preparation differed for each exam. These data may be useful for students and assist faculty in advising about better exam preparation and performance.

 

Student Perspectives on Grade Changes from Test to Test

When students want to improve their exam performance what should faculty tell them? We found no information about influences on student test performance in popular teaching books (e.g., Davis, 1993; McKeachie, 1999) and few empirical studies investigating the relationship between student classroom behavior (e.g., lecture notes; Baker & Lombardi,1985) or external influences on students (e.g., life events; De Meuse,1985) and exam performance.

Equally or more important, when faculty give such advice, how do they maximize the probability that students will follow through on their suggestions? "Letters to Successors" (Brookfield, 1995) is a technique whereby, at the end of a semester, students list what it is essential to know and do to survive or succeed in the course, and the instructor presents this information to students the next time the course is taught. This student perspective may be more influential than faculty advice. We used a variation of this method to seek student perspectives on the specific issue of how their studying and preparation influence exam performance.

 

Method

Participants

Eleven of 12 full-time faculty members in a regional public university sampled undergraduate students whose exam score went up or down substantively (at least one full letter grade) in their psychology courses (n = 26) during the Fall 1998 and Spring 1999 semesters. Participants were 340 students (165 first year, 175 sophomores, juniors, and seniors; 218 women, 122 men). Most students (n = 246) were enrolled in General Psychology, 94 in sophomore and upper level courses.

Questionnaire

Students responded to a questionnaire (based on our own and departmental colleagues' teaching experiences) that listed 19 behaviors that might improve and 17 that might negatively affect test performance; students could add "other" responses. We asked for students' gender, year in college, and course level (first year, second, third, or fourth). The questionnaire took 5 min to complete.

Procedure

Faculty identified students whose exam scores had gone up or down at least one full letter grade (e.g., not a C+ to a B-). Faculty had the freedom to decide whether that definition was too lenient (i.e., a student's full letter grade change was not a substantive one based on the faculty member's grading criteria). For each student with a substantive grade change, faculty highlighted a column on the questionnaire to indicate whether the student "did better" or "did worse," and distributed questionnaires before or after class. Students returned the questionnaire in sealed envelopes to the instructor or department secretary (name and office number listed). Students checked all items they thought contributed to their exam grade as compared to the previous test and identified the one item they believed most important in explaining their grade change. Faculty instructed students to participate only once, and one item asked students to check if they had filled out the questionnaire previously in this or another course.

After the second exam faculty told all students:

I am assisting in a study to determine what factors contribute to a student doing better or worse on exams. We hope the information you provide will help us to better advise students in the future. This study is optional; you do not have to participate. All data are confidential and your responses cannot be attached to your name. There is potential benefit to participating. In reading the list of behaviors related to test grades, you may get ideas on how to do better in this course. By turning in this brief questionnaire, you agree to participate in the research.

We could not sample students in courses with only two exams because students typically took the second exam on the last day of class. We used courses that had three (n = 7, 27%) or four exams (n = 19, 73%), comparing students' performance from the first to second or second to third test. Exams varied from all multiple choice to all essay.

Results

Of 550 questionnaires distributed, 419 were returned (76%). We discarded 79 (student filled one out previously, n = 43; missing demographic information, n = 24; students checked items in both the "did better" and "did worse" columns, n = 8; or students who did better checked "did worse" items or vice versa, n = 4). The useable response rate was 62%; we do not have response rate data separated by better or worse exam grades. Students listed significantly more reasons for test grades going up (M = 5.1, SD = 1.2, range = 1 to 12) than down (M = 4.1, SD = 2.0, range = 1 to 12), t(338) = 4.95, p < .01.

We expected more data for students whose grades had risen than fallen because many students improve as a course progresses, and almost 70% of the useable sample did involve exam grades going up (n = 236, 81 men, 155 women; 124 first year, 112 sophomores, juniors, or seniors; 184 enrolled in General Psychology, 52 in second- third- and fourth-year courses). We compared men and women, first-year students with all others, and General Psychology with students in other courses for the (a) 10 most frequently checked behaviors for test scores rising and (b) most frequently checked "most important" items. We used a Bonferroni correction (Hays, 1988) to minimize Type I error and a significance level of .05. None of the 40 chi square analyses were significant, indicating that gender, course level, and year in school were not related to perceived reasons for doing better. We present overall data in Table 1.

Table 1

Contributions to Improved Examination Grades

Reason

N
(%)a

______________________________________________________________________________________

Studied more hours

156
(66)

Read the book/text/readings more carefully

154
(66)

Studied in a quiet environment, few interruptions

117
(50)

Used the study guide more

75
(32)

Read the text/readings more times

72
(31)

Took better lecture notes

54
(23)

Took notes on readings, had not before

42
(18)

Cut back on time socializing/partying

41
(17)

Got lecture notes I missed from someone

31
(13)

Made flashcards

27
(11)

Studied with someone, had not before

25
(11)

Worked fewer hours at part- or full-time job

18
(8)

Easier/more interesting material, understood the material betterb

18
(8)

Talked with the instructor, clarified how and what to study

17
(7)

Came to class more regularly

16
(7)

Family/personal crisis or problem resolved

13
(6)

Different approach to studying (not defined) b

12
(5)

Reviewed the last exam

10
(4)

Lower stress level, more sleepb

7
(3)

Worked with a tutor

6
(3)

Studied lecture notes more carefullyb

5
(2)

Began using university support services (e.g., Reading Center, Writing Center)

4
(2)

Paid more attention in classb

3
(1)

Bought the text, had not had it previously

2
(1)

Other classes did not take time awayb

2
(1)

_____________________________________________________________________________________
a Based on 236 students' responses.
b Added by students under the "other" category.

 

Students identified studying more hours (n = 38, 16%) as the single most important behavior in improved exam performance, followed by reading more carefully (n = 37, 16%), using the study guide more (n = 24, 10%), and reading material more times (n = 14, 6%). Students rated 19 other behaviors as a "most important" perceived influence on raised exam scores.

Just over 30% of the useable sample (n = 104) returned questionnaires for falling exam grades (41 men, 63 women; 43 first year, 61 sophomores, juniors or seniors; 62 enrolled in General Psychology, 42 in second-, third-, and fourth-year courses). We again used a Bonferroni correction to minimize Type I error. None of 33 chi square analyses for the 10 most frequently checked behaviors related to test scores falling or for most frequently checked "most important" items were significant (p < .05), indicating that gender, course level, and year in school were not related to perceived reasons for doing worse. We present lowered grade data in Table 2.

Table 2

Contributions to Worse Examination Grades

Reason

N
(%)a

_____________________________________________________________________________________

Studied fewer hours

62
(60)

Read the book/text/readings less carefully

50
(48)

Read the text/readings fewer times

43
(41)

Family/personal crisis or problem interfered

32
(31)

Missed more classes

27
(26)

Missing lecture notes

27
(26)

Studied in a noisy environment, many interruptions

24
(23)

Worked more hours at part- or full-time job

24
(23)

Spent more time socializing/partying

22
(21)

Used the study guide less

21
(20)

Took worse lecture notes

19
(18)

Did not make flashcards this time

14
(14)

Stopped taking notes on readings

14
(14)

Took notes on reading, did not help and took time away
from more productive studying

8
(8)

Other tests/assignments due at the same timeb

6
(6)

Studied with someone, actually was a hindrance

5
(5)

No book, did not read the bookb

4
(4)

Less straight forward test questions, more difficult materialb

4
(4)

Do not know, studied the sameb

3
(3)

Stopped studying with someone

2
(2)

Missed class due to extracurricular activityb

2
(2)

Took different kind of notes on readingb

2
(2)

Could not finish test on timeb

1
(1)

Distracted in classb

1
(1)

Studied wrong information, did not know what to focus onb

1
(1)

Other: lack of sleep, did not feel like studying, could not
stay on task, too much TVb

4
(4)

______________________________________________________________________________________
a Based on 104 students' responses.
b Added by students under the "other" category.

 

The single most important behavior related to a poorer test grade was studying fewer hours (n = 22, 21%), followed by personal problems or crises (n = 10, 10%). Students rated 17 other behaviors as a "most important" perceived influence on falling test performance.

Discussion

In most cases, the data complement accepted faculty wisdom on this topic. Some students perceive studying harder as related to rising test grades and less studying to lower grades. But more is going on.

We were surprised by some of the results. Students did not rate some behaviors faculty might expect to affect test grades (e.g., regularly attending class, easier or interesting course material, tests due at the same time) as doing so. Also, students' perceptions did not differ based on year in school, gender, or course level for either rising or falling grades. These findings suggest that the data in Tables 1 and 2 provide good overall lists of what "to do" and "not to do" when studying for exams for any student. Another surprise was that the data did not support student helplessness and lack of control over exam performance as explanations for doing better or worse. We found few comments such as "I don't know what happened" or "I studied the same way, my grade just went down."

Sharing these student perceptions may help all students maximize exam performance, and they may be useful when faculty advise individual students. Because many behaviors perceived as related to exam performance seem independent of any specific course or teacher behavior (e.g., life crises, hours worked), faculty should emphasize to students that problems with exam performance, or strategies to improve them, may have effects beyond the current course, influencing academic performance throughout a student's educational career.

 

References

Baker, L., & Lombardi, B. R. (1985). Students' lecture notes and their relation to test performance. Teaching of Psychology, 12, 28-32.

Brookfield, S. D. (1995). Becoming a critically reflective teacher. San Francisco: Jossey-Bass.

Davis, B. G. (1993). Tools for teaching. San Francisco: Jossey-Bass.

De Meuse, K. P. (1985). The relationship between life events and indices of classroom performance. Teaching of Psychology, 12, 146-149.

Hays, W. L. (1988). Statistics (4th ed.). New York: Harcourt Brace Jovanovich.

McKeachie, W. J. (1999). Teaching tips: Strategies, research and theory for college and university teachers (10th ed.). Boston: Houghton Mifflin.

 

Notes

1. This article is based on a poster presented at the 23rd Annual National Institute on the Teaching of Psychology, January 2001, St. Petersburg Beach, FL.

2. We gratefully acknowledge the assistance of our students and departmental colleagues.

3. A copy of the questionnaire is available from the authors.

4. Send correspondence to Baron Perlman, Department of Psychology, University of Wisconsin, Oshkosh, WI 54901; e-mail: perlman@uwosh.edu.

 [ Return to Teaching Research ]
[
Return to Teaching & Curriculum Index ]
[
Return to Psychology Department Index ]