PLUS MINUS GRADING STUDY, FALL 1994, SPRING 1995, FALL 1995, AND SPRING 1996, AND FALL 1996

Please address any concerns or comments about this report to carol_gosselin@ncsu.edu

Table of Contents

Executive Summary

Plus/minus grading was implemented beginning with new students in the fall of 1994. Instructors have the freedom to use +/- or not to use it as appropriate. However, University policy states that students enrolled in each section of a multiple section course in which the material, the sequencing of its treatment, and the examinations are common to all sections should be graded on the same scale. To that end, the selection of the grading scale to use (i.e., +/- or not) should be a common selection agreed to by every teacher of the course. This summary presents an analysis of the semesters and course sections where +/- was used.

The central research questions are the extent of utilization of +/- and its effect on average GPAs of various student populations. Analysis of the effect cannot be directly accomplished since only grades are available, and not the quality of the student's activity or instructor's measurement in arriving at that grade. In this study, grades are 'adjusted' by removing pluses and minuses and recomputing the GPA. In 'adjusting' grades by removing pluses and minuses, we may not be accurately reflecting the grades which would have been received if the +/- grading system did not exist. Without consulting each individual instructor about each individual student in each individual course, we cannot be sure that, for example, a B- would have been a B and not a C. Therefore, it is possible that our assumptions could have skewed the numbers slightly in one direction or another.

The findings of this study include the following:

• Plus and minus grades were not assigned evenly in the five semesters studied, in the fall of 1994, more + were assigned than -. For the other four semesters, there were more - assigned than +.

• All but one of those multiple-section courses defined by University policy used +/- grading consistently. The single exception used +/- grading in one section for 2 of the 5 semesters analyzed.

• The proportion of sections and credit hours using +/- has been generally increasing. There is a broad variation in implementation among the individual colleges.

• An increased value for A+ would have affected the GPA of some students. However, since more A-s than A+s are given in spring 1995, fall 1995, spring 1996, and fall 1996, some GPAs would not be increased, even with a higher value for A+.

A Short History of the Issue

Chancellor Monteith, in a memorandum dated February 26, 1993, requested that the Office of the Provost develop strategies necessary to implement +/- grading. The scale recommended by the Faculty Senate was adopted, in order that the faculty may discriminate more precisely among levels of academic performance, except that A+ will not have added value above 4.00.

The policy stated that the +/- grades will be reflected on the transcripts of all students at the end of the fall 1994 semester. The values assigned to + and - will be part of the GPA calculation for all new undergraduate and graduate degree students who enroll in the summer and fall of 1994, and thereafter, and for all new or continuing UGS (undergraduate special) or PBS (post-baccalaureate special) students. However, +/- grades without grade points will be used for continuing undergraduate and graduate degree students whose first enrollment was prior to the summer and fall of 1994. This exemption will end in the 1998 fall semester, at which time all students will receive the grade points assigned to the +/- grades they receive.

Concern was expressed by some students that the introduction of +/- grading would significantly reduce grade point averages based on a report from an earlier experiment in which NCSU faculty could assign +/- grades that did not appear on the transcript or effect GPAs. In the experiment, it appeared that twice as many minuses as pluses were assigned.

The scale for +/- grading is as follows:

• A+ 4.00

• A 4.00
• A- 3.67
• B+ 3.33
• B 3.00
• B- 2.67
• C+ 2.33
• C 2.00
• C- 1.67
• D+ 1.33
• D 1.00
• D- 0.67
• F 0.00

The Research Questions Addressed

The following data are presented in this report for both graduate and undergraduate students to show the utilization of +/- grading.

1. Frequency distribution of grades (Table 1). The tables show how many pluses and minuses were given in each semester. The assumption was that more pluses would be given than minuses. In all semesters studied except fall 1994, this is not the case. Also shown is the breakdown of grades and the mean GPA for each semester.
2. Consistency of use of + and - grades in multiple section courses (Question 2). There was a policy that stated that all students enrolled in each section of a multiple section course in which the material, the sequencing of its treatment, and the examinations are common to all sections should be graded on the same scale. To that end, the selection of the grading scale to use (i.e., +/- or not) should be a common selection agreed to by every teacher of the course. Only one course was found that did not adhere to this policy over 2 of the 5 semesters analyzed.
3. Proportion of sections for which faculty assigned + and - grades and proportion of credit hours delivered in sections for which + and - grades were assigned (Table 3).

For items 2 & 3 involving proportions of sections and hours and consistency over multiple section courses, all registered students in the studied semesters were used in the calculations.

The following data are presented as evidence of the effect of +/- on various student populations. The direct comparison of grades with and without +/- is problematic since such an analysis is not feasible given the need to determine intent of the instructor in assigning a grade.

1. Comparison between GPAs with and without plus and minus grades (Table 4).
2. Distribution of differences between GPA computed with +/- grades and with all A+ given a value of 4.33. This would show if the lack of an increased value for A+ is the main reason for any significant differences in GPA (Table 5).
3. Segmentation of the student body at the end of each semester with respect to those whose GPA >= 2.0 (3.0 for Graduate students), those on academic warning, and those suspended.

   The fall 1994, fall 1995, and fall 1996 cohorts are the population which is affected by +/- grading and, therefore, was used for items 4, 5, and 6, and for the frequency distributions of grades (Item 1). The relevant course data for these students, by individual student, was used for the fall 1994, spring 1995, fall 1995, spring 1996, and fall 1996 semesters. For each grade received, the quality points were calculated by multiplying the numeric value of the letter grade by the number of credit hours for the course.
   

   In items 4 and 6, the 'adjusted' quality points were calculated using no + and -. For item 5, the 'adjusted' quality points were calculated by assigning a value of 4.33 to A+, thus negating any effect of having no higher value added for A+.
   

4. Breakdown of Grades and Mean GPA by Semester for all students for the 21 semesters from fall 1986 through fall 1996, by graduate and undergraduate students. All grades in all classes were used in this analysis. Tables show the number of each grade received in each semester, and graphs show the comparison on the semesters. All plus and minus grades were counted in this analysis, even though there may have been some plus and minus grades awarded which were not included in the calculation of the student's GPA. A table and graph are included showing the progression of the GPA over the course of the 21 semesters for both graduate and undergraduate students.

QUESTION 1: Frequency Distribution of Grades and Mean GPA, at Undergraduate and Graduate Level

The counts and percentages shown for each group of students are only for those students for whom +/- grading was used in calculating the GPA. 'Count' represents each instance of student and course where a grade was assigned. Except for fall 1994, more minuses were given than pluses. Many more A- grades were given than A+ grades.

The following tables show the mean GPA by semester for graduate and undergraduate students in the fall 1994, fall 1995, and fall 1996 cohorts, students entering the university in those 3 semesters. These are the students who were affected by plus/minus grading. The first table shows the actual mean GPA, calculated by summing all of the quality points and dividing by the total credit hours. The second table shows the mean GPA computed the same way but without using the plusses and minuses.

QUESTION 2: Was +/- Grading Used Consistently In Multiple Section Courses with a Common Final Examination?

In the +/- implementation, the stated policy requires that plus and minus grades be used consistently in multiple section courses in which the material, the sequencing of its treatment, and the final examination are all common.

Only one course, ACC 210, did not use +/- grading consistently. In fall 1994, no plus or minus grades were given in the 16 sections taught. In spring 1995, one C- was given in one section out of the 15 sections taught. In fall 1995, 6 C-'s and 3 A+'s were given in one section out of the 13 sections taught. In spring 1996 and fall 1996 , no plus or minus grades were given in the sections taught.

QUESTION 3: Proportion of Sections and Credit Hours using +/- Grading

The following tables, organized by college/school, show the proportion of total sections and total credit hours for which +/- grading was used in each semester, for graduate and undergraduate level courses.

For most of the schools, an increase is evident each semester in the number of sections using +/- grading.

For most of the schools, an increase is evident each semester in the number of credit hours using +/- grading. Where the percentage of credit hours is greater than the percentage of sections for the same school in the same semester, we can assume that the larger sections are the ones that are using +/- grading. Likewise, if the percentage of credit hours is less than the percentage of sections for the same school in the same semester, we can assume that the smaller sections are the ones that are using +/- grading.

QUESTION 4: Effect on GPA Using +/- Grading

The effect on GPA was calculated by starting with the actual GPA and subtracting a GPA computed by removing all +'s and -'s and assigning a numerical value to the resulting grade. A negative difference implies that the actual GPA using +/- grading is lower than it would have been without +/- grading. In this case, the student received more - than +. A positive difference implies that the actual GPA using +/- grading is higher than it would have been without +/- grading. In this case, the student received more + than -. Of course there is no way of knowing if a B+ would have been an B instead of an A. The assumption is that a B+ would have been a B.

The percent of students in each column category is recorded here. It is important to remember that this is analysis of apparent effect, the intent of instructor in assigning original grade cannot be construed from this data.

There is, of course, a wide variation in the size of the population among the colleges. The population used in these calculations is increasing in size over time, since it includes only those students eligible for +/- grading, a number which increases with each new cohort after fall 1994.

The majority of students showed no difference in this calculation. Plus/minus grading produced a lower GPA for about 11% of the graduate students and 11% of the undergraduate students in fall 1994, 15% of the graduate students and 33% of the undergraduate students in spring 1995, 20% of the graduate students and 33% of the undergraduate students in fall 1995, 20% of the graduate students and 34% of the undergraduate students in spring 1996, and 22% of the graduate students and 34% of the undergraduate students in fall 1996.

The group with a higher GPA using +/- grading has also increased over time starting with 7% of graduate students and 8% of undergraduate students in fall 1994, increasing to 11% of graduate students and 23% of undergraduate students by fall 1996.

Fall 1994

Percent of Graduate students

Percent of Undergraduate Students

Spring 1995

Percent of Graduate students

Percent of Undergraduate Students

Fall 1995

Percent of Graduate students

Percent of Undergraduate Students

Spring 1996

Percent of Graduate students

Percent of Undergraduate Students

Fall 1996

Percent of Graduate students

Percent of Undergraduate Students

QUESTION 5: Distribution of Differences in GPA, with A+ Given Value of 4.33

The potential effect of giving A+ a value of 4.33 was calculated by computing the student's GPA with A+ given a value of 4.33, to conform to the +/- grades used for B, C, D, and F, and then subtracting the result from the recorded GPA using +/- grading, as defined. A negative difference implies that the actual GPA is lower than it would have been with A+ having a higher value.

The percent of students in each column category is recorded here. It is important to remember that this is analysis of apparent effect, since the intent of instructor in assigning original grade cannot be construed from this data.

With the addition of a value of 4.33 for A+, about 4% of the graduate students and 4% of the undergraduate students in fall 1994 would see an increase in the GPA with +/- grading. The figures for spring 1995 are about 7% for graduate students and 7% for undergraduate students, for fall 1995 about 7% for graduate students and 11% for undergraduate students, for spring 1996 about 8% for graduate students and 9% for undergraduate students, and for fall 1996 about 9% for graduate students and 11% for undergraduate students.

Fall 1994

Percent of Graduate students

Percent of Undergraduate Students

Spring 1995

Percent of Graduate students

Percent of Undergraduate Students

Fall 1995

Percent of Graduate students

Percent of Undergraduate Students

Spring 1996

Percent of Graduate students

Percent of Undergraduate Students

Fall 1996

Percent of Graduate students

Percent of Undergraduate Students

QUESTION 6: Segmentation of the Student Body With Respect to Academic Warning and Suspension

The following tables, organized by undergraduate and graduate students, show the segmentation of the students with respect to academic warning, suspension, and probation

for each semester. Only those students for whom + and - grades are part of the GPA calculation are included in these tables.

The "calculated" GPA was computed by adding the quality points for each course (using +/- grading) and dividing by the total credit hours taken at that point. The "adjusted" GPA was computed the same way, but without using +/- grading. Transfer hours, if present, were then added to the total credit hours taken to find the total hours for each student. The students were then divided into categories (Academic Warning, Suspension, and Probation) based on the policies in the Advisor's Handbook for 1994-1995. This policy is no longer in effect, but provides a clear delineation of the population.

For undergraduates, the difference in segmentation from the calculated GPA to the adjusted GPA is less than 1% for each category. For graduates, the difference is as high as 1.6% for fall 1996. Therefore, it appears that the use of +/- grading is having an impact on a small percentage of this population.

Fall 1994

Spring 1995

Fall 1995

Spring 1996

Fall 1996

QUESTION 7: Breakdown of Grades and Mean GPA by Semester

The following tables show the breakdown of grades for undergraduate and graduate classes for each of the 21 semesters from fall 1986 through fall 1996. All plus and minus grades are included in these totals, whether or not the student actually received the point value assigned to plusses and minuses.

Undergraduate

Sem	A+	A	A-	B+	B	B-	C+	C	C-	D+	D	D-	F
F86		18561			26115			19616			6538		6882
S87		28098			24094			17763			5488		4052
F87		20089			26346			19324			5886		6397
S88		18994			27910			17989			5145		5576
F88		21914			27491			19725			6075		6683
S89		20415			25463			17579			5187		5452
F89		22357			27556			19289			6031		6550
S90		21615			26235			17699			5426		5384
F90		23999			28258			19560			6115		6051
S91		22705			26764			17757			5588		5238
F91		26048			29957			19322			6045		5958
S92		25065			27697			17762			5643		5358
F92		27282			29398			18787			5673		5693
S93		25751			27063			17139			5134		5344
F93		27852			28936			18320			5360		5835
S94		25300			26532			16719			5064		5465
F94	797	25728	1362	1081	26522	965	621	17089	419	121	5590	85	6313
S95	1412	20224	3911	3270	20535	3195	2086	12780	1599	535	4041	319	5569
F95	1605	22969	4379	3596	21269	3783	2548	13215	1709	531	4217	425	5636
S96	1848	20063	4387	3818	19257	3412	2273	12012	1682	491	3798	388	4479
F96	2209	21871	4670	4038	21305	3701	2577	12838	1857	658	4072	462	5463

Graduate

Sem	A+	A	A-	B+	B	B-	C+	C	C-	D+	D	D-	F
F86		3988			3158			567			63		116
S87		3917			2793			506			50		109
F87		3975			3075			561			41		125
S88		4004			2667			472			45		118
F88		4147			3133			535			75		146
S89		4051			2748			503			54		129
F89		4289			3218			535			59		133
S90		4309			2913			471			46		133
F90		4586			3175			453			47		110
S91		4557			2931			463			55		126
F91		4798			3196			542			64		120
S92		5712			2979			470			45		106
F92		5317			3487			527			43		134
S93		5127			3091			444			31		122
F93		5650			3414			555			64		137
S94		5503			2930			377			26		115
F94	193	5071	346	253	3076	107	28	479	19	1	52	1	142
S95	282	4592	711	406	2364	148	29	344	9	1	38	2	111
F95	322	4959	922	601	2612	201	52	289	18	5	33	0	68
S96	355	4586	797	539	2193	160	39	328	22	0	33	1	53
F96	370	4575	887	555	2387	201	64	294	21	2	38	0	65

Below is a graphic representation of the distribution of grades for undergraduates across the 21 semesters, where plusses and minuses are ignored. It shows the distribution of A's, B's, C's, D's, F's across the 21 semesters from fall 1986 through fall 1996. The second graph shows the distribution of plus/minus grades for the undergraduates for the 5 semesters in which it was used.

Below is a graphic representation of the distribution of grades for graduate students across the 21 semesters, where plusses and minuses are ignored. It shows the distribution of A's, B's, C's, D's, F's across the 21 semesters from fall 1986 through fall 1996. The second graph shows the distribution of plus/minus grades for the graduate students for the 5 semesters in which it was used.

The following table shows the mean GPA calculated using all plus and minus grades over the 21 semesters from fall 1986 through fall 1996, for both undergraduate and graduate students. All plus and minus grades were used in these calculations. However, there were probably some students who received plus and minus grades for whom the value associated with these grades was not used in the computation of their individual semester GPA. Following the table is a graphical representation of the changes in GPA over the 21 semesters for undergraduate and graduate students.