Gender Differences in College Mathematics Grades and SAT-M Scores: A Reanalysis of Wainer and Steinberg

Wainer and Steinberg (1992) showed that within broad categories of first-year college mathematics courses (e.g., calculus) men had substantially higher average scores on the mathematics section of the SAT (SAT-M) than women who earned the same letter grade. However, three aspects of their analyses may lead to unwarranted conclusions. First, they focused primarily on differences in SAT-M scores given course grades when the more important question for admissions officers is the difference in course grades given scores on the predictor. Second, they failed to account for differences among calculus courses (e.g., calculus for engineers versus calculus for liberal arts students). Most importantly, Wainer and Steinberg focused on the use of SAT-M as a single predictor. A reanalysis presented here indicated that a more appropriate composite indicator made up of both SAT-M and high school grade point average demonstrated minuscule gender differences for both calculus and precalculus courses.     

Analysis of Student Performance in Large-Enrollment Life Science Courses

This study examined the historical performance of students at Michigan State University in 12 life sciences courses over 13 yr to find variables impacting student success. Hierarchical linear modeling predicted 25.0–62.8% of the variance in students’ grades in the courses analyzed. The primary predictor of a student''s course grade was his or her entering grade point average; except for the second course in a series (i.e., Biochemistry II), in which the grade for the first course in the series (i.e., Biochemistry I) was often the best predictor, as judged by β values. Student gender and major were also statistically significant for a majority of the courses studied. Female students averaged grades 0.067–0.303 lower than their equivalent male counterparts, and majors averaged grades were 0.088–0.397 higher than nonmajors. Grades earned in prerequisite courses provided minimal predictive ability. Ethnicity and involvements in honors college or science residential college were generally insignificant.     

Effects of Improving the Reliability of the GPA on Prediction Generally and on Comparative Predictions for Gender and Race Particularly

The reliability of a method of adjusting grade point averages for differences in departmental grading standards was examined, as were the effects of such adjustments on the predictive validity of high school grades, SAT scores, and achievement test scores. The index of differential grading standards for all on-time graduates of the Dartmouth College class of 1986 was quite reliable, and its use in adjusting grade averages increased predictive validity, reduced its erosion over years, reduced the apparent underprediction of women, and improved predictions for blacks. Differential group enrollment in courses in the science division seems to account for much of the effect of adjustment on grades. Improvement in the reliability of the criterial grade averages also was shown to have similar effects on gender and race prediction in another data set     

Measuring teaching quality in higher education: assessing selection bias in course evaluations

Student evaluations of teaching (SETs) are widely used to measure teaching quality in higher education and compare it across different courses, teachers, departments and institutions. Indeed, SETs are of increasing importance for teacher promotion decisions, student course selection, as well as for auditing practices demonstrating institutional performance. However, survey response is typically low, rendering these uses unwarranted if students who respond to the evaluation are not randomly selected along observed and unobserved dimensions. This paper is the first to fully quantify this problem by analyzing the direction and size of selection bias resulting from both observed and unobserved characteristics for over 3000 courses taught in a large European university. We find that course evaluations are upward biased, and that correcting for selection bias has non-negligible effects on the average evaluation score and on the evaluation-based ranking of courses. Moreover, this bias mostly derives from selection on unobserved characteristics, implying that correcting evaluation scores for observed factors such as student grades does not solve the problem. However, we find that adjusting for selection only has small impacts on the measured effects of observables on SETs, validating a large related literature which considers the observable determinants of evaluation scores without correcting for selection bias.     

The Relationship of Essay and Multiple-Choice Scores With Grades in College Courses

Essay and multiple-choice scores from Advanced Placement (AP) examinations in American History, European History, English Language and Composition, and Biology were matched with freshman grades in a sample of 32 colleges. Multiple-choice scores from the American History and Biology examinations were more highly correlated with freshman grade point averages than were essay scores from the same examinations, but essay scores were essentially equivalent to multiple-choice scores in correlations with course grades in history, English, and biology. In history courses, men and women received comparable grades and had nearly equal scores on the AP essays, but the multiple-choice scores of men were nearly one half of a standard deviation higher than the scores of women.     

Announced Repetitive Tests as a Basis for Self-Directed Study and Evaluation

College students voluntarily took all their courses or one course on a pass-fail basis. The mean grade point average (GPA) before conversion to pass-fail for freshmen taking all their courses on a pass-fail basis was 1.67 (C-), which is significantly lower than the 2.26 (C+) for controls who wanted but were denied pass-fail grading. Even after returning to conventional grading the former pass-fail students continued to get significantly lower grades than controls. Juniors taking one course on a pass-fail basis received significantly lower grades, before conversion, in their pass-fail course (mean 2.07) than did controls who wanted but were denied pass-fail grading (mean 2.40). There was no compensatory improvement in the grades received in non-pass-fail courses.     

Grade stability in a criterion‐referenced grading system: the Swedish example

This study investigates empirically the mechanisms behind the increasing grade point averages in Swedish upper secondary schools. Four hypotheses are presented as plausible explanations; improved student achievements, student selection effects, strategic behaviour in course choices, and lowering of grading standards. The analysis is based on extensive data, and focuses on grades and test scores from upper secondary school graduates over a 6‐year period. The result shows that the increase in grade point averages cannot be explained by better achievements, selection effects or course choices, which means that standards have been lowered, which is interpreted here as grade inflation. The grade inflation is most likely an effect of the leniency in the grading system in combination with pressure for high grading, related to the upper secondary school grades’ function as an instrument for selection to higher education.     

Academic Performance,Age, Gender,and Ethnicity in Online Courses Delivered by Two-Year Colleges

This study investigated the effects the demographic variables age, gender, and ethnicity and their interactions had on academic performance in online courses delivered by public two-year colleges in Kentucky. The study controlled for previous academic performance measured by cumulative grade point average (GPA). The study used a random sample (N = 320) of all students who had enrolled in at least one online course delivered by the institutions of the Kentucky Community and Technical College System in the spring 2008 semester. A linear hierarchical multiple regression acting as ANCOVA served as the main analysis, with the order entry as follows: cumulative GPA; independent variables (age, gender, ethnicity); interaction vectors; and product vectors. Final course grade served as the dependent variable. The results of the analysis indicated that only cumulative GPA was a significant predictor, explaining approximately 40% of the variance of the final grade. Although differences in final grades were present among the variables age and ethnicity, these differences disappeared when controlling for cumulative GPA. Significance of the results and recommendations for future research are discussed.     

Assessing the Effectiveness of a College Freshman Seminar Using Propensity Score Adjustments

Researchers investigated the impact that a first-year college experience course had on students?? first-year grade point averages (GPAs) and retention rates. A sample of 109 first-year students enrolled in the course was compared to a sample of 326 students from the same university who had not taken the course. The goals of the experience course were to reduce attrition, increase grade point averages, and enhance academic skills. Without accounting for selection bias, those who took the course had similar retention rates and lower GPAs than those who did not take the course. After matching on propensity scores, the negative effects of the program on GPA were nullified and those in the program were more likely to enroll for a second year. Although the benefits from the course were weak, the positive impact of the program was more apparent after accounting for individual differences.     

Correlation between grade point averages and student evaluation of teaching scores: taking a closer look

One of the most contentious potential sources of bias is whether instructors who give higher grades receive higher ratings from students. We examined the grade point averages (GPAs) and student ratings across 2073 general education religion courses at a large private university. A moderate correlation was found between GPAs and student evaluations of teaching (SETs); however, this global correlation did not hold true for individual teachers and courses. In fact, there was a large variance in the correlations between GPAs and SETs, including some teachers with a negative correlation and a large variance between courses.     

Comparison of anxiety levels of students in introductory earth science and geology courses

Students in geology and earth science courses were not anxious about taking these courses as measured by the State-Triat Anxiety Inventory. Students in each course were split into two grade groups. It was shown that students whose grades were in the higher group had low anxiety, which continued to be reduced during the course. Students in the lower grade group had high anxiety, which increased during the course. There was no difference between anxiety scores of males and females. In general, students who planned to elect additional courses had lower state anxiety and higher grades than students who did not plan to elect additional courses in geology and earth science.     

Changes in race and sex stereotype threat among diverse STEM students: Relation to grades and retention in the majors

In laboratory studies, induced stereotype threat shows negative effects on academic performance and learning. Is the relation between stereotype threat and grades robust in naturalistic settings, specifically in introductory STEM courses? We gathered data on two new measures we term race and sex stereotype bias, which were administered four times over the course of introductory chemistry and biology courses for STEM majors (N = 1358). Patterns of growth for all stereotype bias measures showed a discontinuous pattern, with increases during each semester (fall and spring) and decreases between semesters. For all stereotype bias measures, sophomores scored significantly higher than freshmen, and juniors scored in between. For the sex stereotype bias measure, females scored significantly higher than males. There were no race or sex differences on slopes of growth; though groups began at different levels, all grew at the same rate. There was little relation between grades and stereotype bias when analyzed by race; Asian students showed the largest number of significant – albeit small – correlations (3) and Black students the fewest (none). Correlations between grades and sex stereotype bias were significant and negative – but small – only for males. Results support a point made by Steele in 1997 but neglected since then; stereotype threat may affect only a small sub-portion within stereotyped groups. We argue that variables other than stereotype threat might be better targets for research attempting to explain gaps in STEM achievement and retention.     

Examination of Gender Bias in University Admissions

In this study we examined whether the measures used in the admission of students to universities in Israel are gender biased. The criterion used to measure bias was performance in the first year of university study; the predictors consisted of an admission score, a high school matriculation score, and a standardized test score as well as its component subtest scores. Statistically, bias was defined according to the boundary conditions given in Linn (1984). No gender bias was detected when using the admission score (which is used for selection) as a predictor of first-year performance in the university. Bias in favor of women was found predominantly using school grades as predictor whereas bias against women was found predominantly in using the standardized test scores. It was concluded that the admission score is a valid and unbiased predictor of first-year university performance for the two genders.     

Multilevel analysis of the effects of student and course characteristics on satisfaction in undergraduate liberal arts courses

The purpose of this study is to analyze the effects of student and course characteristics on student satisfaction in courses by using a two-level hierarchical linear model (HLM). Based on literature reviews, 13 research variables (1 student gender, 2 student academic year, 3 student major, 4 student reason for taking the course, 5 student level of course participation, 6 student expected grade, 7 student achieved grade, 8 faculty gender, 9 faculty age, 10 faculty status, 11 academic field of the course, 12 class size, and 13 course type) were selected and specified as fixed effects in the analysis model. Data were 57,216 ratings of 1,481 undergraduate liberal arts courses at Seoul National University in 2006. The result of unconditional model analysis revealed that student characteristics’ effects on student satisfaction in courses (within-course effects, 88.1%) were much larger than course characteristics’ effects (between-courses effects, 11.9%). The result of conditional model analysis specifying student and course level predictors revealed that those 12 research variables, with the exception of student gender, had statistically significant effects on student satisfaction in courses. The explained variance was 22.0% in student level, 65.8% in course level, and 27.2% of the total variance. An earlier version of this paper was presented at the 8^th International Conference on Education Research, Seoul National University, Seoul, Korea, October 23–25, 2007.     

Relationships between student characteristics and self‐, peer and tutor evaluations of oral presentations

There are many influences on how assessors grade themselves and others. Oral presentations are useful for exploring such factors in peer, self‐ and tutor marked assessments, being rapidly completed and assessed, commonly used in HE and very difficult to anonymize. This opportunistic study examined the effects of gender and level of attainment on the triangulation of marks awarded to student presenters. Grades generated by peer assessment were associated more strongly with tutor‐awarded marks than those from self‐assessment. For self‐assessment there was a strong effect of gender (female students undervalued their performance compared with tutor grades). Peer assessment produced higher marks than from tutors, perhaps because of the close‐knit community developed during residential courses. For tutor marks, the greatest variability was at the lower end of the scale, whereas peer assessors were most variable when marking students who self‐evaluated or peer assessed highly. Students awarded a narrower range of marks to peers compared with tutors, but when self‐assessing used a larger range. Presentations by students who admitted to little sleep the night before received lower grades from both peers and tutors, but this was not reflected by self‐assessments, suggesting they were unaware of their poorer performances. Sessions with fewer talks (four rather than seven) reduced the ‘dip’ in marks previously observed in the middle of sessions. Findings are discussed in the context of bias in this mode of assessment.     

A study of the relationships between faculty characteristics,subject field,and course grading patterns

This paper describes an analysis of faculty grading patterns at a large public university. It introduces a methodology to both normalize and stabilize grade data by courses. Using this method, we report on grading patterns for over 1,000 teachers, giving more than 40,000 grades in approximately 2,000 courses. The findings indicate that the academic field of the course is strongly related to the types of grades assigned. Courses emphasizing quantitative and factual learning tend to have assigned lower grades. The higher grades are found in career-oriented courses, such as teacher training. This was shown in separate analyses for undergraduate and graduate courses. For undergraduate courses, lower grades were found in freshman and sophomore courses as compared with junior and senior courses. The academic credentials and personal characteristics of the teachers were only moderately related to grades. Temporary teachers had higher grades in their courses, but such personal characteristics as sex, marital, and minority status were not systematically related to grades given. There was a tendency in undergraduate courses that as class size increased, the proportion of higher grades decreased.     

“品德与公民”课程在多大程度上影响学生的国家认同？——基于澳门中小学生的实证调查

自编国家认同、思政类课程感知问卷,采用整群抽样法对澳门中小学生进行调查。研究发现,澳门中小学生国家认同整体水平中等偏上,情感领域略高于认知和行为领域;其因子在性别、年级、父亲国籍、父母亲文化程度变量上差异显著,尤以年级差异为甚。大多数学生能较好感知"品德与公民"课程目标、课程内容,但教学方式和评价方式缺乏多元性;课程感知的因子在性别、年级、母亲国籍、母亲文化程度变量上存在显著差异,年级差异最为显著。国家认同和"品德与公民"课程感知联系紧密,课程目标感知程度是影响国家认同的关键因素,但教师的教学方式和评价方式影响甚微。建议关注国家认同年级差异,低年级夯实基础,高年级查漏补缺;家庭教育是落实国家认同教育的关键一环,开设"家长课程"注重母亲教育势在必行;"品德与公民"是落实立德树人根本任务的关键课程之一,要完善基本学力要求文件以指导教师优化教学方式和评价方式。     

Why go modular? A review of modular A‐level Mathematics

Modular A‐level courses, indeed modular forms of assessment in many areas of education, have gained in popularity over the past few years. This research looked at data collected as part of the ALIS (A‐level Information System) project. In particular, the attitudes, future academic intentions and attainment of students gaining a grade in A‐level Mathematics were compared, by gender, between those who followed a modular course and those who were assessed at the end of two years of study. The two groups were found to have a similar average GCSE grade, but overall, the final A‐level grades of those assessed modularly were half a grade higher than their linearly assessed contemporaries. Possible reasons for these differences are discussed. The types of teaching style employed in modular and non‐modular A‐level courses were compared. Initial findings suggest that modularly assessed courses are relying even more on a didactic approach.     

Student Grades and Average Ratings of Instructional Quality: The Need for Adjustment

Abstract

The authors compared the average grades given in 165 behavioral and social science courses with the average ratings given by students to the instructors who taught the courses. Significant positive correlations were found between the average ratings for instructional quality and the average grades received by students. The courses in which the average grades were the highest were also those in which students gave teachers the highest ratings. Among possible reasons for the correlations are that better teachers attracted better students or that quality teachers provided more effective instruction, resulting in more student learning and, thus, higher average grades. Another explanation is that most college students tend to bias their ratings of instructional quality in favor of teachers who grade leniently (I. Neath, 1996). If correct, the latter reasoning begins to explain why the widespread use of student evaluations in the United States in recent decades has been accompanied by increases in the average grades that university students received. To prevent grade inflation, and particularly to avoid rewarding and promoting instructors who use increasingly lax grading standards, administrators should adjust student ratings of instructional quality for the average grades given for a course. In general, only courses near the extremely high and low ends in terms of students' average grades were significantly affected by the statistical adjustment.     

The Grass Isn’t Always Greener: Perceptions of and Performance on Open-Note Exams

Undergraduate biology education is often viewed as being focused on memorization rather than development of students’ critical-thinking abilities. We speculated that open-note testing would be an easily implemented change that would emphasize higher-order thinking. As open-note testing is not commonly used in the biological sciences and the literature on its effects in biology education is sparse, we performed a comprehensive analysis of this intervention on a primary literature–based exam across three large-enrollment laboratory courses. Although students believed open-note testing would impact exam scores, we found no effect on performance, either overall or on questions of nearly all Bloom’s levels. Open-note testing also produced no advantage when examined under a variety of parameters, including research experience, grade point average, course grade, prior exposure to primary literature–focused laboratory courses, or gender. Interestingly, we did observe small differences in open- and closed-note exam performance and perception for students who experienced open-note exams for an entire quarter. This implies that student preparation or in-test behavior can be altered by exposure to open-note testing conditions in a single course and that ­increased experience may be necessary to truly understand the impact of this intervention.