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.     

THE COMPARISON OF GPA REGRESSION EQUATIONS FOR REGULARLY ADMITTED AND DISADVANTAGED FRESHMEN AT THE UNIVERSITY OF ILLINOIS

The regression equations of first semester grade point average (GPA) on high school percentile rank (HSPR) and verbal and quantitative score on the Cooperative School and College Ability Tests (SCAT) were significantly different for men and women regularly admitted freshmen and men and women freshmen admitted to the Special Educational Opportunities Program (SEOP) at the University of Illinois. HSPR and SCAT verbal scores were useful predictors of GPA for all groups. However, separate regression equations for the prediction of GPA were indicated, since there were significant differences in the regression coefficients of all of the independent variables among the four groups.     

Assessing Differential Prediction of College Grades by Race/Ethnicity with a Multilevel Model

Previous research notes the importance of understanding racial/ethnic differential prediction of college grades across multiple institutions. Institutional variation in selection indices is especially important given some states' laws governing public institutions' admissions decisions. This paper employed multilevel moderated multiple regression to study the variation of selection indices across 30 institutions and the accuracy of selection indices in predicting college grades for students of different racial/ethnic backgrounds. Several benefits of multilevel models for cross-institutional differential prediction studies were described and include: controlling for institutional differences in range restriction, providing reliability estimates of least squares estimates, and adjusting criterion scores for differences in coursework difficulty. The findings from this study provide evidence of institutional variation in selection indices, which challenges current laws aimed at standardizing them. Specifically, there was evidence that the predictor slope coefficients varied across institutions, in addition to the estimates that measured intercept differences for African and Asian American students. Across universities, the results mirrored previous findings: high school grade point average (GPA) differentially predicted grades for African Americans, SAT verbal scores differentially predict grades for Asian Americans, and SAT math scores were better predictors of Asian Americans' grades.     

Some Impacts of Science Telecasts

Abstract

The problem investigated in this study was to determine how accurately three related, yet distinct, groups could predict first semester college freshman GPA Data on seventy-six students, parents, and counselors were analyzed.

Correlation coefficients computed between predicted GPA by students, parents, and counselors and GPA actually achieved by the students were found to be 0.5443, 0.6010, and 0.5867 respectively. An analysis of variance of .4265 demonstrated no significant differences between the three groups' predicted GPA. The differences between the students' achieved GPA and the predicted GPA by each group were all significant at the .01 level.

Each group predicted about as well as the more objective measures of academic success; each group tended to predict about the same; and each group tended to over-predict.     

How Well Do Undergraduate Research Programs Promote Engagement and Success of Students?

Assessment of undergraduate research (UR) programs using participant surveys has produced a wealth of information about design, implementation, and perceived benefits of UR programs. However, measurement of student participation university wide, and the potential contribution of research experience to student success, also require the study of extrinsic measures. In this essay, institutional data on student credit-hour generation and grade point average (GPA) from the University of Georgia are used to approach these questions. Institutional data provide a measure of annual enrollment in UR classes in diverse disciplines. This operational definition allows accurate and retrospective analysis, but does not measure all modes of engagement in UR. Cumulative GPA is proposed as a quantitative extrinsic measure of student success. Initial results show that extended participation in research for more than a single semester is correlated with an increase in GPA, even after using SAT to control for the initial ability level of the students. While the authors acknowledge that correlation does not prove causality, continued efforts to measure the impact of UR programs on student outcomes using GPA or an alternate extrinsic measure is needed for development of evidence-based programmatic recommendations.     

A comparison of white and black student academic success using noncognitive variables: A LISREL analysis

The structural relation of the seven noncognitive dimensions proposed by Sedlacek and Brooks in 1976 and traditional definitions of academic ability, as indicated by Scholastic Aptitude Test (SAT) scores, to first semester grade-point average (GPA) and persistence after three and five semesters was examined in this study. Random samples of entrants at one predominantly white state university were administered the Non-cognitive Questionnaire (NCQ) during summer orientation in 1979 and 1980. The NCQ results and the SAT scores were used to derive structural models (using LISREL) or early academic success for both black and white students. The structural models for the black and white students were found to be very different. For black students, traditional academic ability was related to first semester GPA, but neither GPA nor academic ability was related to persistence. Only the noncognitive dimensions were predictive of black student persistence. For white students, academic ability was the best predictor of first semester grades, and these grades were the major predictor of subsequent persistence. The noncognitive dimensions were not important in white student academic success, whereas they were crucial in black student academic success.     

Motivation and Self-regulation in Community College Transfer Students at a Four-year Online University

ABSTRACT

Motivation and self-regulation were examined in a sample of community college transfer students enrolled in a 4-year, online university. The relation between motivation and self-regulation and students’ performance was examined, as was the association between these learner characteristics (i.e., motivation and self-regulation) and sociodemographic factors (e.g., marital status, employment status). Motivation was found to be significantly correlated with both semester and cumulative grade point average (GPA), while associations between self-regulation and performance were more limited. Further, motivation was found to be a significant predictor of semester GPA in a model controlling for sociodemographic factors and prior achievement. Motivation and self-regulation were also found to differ according to students’ sociodemographic status. For instance, transfer students with children under 18 were found to have significantly higher levels of motivation than nonparents. Those employed full-time had lower levels of self-regulation than did their nonemployed peers. Implications for further research on community college transfer students and online learning are discussed.     

Academic Factors Related to Student Achievement in a Capstone Food Chemistry Course

The 1st step in successfully intervening with students who may fail a course is to identify them as early as possible in the semester. The objective of this study was to create a model to predict student performance in FDSC 4304, the required capstone Food Chemistry class, using academic performance in prerequisite courses as potential predictors. We analyzed data for 116 undergraduates who completed Food Chemistry (FDSC 4304) between 2008 and 2015. Data included semester of enrollment and grade earned in FDSC 4304; transfer status; grades in prerequisite classes in science, math, and statistics courses and an introductory Food Science course, FDSC 1103; and the students’ university GPA at the time of enrollment in FDSC 4304. Cumulative GPA had the strongest significant (P < 0.001) positive correlation with FDSC 4304 grade (r = 0.64), followed by grade in statistics GPA (r = 0.52), FDSC 1103 grade (r = 0.45), pre‐requisite chemistry GPA (r = 0.44), and biology GPA (r = 0.42). When using partial correlations to control for cumulative GPA, only grades in FDSC 1103 (completed by 62.9% of students) were significantly correlated with grades in FDSC 4304. Linear regression indicated cumulative GPA and FDSC 1103 grades explained 35.5% of the variance in FDSC 4304 grades. When cumulative GPA (available for 91.6% of students) alone was regressed on FDSC 4304, it explained 40.6% of the variance for the larger group. Lower cumulative GPAs and FDSC 1103 grades are suggestive but not determinative of potential student struggles in FDSC 4304. Instructors should use cumulative GPAs and introductory food science course grades (either alone or in combination) with actual early course performance measures to identify students in need of additional help.     

Reducing Inequality in Academic Success for Incoming College Students: A Randomized Trial of Growth Mindset and Belonging Interventions

Light-touch social psychological interventions have gained considerable attention for their potential to improve academic outcomes for underrepresented and/or disadvantaged students in postsecondary education. While findings from previous interventions have demonstrated positive effects for racial and ethnic minority and first-generation students in small samples, few interventions have been implemented at a larger scale with more heterogeneous student populations. To address this research gap, 7,686 students, representing more than 90% of incoming first-year students at a large Midwestern public university, were randomly assigned to an online growth mindset intervention, social belonging intervention, or a comparison group. Results suggest that after the fall semester, the growth mindset intervention significantly improved grade point averages for Latino/a students by about .40 points. This represents a 72% reduction in the GPA gap between White and Latino/a students. Further, this effect was replicated for both spring semester GPA and cumulative GPA. These findings indicate that light-touch interventions may be a minimally invasive approach to improving academic outcomes for underrepresented students. Our findings also highlight the complexity of implementing customized belonging interventions in heterogeneous contexts.     

The Influence of Several Factors on Reliability for Complex Reading Comprehension Tests

The purpose of this study was to investigate the effects of items, passages, contents, themes, and types of passages on the reliability and standard errors of measurement for complex reading comprehension tests. Seven different generalizability theory models were used in the analyses. Results indicated that generalizability coefficients estimated using multivariate models incorporating content strata and types of passages were similar in size to reliability estimates based upon a model that did not include these factors. In contrast, incorporating passages and themes within univariate generalizability theory models produced non-negligible differences in the reliability estimates. This suggested that passages and themes be taken into account when evaluating the reliability of test scores for complex reading comprehension tests.     

Are validity coefficients understated due to correctable defects in the GPA?

The predictive validity of preadmissions measures such as standardized test scores and high school grades may be understated because of correctable defects in both the freshman year and cumulative grade point average (GPA). Measurement error in the criterion artificially depresses the size of observed validity coefficients. A study was conducted using item response theory (IRT) to develop a more reliable measure of performance, called an IRT-based GPA, and tested in a predictive validity study using data from Stanford University. Results indicate increased predictability when the IRT-based GPA is compared with the usual GPA.This article is based, in part, on the doctoral dissertation of the author, which was completed at the School of Education at Stanford University.     

Relationships among properties of college students’ self-set academic goals and academic achievement

The major purpose of this study was to investigate the relationships among properties of college students’ self-set academic goals and academic achievement, using multiple theoretical perspectives. Using a personal goal-based research methodology, college students enrolled in a learning-to-learn course (N?=?130) were asked to list 20 of their goals (academic and/or non-academic). For each of their goals, goal specificity, value, expectation of success and autonomous and controlled motivation were measured and then ratings on each goal property were averaged across students’ academic goals (24.75% of all goals) to predict students’ grade point average (GPA) for the semester. Regression results suggested a positive affect on students’ semester GPA for goal specificity and a negative effect for controlled motivation; the model explained 19% of the variation in GPA. This research may help to inform motivation researchers and educational practitioners who assist college students with goal setting.     

Value-added models and the measurement of teacher productivity

Research on teacher productivity, as well as recently developed accountability systems for teachers, relies on “value-added” models to estimate the impact of teachers on student performance. We consider six value-added models that encompass most commonly estimated specifications. We test many of the central assumptions required to derive each of the value-added models from an underlying structural cumulative achievement model and reject nearly all of them. While some of the six popular models produce similar estimates, other specifications yield estimates of teacher productivity and other key parameters that are considerably different.     

Evaluation of Dimensionality in the Assessment of Internal Consistency Reliability: Coefficient Alpha and Omega Coefficients

In the lead article, Davenport, Davison, Liou, & Love demonstrate the relationship among homogeneity, internal consistency, and coefficient alpha, and also distinguish among them. These distinctions are important because too often coefficient alpha—a reliability coefficient—is interpreted as an index of homogeneity or internal consistency. We argue that factor analysis should be conducted before calculating internal consistency estimates of reliability. If factor analysis indicates the assumptions underlying coefficient alpha are met, then it can be reported as a reliability coefficient. However, to the extent that items are multidimensional, alternative internal consistency reliability coefficients should be computed based on the parameter estimates of the factor model. Assuming a bifactor model evidenced good fit, and the measure was designed to assess a single construct, omega hierarchical—the proportion of variance of the total scores due to the general factor—should be presented. Omega—the proportion of variance of the total scores due to all factors—also should be reported in that it represents a more traditional view of reliability, although it is computed within a factor analytic framework. By presenting both these coefficients and potentially other omega coefficients, the reliability results are less likely to be misinterpreted.     

Gender Bias in Predicting College Academic Performance: A New Approach Using Item Response Theory

The possibility of differential prediction of college academic performance for men and women is, at present, an issue of concern both to measurement specialists and to the general public. In prior studies, the use of a single equation to predict grade point average (GPA) from preadmissions measures has generally led to the systematic underprediction of women's college grades. In the following study, Item Response Theory (IRT) was used to develop a form of adjusted cumulative GPA, called the IRT-based GPA. Significant underprediction for women occurred using a version of the cumulative GPA as the outcome measure. In contrast, the use of the IRT-based GPA indicated no significant underprediction for men or women. A single regression equation worked best in predicting both men's and women's IRT-based GPA. In addition, the IRT-based GPA was substantially more predictable from preadmissions measures than the cumulative GPA.     

IRT‐Estimated Reliability for Tests Containing Mixed Item Formats

As a global measure of precision, item response theory (IRT) estimated reliability is derived for four coefficients (Cronbach's α, Feldt‐Raju, stratified α, and marginal reliability). Models with different underlying assumptions concerning test‐part similarity are discussed. A detailed computational example is presented for the targeted coefficients. A comparison of the IRT model‐derived coefficients is made and the impact of varying ability distributions is evaluated. The advantages of IRT‐derived reliability coefficients for problems such as automated test form assembly and vertical scaling are discussed.     

Inferences About Variance Components and Reliability-Generalizability Coefficients in the Absence of Random Sampling

Generalizability theory (G theory) employs random-effects ANOVA to estimate the variance components included in generalizability coefficients, standard errors, and other indices of precision. The ANOVA models depend on random sampling assumptions, and the variance-component estimates are likely to be sensitive to violations of these assumptions. Yet, generalizability studies do not typically sample randomly. This kind of inconsistency between assumptions in statistical models and actual data collection procedures is not uncommon in science, but it does raise fundamental questions about the substantive inferences based on the statistical analyses. This article reviews criticisms of sampling assumptions in G theory (and in reliability theory) and examines the feasibility of using representative sampling, stratification, homogeneity assumptions, and replications to address these criticisms.     

Building bridges to student learning: Perceptions of the learning environment,engagement, and learning outcomes among Chinese undergraduates

The study examined the relationships among students’ perceptions of the learning environment, prior academic achievement, engagement, and learning outcomes (cumulative university GPA, generic skills development, and learning satisfaction) with a sample of 2,616 seniors from a full-time research-oriented university in Mainland China. The results supported a model which showed that students’ perceptions of the learning environment and prior academic achievement had direct effects on learning outcomes, and indirect effects via their engagement. The effects, however, varied depending on the type of the outcome in question: (1) Student engagement mediated the relationship between course experience and generic skills development, (2) cocurricular experience positively predicted learning satisfaction, (3) first semester GPA positively predicted cumulative university GPA, and (4) National College Entrance Examination scores did not correlate with any other factor. A major proportion of the variance in the three types of learning outcomes was accounted for by the model, showing its effectiveness in predicting university students’ learning. Implications for improving undergraduate education in China are provided.     

A review of reliability procedures for measuring observer agreement

The present article reviews the procedures that have been developed for measuring the reliability of human observers' judgments when making direct observations of behavior. Measures such as the percentage of agreement, Cohen's kappa, and phi have been used to measure observer agreement; however, these coefficients have serious limitations. In addition to specifying the deficiencies that exist with these excessively used reliability measures, the present article discusses recently developed univariate and multivariate agreement measures that are based on quasi-equiprobability and quasi-independence models and estimates. Improvements in precision are provided by such models and estimates since they (1) yield a probability based coefficient of agreement with a directly interpretable meaning, (2) correct for the proportion of “chance” agreement, and (3) allow the partitioning of the agreement and disagreement estimates within the models.     

Analyzing the long-term performance of minority and nonminority students: A tale of two studies

This is a comparative longitudinal analysis of the performance of minority and nonminority students in four successive entering classes at a highly selective college. Graduation rates were comparable, but the two groups were found to be as sharply differentiated with respect to cumulative grade point average (GPA) at the end of 4 years as they were at the end of the first year. Unanticipated across-class increases in mean GPA relative to ability, greater for minority than for nonminority students, were found and their implications considered. The validity of admissions tests with respect to long-term cumulative GPA was comparable to that obtained when first-year GPA was used. Questions are raised regarding the possible effect on minority students of persistent minority-nonminority differences in academic performance, especially in settings in which the two groups are sharply differentiated on academic ability measures. Findings of a similarly designed study in a less selective setting are also considered.