John W. Young

A Family Literacy Program Connecting School and Home: Effects on Attitude, Motivation, and Literacy Achievement.

This study connected home and school literacy contexts by involving parents in developmentally appropriate and culturally sensitive literacy activities with their children. The purpose of the program was to enhance childrens achievement and interest in literacy. The family program was similar to a literature-based school program that included classroom literacy centers, teacher-modeled literature activities, and literacy center time. Meetings were held monthly, with parents, teachers, and children all working together. The program was in an inner-city school district including African American and Latino families. There were 56 children in 1st through 3rd grades (28 in the experimental group and 28 in the control). Pre- and posttest data determined achievement and motivation differences favoring the children in the family program.

Differential Prediction of College Grades by Gender and by Ethnicity: A Replication Study

Differential predictive validity in forecasting academic performance in college has been observed for a number of years. A common finding is that the use of a single regression equation generally under predicts the grades of women and over predicts the grades of minority students. This study was undertaken to determine whether differential prediction was evident for a recent cohort of students at a state university. The results confirm that the phenomenon still exists. For women, but not for minority students, the difference in predictive validity appears to be related to the effects of course selection.

Sex Differences on the SAT: An Analysis of Demographic and Educational Variables.

In comparing the average scores for men and women on the Scholastic Aptitude Test (SAT, since revised and renamed as the Scholastic Assessment Test), sex differences in demographic and educational variables that influence test performance are usually ignored. In this study of 69,284 high school seniors (12th graders) who took the SAT in November of 1990, self-reported background information was used to compute adjusted scores for men and for women. On the Verbal section, the difference in observed means was 4.68 points while the difference in adjusted means was 9.87 points. On the Mathematical section, the difference in observed means was 45.38 points but the difference in adjusted means was 33.76 points (a reduction of over 25%). In all comparisons, the mean was higher for men than for women. It is argued that adjusted means may provide more appropriate comparisons of the performance of men and women on the SAT depending on the type of comparison to be made.

Technology and Early Literacy: The Impact of an Integrated Learning System on High-Risk Kindergartners'Achievement

Two hundred sixty-five kindergarten children from a high-risk community participated in an 10-month evaluation of the effectiveness of the Waterford Early Reading Program, a software program designed to facilitate early literacy development. One hundred fifty-one students in eight experimental classrooms used the program for approximately 15 minutes per day. One hundred fourteen students in seven nonintervention classrooms had varying amounts of access to older hardware and software that was not systematically utilized by their teachers. Students were individually pre-and posttested using the TERA-2, the Lindamood Auditory Conceptualization Test, and the Waterford Inventory. Results indicated that students in the experimental classrooms performed significantly better than nonintervention students on the TERA-2 and the Waterford Inventory. Students in the experimental classrooms also showed a trend to outperform nonintervention students on the Lindamood, although not to a significant degree.

THE IMPACT OF AN SES-BASED MODEL ON A COLLEGE'S UNDERGRADUATE ADMISSIONS OUTCOMES*

In Grutter vs. Bollinger, the U.S. Supreme Court ruled that the use of race as one factor, among many in admissions decisions is constitutional. It is not known, however, whether future legal opinions will continue to uphold the use of affirmative action policies. Some have argued that class-based preferences can achieve many of the same goals as in affirmative action while being more likely to withstand legal challenges. To date, no empirical studies have been conducted on the potential impact of a class-based admissions policy if implemented at an undergraduate institution. This paper reports on a study at a selective public college and compares a number of outcomes under three admissions models: the original admissions decisions, a purely academic model, and an socio-economic status (SES)-based model. The findings showed that use of the SES-based model would have led to a more academically qualified class than in the original admitted class while maintaining substantially greater student diversity that was found under the academic model. An admissions policy based on preferences for socio-economically disadvantaged applicants appears to hold promise for other colleges and universities with similar institutional and applicant characteristics. The ideas and research design reported in this paper are based on the doctoral dissertation study of the second author, “Undergraduate Admissions Models Incorporating Socioeconomic Factors” (Johnson, 2000).

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.

A Collaborative Family Literacy Program: The Effects on Children's Motivation and Literacy Achievement.

The purpose of the study was to bridge home and school literacy contexts by involving families in literacy activities with their children to enhance childrens achievement and interest in reading and writing. The literature‐based school program included, literacy centers in classrooms, teacher modelled literature activities, and writing and reading appreciation periods called WRAP Time when children worked on literacy activities in a social context. The home program had similar features with parents and children engaging in storybook reading, recording very own words from the environment, writing journals, storytelling and the use of Highlights for Childrera magazine. Monthly meetings with parents and children allowed for, sharing ideas, finding out what parents and children wanted to learn, and the opportunity for families to work together. The program was carried out in an urban school district which included African‐American and Latino families of children in the first grade. Preand posttests were admi...

Predicting College Grades: The Value of Achievement Goals in Supplementing Ability Measures.

Achievement goal theory is an important theoretical framework for understanding achievement motivation. In previous studies, a mastery orientation has been shown to be related to students’ interest, while a performance orientation has been found to be predictive of academic performance outcomes such as course grades. In this study, the two mastery sub‐scores from the Multiple Goals Theory Measure (MGTM) of academic motivation, which was developed specifically from achievement goal theory, was found to be predictive of college grades for a sample of 257 undergraduates at a public university in the north‐eastern United States. Additionally, the results support a trichotomous model of achievement orientations comprising mastery approach, performance approach, and avoidance. The MGTM appears to hold promise as a diagnostic tool, but additional research is required on its resistance to faking and other threats to validity.

A Comparison of Two Adjustment Methods for Improving the Prediction of Law School Grades

In this study, two statistical approaches for adjusting grades were tested on data obtained from four American law schools. These approaches were previously validated using data on undergraduates but have not been used in a study of postgraduate performance. Neither method yielded consistent improvements in the predictive validity of Law School Admission Test (LSAT) scores and undergraduate grades. The single exception was for School D, where a significant improvement in the correlation of test scores with law school grades was observed. It is hypothesized that the greater variability of LSAT scores in School D was the main reason for this result.