University of Minnesota
Driven to Discover



 

In a time of economic uncertainty, recession, and pessimism, colleges and universities still need to engage and empower students in constructing their career futures. Using a pre/post-test design including a comparison group, we explored changes in cognitive and affective engagement in school, self-defeating career thoughts, and career decision self-efficacy for students who were enrolled in a career course and students who were not. Including a sample of 76 college students, we explored constructivist career coursework as a mechanism of engagement and empowerment. Suggesting the potential of a constructivist career course to empower but not necessarily engage students, we found a trend toward increased empowerment for those in the career course with no change in students’ cognitive and affective engagement in school.
Author(s)/Presenter(s): Zoila M. Ganuza M.A., Tabitha Grier Ph.D., L.P.

Over the past two decades students’ engagement in school and learning has emerged as the “bottom line” issue across hundreds of dropout prevention and recovery programs in the U.S. Schools, districts, and states need to implement scientifically-based models that keep youth engaged in school and on the road to graduation. Check & Connect is one such model. One of 22 dropout prevention interventions rated by the U.S. Department of Education’s What Works Clearinghouse, Check & Connect is the only program found to have strong evidence of positive effects for staying in school.
Check & Connect is based on the following assumptions: (1) Leaving school prior to graduation is not an instantaneous event; (2) solving the dropout problem will require a multifaceted effort of home, school, community, and youth; (3) students must be empowered to take control of their own behavior; and (4) schools must be designed to reach out to families in partnership with community. The Check & Connect intervention has two main components: The Check component involves continually assessing student engagement through close monitoring of student performance and progress indicators (e.g., attendance, suspension, grades, credit accrual); and the Connect component involves a mentor intervening in a timely fashion in partnership with school personnel, family members, and community service providers based on individual student performance information obtained through the Check component
Author(s)/Presenter(s): Sandra Christenson, professor; Karen Stout, research associate; and Ann Mavis, coordintor.

In computerized adaptive test (CAT), many termination criteria have been proposed such as fixed-length, pre-specified standard error of measurement (SEM), information function, mastery/classification and administering all items criteria. Each of them has at least an advantage in a specific area. We have studied the first two different termination criteria, since those methods may not be compatible with each other but comparable in terms of efficiency of CAT. For example, the smaller SEM requires larger NOI and vice verse. Examiners who make a decision of these criteria often face a difficulty in making decision on the selection.
Through the Post-hoc Simulation (POSTSIM) program, we have researched the efficiency between these criteria. The efficiency was measured by the root mean square difference (RMSD). Instead of providing a way of selection, we expose results from the comparison. Our study has also been expanded with Classification Consistency (CC), which is more practical issue, for instance, with standardized achievement test. Applied were real data that consisted of 2889 students on 80 mathematics test items. Furthermore, we propose new criterion combining these two criteria as new criterion. In this study, we applied three unidimensional item response models such as one-, two- and three-parameter logistic models.
Author(s)/Presenter(s): NamKeol Kim and Jihoon Ryoo, graduate students

This study investigated the trajectory of internalizing behaviors as well as the effects of gender and perceived academic achievement on changes in internalizing behaviors. It did so through latent growth curve modeling procedures. The participants were adolescents, taken from a sample of 4th ~ 7th graders from the Korean Youth Panel Survey. There were three significant findings. First, there was evidence of both intraindividual and interindividual variability in internalizing behaviors between the 4th and 7th grades. On average, early adolescents’ internalizing behaviors increased over time. Second, there were gender differences for the changes in internalizing behaviors. On average, girls’ internalizing behaviors increased more rapidly over time than those of boys. Finally, adolescents with higher levels of perceived academic achievement at Grade 4 had more rapid increases in internalizing behaviors between the 4th and 7th grades when compared to peers with lower levels of perceived academic achievement. The limitations of the study are noted, and suggestions are made for future research.
Author(s)/Presenter(s): Mijung Seo, Visiting Scholar in Educational Psychology

College of Education & Human Development
cehd@umn.edu | 612-626-9252 | 104 Burton Hall, 178 Pillsbury Dr SE, Minneapolis, MN 55455

© 2009 Regents of the University of Minnesota. All rights reserved.
The University of Minnesota is an equal opportunity educator and employer.