Evaluation 2008 Banner

Return to search form  

Contact emails are provided for one-to-one contact only and may not be used for mass emailing or group solicitations.

Session Title: Longitudinal and Growth Curve Analysis
Multipaper Session 603 to be held in Centennial Section B on Friday, Nov 7, 1:35 PM to 3:05 PM
Sponsored by the Quantitative Methods: Theory and Design TIG
Chair(s):
Patrick McKnight,  George Mason University,  pmcknigh@gmu.edu
Discussant(s):
Frederick Newman,  Florida International University,  newmanf@fiu.edu
Evaluating Mental Health Recovery: A Latent Growth Curve Modeling Approach
Presenter(s):
Kathryn DeRoche,  Mental Health Center of Denver,  kathryn.deroche@mhcd.org
Antonio Olmos-Gallo,  Mental Health Center of Denver,  antonio.olmos@mhcd.org
Christopher McKinney,  Mental Health Center of Denver,  christopher.mckinney@mhcd.org
Abstract: The field of adult mental health has been evolving in the last two decades to focus on consumer-centered mental health recovery. The current study evaluated change across time among two measures of mental health recovery, through the use of a multivariate latent growth curve model. The presentation will discuss the influence of moderators of recovery including: level of services being received, characteristics of staff members that promote recovery, and the consumers’ level of daily functioning. The clinical implications regarding the initial level of recovery, the rate of change in recovery across time, and the potential moderators of change for community-based mental health centers and their consumers will be discussed. In addition, the benefits of applying latent growth curve modeling techniques for evaluating change in the social and behavioral science disciplines will be highlighted.
A Longitudinal Examination of the Academic Year and Summer Learning Rates of Full and Half-Day Kindergartners
Presenter(s):
Keith Zvoch,  University of Oregon,  kzvoch@uoregon.edu
Joseph Stevens,  University of Oregon,  stevensj@uoregon.edu
Abstract: Literacy data collected over the course of two academic years were used to estimate the rate at which full and half-day kindergartners acquired literacy skills during kindergarten, first grade, and the intervening summer. Application of piecewise growth models to the time series data obtained on students from a large southwestern school district revealed that economically disadvantaged full-day kindergartners gained literacy skills at a faster rate than their more economically advantaged and initially higher scoring half-day peers during the kindergarten year. However, over the summer between kindergarten and first grade, the literacy performance of full-day dropped while their half-day peers maintained the literacy gains acquired during kindergarten. Full and half-day alumni growth rates then remained equivalent over the first grade school year. Implications for evaluating the short and long term efficacy of school-based initiatives like full-day kindergarten and more generally, the effectiveness of schools and schooling are discussed.
Multilevel Longitudinal Analysis of Teacher Effectiveness and Reading Fluency in Native American Students
Presenter(s):
Heather Chapman,  EndVision Research and Evaluation,  hjchapman3@gmail.com
Abstract: All students need to learn to read in order to be successful in school and in life in general. Unfortunately, many students do not learn to read at grade level by the time they finish high school. In the Native American population, the number of students reading at grade level has been reported to be as little as 26%. Many different interventions have been used to increase achievement, but the analyses used to determine success of these interventions have often not been adequate. Often these methods have been cross-sectional in nature and have failed to account for the clustering of students within classrooms and teachers within schools. The proposed paper aims to investigate the relationship between reading ability and several other factors using more advanced multilevel longitudinal methods. These methods have the potential to decrease the bias introduced into many traditional analyses, which leads to increased accuracy of results.

 Return to Evaluation 2008

Add to Custom Program