Search Results

Author: Wisnicki, Kathleen Sherlock
Resulting in 1 citation.
1. Wisnicki, Kathleen Sherlock
The Impact of Maternal Characteristics on Child Academic Achievement as Mediated by Home Environment: A Cohort-Sequential Approach to Latent Growth Modeling
Ph.D. Dissertation, University of California - Los Angeles, 1998.
Also: http://en.scientificcommons.org/6149228
Cohort(s): Children of the NLSY79, NLSY79
Publisher: UMI - University Microfilms, Bell and Howell Information and Learning
Keyword(s): Age at Birth; Age at First Birth; Children, Academic Development; Children, Home Environment; Ethnic Groups/Ethnicity; Gender; Home Observation for Measurement of Environment (HOME); Modeling, Growth Curve/Latent Trajectory Analysis; Modeling, Multilevel; Mothers, Behavior; Mothers, Education; Poverty; Welfare

Research in child development consistently demonstrates the importance of both maternal and child characteristics and their influence on children's cognitive outcomes. Utilizing data from the NLSY Mother/Child Assessment, this study employs powerful, multivariate statistical methodology to examine growth in children's achievement over time with respect to important background variables. These include both maternal and child characteristics such as Poverty Status, Ethnicity, Gender, Maternal age at birth, Maternal Education, Maternal Support, and Maternal Drinking Behavior. This data set, which consists of data for over 9000 children and their mothers, allows for examination of children's growth at four time points (1986-1992) over a wide range of child ages (5-17). The complex structure of this data set provides a great opportunity to build on previous substantive findings, while demonstrating a cohort-sequential approach to latent variable growth modeling (LGM). Specifically, the LGM framework assumes that there is variable growth modeling (LGM). Specifically, the LGM framework assumes that there is individual variation in both the initial status of an individual and individual variation in growth rate over time. In this approach, random slopes and intercepts are incorporated into the model as latent variables, and growth in achievement is based on these growth parameters as influenced by background characteristics. Further, growth in achievement may be modeled independently for each birth cohort, as in a multiple-group design, allowing for parameter estimates based on the child's age at the time of assessment. In addition, the influences of background variables that change over time, such as poverty status, are examined at each assessment and are estimated for each cohort. Finally, the impact of home environment as a mediator for child outcomes and the manner in which the influence of home environment changes over time are examined. According to the results, aassumptions regarding individual variation in growth and initial status are valid. Further, the modeling sequence clearly demonstrates the need for the inclusion of both time varying and time invariant background variables into the growth model. Although the home mediator did not produce as big an effect on the outcome measure as initially anticipated, its usefulness within the latent variable growth model is easily demonstrated. Overall, this study provided a comprehensive framework for the study of growth over time with encouraging results. The importance of including background variables and the mediator is discussed in terms of relevance to many other applications in Social Science research.
Bibliography Citation
Wisnicki, Kathleen Sherlock. The Impact of Maternal Characteristics on Child Academic Achievement as Mediated by Home Environment: A Cohort-Sequential Approach to Latent Growth Modeling. Ph.D. Dissertation, University of California - Los Angeles, 1998..