Sehee Hong

The Reciprocal Relationship Between Parental Involvement and Mathematics Achievement: Autoregressive Cross-Lagged Modeling

This study focused on comparing the longitudinal associations between two types of parental involvement (i.e., mathematics value and academic reinforcement) and high school students’ mathematics achievement, using data from the Longitudinal Study of American Youth(LSAY). Results, based on multivariate autoregressive cross-lagged modeling, indicated that parents’ academic reinforcement had no effect on students’ mathematics achievement and vice versa; however, a statistically significant positive reciprocal influence existed between parents’ mathematics value and students’ mathematics achievement throughout high school. This result not only reaffirms that parental involvement is a multidimensional construct but also implies that parental involvement has a domain-specific effect. Results from multigroup analyses revealed that students’ gender did not have a differential effect on these associations.

Longitudinal Effects of Perceived Control on Academic Achievement

ABSTRACT It is well established that perceived control plays an important role in student academic achievement, but little is known about its longitudinal stability, ethnic variation, and developmental effects on subsequent achievement during adolescence. Findings from this study indicated (a) perceived control remains stable during adolescence for each of the four major ethnic groups in the United States; (b) perceived control has a direct effect on subsequent academic achievement as well as an indirect effect, which is mediated by high school students academic engagement behaviors for all 4 ethnic groups; (c) regarding social contextual factors, students’ perception of teacher and parental support had a positive effect on perceived control, which ultimately impacted the academic achievement of high school students across all 4 ethnic groups.

Understanding Latino Children's Heterogeneous Academic Growth Trajectories: Latent Growth Mixture Modeling Approach

ABSTRACT Addressing the academic needs of a growing student population with culturally and linguistically diverse characteristics is one of the challenges facing educators. This study used data from the Early Childhood Longitudinal Study to test for differences in patterns of mathematics growth (e.g., high, middle, and low performance groups) in Latino children. Analysis through the latent growth mixture method yielded 4 distinct mathematics development profiles. Examinations into positive or negative factors related to successful mathematics achievement were also conducted. Results indicate that students in the highest performing group were associated with education programs in which language of instruction and home language were English.

Multivariate Autoregressive Cross-Lagged Modeling of the Reciprocal Longitudinal Relationship between Perceived Control and Academic Achievement

Although many studies have demonstrated positive associations between perceived control and academic achievement, few studies have actually explored which of the two constructs is the determinant of the other. There are only a few longitudinal studies on the relationship of perceived control and academic achievement. The present study examined the reciprocal longitudinal relation between perceived control and academic achievement. Further, considering the multiethnic background of the USA, this study investigated how the relationship between two constructs varies with ethnicity. Using a randomly selected sample of 1,500 students from Asian, Black, Hispanic, and White groups in the National Education Longitudinal Study data, Autoregressive Cross-lagged Modeling was performed to get a complete picture of the longitudinal relationship. Results showed a positive longitudinal effect of academic achievement on perceived control across the ethnic groups. Explanations for these findings are discussed.

Comparisons of Multilevel Modeling and Structural Equation Modeling Approaches to Actor–Partner Interdependence Model

There are basically two modeling approaches applicable to analyzing an actor–partner interdependence model: the multilevel modeling (hierarchical linear model) and the structural equation modeling. This article explains how to use these two models in analyzing an actor–partner interdependence model and how these two approaches work differently. As an empirical example, marital conflict data were used to analyze an actor–partner interdependence model. The multilevel modeling and the structural equation modeling produced virtually identical estimates for a basic model. However, the structural equation modeling approach allowed more realistic assumptions on measurement errors and factor loadings, rendering better model fit indices.

A Comparison of Mixture Modeling Approaches in Latent Class Models With External Variables Under Small Samples

The purpose of the present study is to compare performances of mixture modeling approaches (i.e., one-step approach, three-step maximum-likelihood approach, three-step BCH approach, and LTB approach) based on diverse sample size conditions. To carry out this research, two simulation studies were conducted with two different models, a latent class model with three predictor variables and a latent class model with one distal outcome variable. For the simulation, data were generated under the conditions of different sample sizes (100, 200, 300, 500, 1,000), entropy (0.6, 0.7, 0.8, 0.9), and the variance of a distal outcome (homoscedasticity, heteroscedasticity). For evaluation criteria, parameter estimates bias, standard error bias, mean squared error, and coverage were used. Results demonstrate that the three-step approaches produced more stable and better estimations than the other approaches even with a small sample size of 100. This research differs from previous studies in the sense that various models were used to compare the approaches and smaller sample size conditions were used. Furthermore, the results supporting the superiority of the three-step approaches even in poorly manipulated conditions indicate the advantage of these approaches.

A Meta-analysis on relationship between adolescents’ suicidal ideation and risk factors

본 연구는 메타분석을 통해 청소년의 자살생각과 이와 관련된 위험요인들을 체계별로 구분하여 평균 효과크기를 추정하고, 각 연구별 효과크기에 영향을 미치는 연구특성을 검증하는데 그 목적이 있다. 이를 위해 2000년부터 2014년까지 출판된 학위논문과 국내학술지논문, 연구보고서 등을 수집하여 분석을 수행하였다. 개인, 가족, 또래, 학교체계별 주요 위험요인은 각각 우울, 부모학대, 따돌림, 학업스트레스이며 연구별 효과크기의 이질성을 파악하기 위해 연구특성 변수(출판유형, 자살사고 척도 유형, 학교급, 표본크기, 출판연도)를 조절변수로 하여 메타회귀분석을 실시하였다. 더불어 출판 편의 가능성을 검토하였다. 본 연구를 통해 밝혀진 연구결과를 보면 가장 큰 효과크기를 보인 변수는 우울(1.147)이었고, 부모학대(0.676), 학업스트레스(0.570), 따돌림(0.569) 순서로 나타났으며 모두 유의하였다. 메타회귀분석 결과 우울에서는 학교급에 따른 차이를 보였고 부모학대에서는 출판연도에 따른 차이를 확인하였다. 모든 체계에서 출판 편의는 나타나지 않았다. 본 연구결과는 향후 효과적인 청소년 자살생각 예방 프로그램 개발 및 개입에 도움이 될 수 있을 것이다.