Hierarchical linear models in education sciences: an application (original) (raw)
Related papers
An Application of Hierarchical Linear Modeling: OKS-2006 Science Test Achievement
Recent research has commonly used hierarchical linear models (HLMs). Also known as multilevel models, HLMs can be used to analyze a variety of questions with either categorical or continuous dependent variables. With hierarchical linear models, each level (e.g., student, classroom, and school) is formally represented by its own submodel. This study presents detailed descriptions of practical procedures to conduct nested data analysis using HLM.
1995
School effectiveness was studied in New Zealand schools, concentrating on Type A effects (given average background characteristics, how well would a student perform in any particular school?) and Type B effects (given similar student populations, are some schools more effective in achieving specified results than others?). Hierarchical linear modeling was used on data for 37 schools and 5,391 students to establish that 18.2% of total variance in mathematics, 14.37. in science, and 14.67, in English was related to the characteristics of the schools rather than inter-individual variability within the population sampled. Some schools were significantly more effective than others in that school means for the School Certificate examinations varied significantly between schools. However, a ronsiderable proportion (around 70%) of the between-school variance in all three subject areas was due to the initial ability and social and ethnic characteristics of the student populations. This Type A effect varied. Controlling for these factors (to produce a better Type B estimate) shrank the differences between schools and changed the rank order of effective schools from that based on raw scores alone. (Contains 6 figures, 10 tables, and 19 references.
We propose a multivariate multilevel model for the analysis of the Italian sample of the TIMSS & PIRLS 2011 Combined International Database on 4th grade students. The multivariate model jointly considers educational achievement on Reading, Mathematics and Science, thus allowing us to test for differential effects of the covariates on the three scores. The multilevel approach allows us to disentangle student and contextual factors affecting achievement, also considering territorial differences in wealth. The proportion of variability at class level is relevant even after controlling for the observed factors. The residual analysis allows us to locate classes with extremely high or low effectiveness.
An analysis of Turkey’s Pisa 2015 results using two-level hierarchical linear modelling
In the field of education, most of the data collected are multi-level structured. Cities, city based schools, school based classes and finally students in the classrooms constitute a hierarchical structure. Hierarchical linear models give more accurate results compared to standard models when the data set has a structure going far as individuals, groups of individuals and communities of groups. In this study, the effects of the school level indicators on overall reading skills performance of 15 year-old group students within PISA 2015 Turkey application, are analyzed by using a two level hierarchical linear model. In the study, socioeconomic indicators of family, education level of the parents and some student level indicators regarding reading skills, school level indicators such as school type, number of students and number of teachers are also integrated to modelling process to reflecting the hierarchical data structure into the statistical model. At the end of the modelling process, factors that effects the reading skills are determined.
Application of HLM to data with multilevel structure
Discussiones Mathematicae Probability and Statistics, 2011
Many data sets analyzed in human and social sciences have a multilevel or hierarchical structure. By hierarchy we mean that units of a certain level (also referred micro units) are grouped into, or nested within, higher level (or macro) units. In these cases, the units within a cluster tend to be more different than units from other clusters, i.e., they are correlated. Thus, unlike in the classical setting where there exists a single source of variation between observational units, the heterogeneity between clusters introduces an additional source of variation and complicates the analysis. Collecting data on Educational Research often does not follow the principles of simple random sample, suspected by classical regression, but rather a sample by nested clusters. Selected to students and also the contextual units to which they belong such as classes, courses, schools, neighborhoods or regions, and so forth. Using classical regression bias is produced in the typical error of measurement and an increased likelihood of committing errors of inference.
2011
The quality of the educational system is crucial to the most diverse socioeconomic and cultural aspects of a country. The objective of this paper is highlighting the importance of quality of education to explain the wide income disparities between Brazilian people. In face of such evidence, and considering the need to better understand the factors associated with education quality and efficiency, this study aims to investigate which factors are the most significant to determine elementary school students' performance in capital and small towns. For this purpose, it will be used the Brazilian Educational Search (SAEB) for 4th grades of elementary education. The estimation method is the hierarchical regression, which, in this work, will be structured into 2 levels, respectively: students, and school. The results suggest, after socioeconomic control, the positive correlation of principal's characteristic; class-time, kindergarten and presence of library to a better performance. In opposite way, the influence of professor's turnover seems to be quite harmful to determine the student´s score in capital city and small towns respectively. These results can serve as aid to public policies aimed at equity and educational qualification.
AIP Conference Proceedings, 2017
Individual and environment are a hierarchical structure consist of units grouped at different levels. Hierarchical data structures are analyzed based on several levels, with the lowest level nested in the highest level. This modeling is commonly call multilevel modeling. Multilevel modeling is widely used in education research, for example, the average score of National Examination (UN). While in Indonesia UN for high school student is divided into natural science and social science. The purpose of this research is to develop multilevel and panel data modeling using linear mixed model on educational data. The first step is data exploration and identification relationships between independent and dependent variable by checking correlation coefficient and variance inflation factor (VIF). Furthermore, we use a simple model approach with highest level of the hierarchy (level-2) is regency/city while school is the lowest of hierarchy (level-1). The best model was determined by comparing goodness-of-fit and checking assumption from residual plots and predictions for each model. Our finding that for natural science and social science, the regression with random effects of regency/city and fixed effects of the time i.e multilevel model has better performance than the linear mixed model in explaining the variability of the dependent variable, which is the average scores of UN.