Multinomial Logistic Regression Model to Identify the Factors Associated with Academic Performance of Hearing Impaired Students of Some Selected Districts of Nepal

  • Tara Devi Rijal Tribhuvan University, Kirtipur, Kathmandu, Nepal
  • Gauri Shrestha Tribhuvan University, Kirtipur, Kathmandu, Nepal
Keywords: Academic performance, chi-square test, hearing impaired students, multinomial logistic regression, odds ratio, two-stage stratified random sampling

Abstract

Background: Education for hearing impaired (HI) students is always challenging for schools, teachers, parents and even for students themselves. For their education, government has established special schools and included them in mainstream school with hearing peers also called integrated school. Although all these efforts for their education, their academic performance is affected by different factors.

Objective: This study is focused to assess factors which affects the academic performance of HI students of Jhapa, Morang and Sunsari districts.

Materials and Methods: This study is a cross-sectional study and based on primary data collected through structured questionnaire. Academic Performance is categorized as Below Average, Average and Good. A sample of 238 HI students from eleven schools of three districts were selected using two stage stratified random sampling methods. Sixty teachers were also chosen for assessing some variables. Data were analyzed using descriptive and inferential statistical methods. To determine the significant factors influencing the academic performance of HI students, Multinomial Logistic Regression (MNLR) model was used.

Results: From the fitted MNLR model, variables like attendance [Odds Ratio (OR) =0.951], type of school (OR=27.39), level of study (OR=4.551), additional handicapping condition (OR=5.202), communication capacity of students (OR=9.477) and instructional material (OR=3.028) are found to be significant in the model Below Average versus Average level of academic performance of HI students. Similarly, the variables like type of school (OR=7.875 and 9.572), level of study (OR= 0.556), purpose of parents' visit (OR= 0.410) and instructional material (OR= 0.304) are found to be significant in the model Good versus Average level of Academic performance of HI students.

Conclusion: The results of the study illustrates that several factors are associated with academic performance of HI students. Concerned authorities are suggested to provide suitable infrastructure, HI students friendly curriculum, special training to teacher and awareness to family, society and hearing peers to improve the academic performance of HI students.

Downloads

Download data is not yet available.
Abstract
411
PDF
142

Author Biographies

Tara Devi Rijal, Tribhuvan University, Kirtipur, Kathmandu, Nepal

Central Department of Statistics

Gauri Shrestha, Tribhuvan University, Kirtipur, Kathmandu, Nepal

Central Department of Statistics

Published
2019-09-16
How to Cite
Rijal, T., & Shrestha, G. (2019). Multinomial Logistic Regression Model to Identify the Factors Associated with Academic Performance of Hearing Impaired Students of Some Selected Districts of Nepal. Nepalese Journal of Statistics, 3, 41-56. https://doi.org/10.3126/njs.v3i0.25577
Section
Research Articles