Development And Validation Of Parent Action Measurement Model

Loading...
Thumbnail Image
Date
2012-03
Authors
Koh, William Saw Yen
Journal Title
Journal ISSN
Volume Title
Publisher
Universiti Sains Malaysia
Abstract
The present research is aimed at developing and validating an instrument (SPAQ) to measure the impact of parents’ actions on students’ academic achievement and discipline in Malaysia. A survey using SPAQ was administered to 3425 secondary school students of various races from 24 schools in Kedah, Penang and Perak. This study was conducted in two stages. The processes in the development stage of the “Student’s Parents-Actions Questionnaire” (SPAQ) included the formulation and operationalisation of constructs, formulation and refinement of items and translation of the instrument. The content validity and construct validity were established. The model factor structure and SPAQ measurement model invariance were validated across the Malay, Chinese and Indian ethnic groups. The second part of this study explored the impact between the various constructs of parents’ actions on students’ academic achievement and discipline by utilising the structural model. SPAQ exhibited high reliability with Cronbach’s Alpha, α = 0.93 and coefficient rho, ρ = 0.94. It also exhibited sufficient content validity and construct validity. Content validity was evaluated from the average rating for each item by the experts. Item loadings with λ > 0.60 and acceptable values of goodness-of-fit indices such as AGFI, RMR, IFI, TLI, PNFI, CFI and RMSEA indicated convergent validity. The correlation coefficient between constructs, r < 0.85 indicated the existence of divergent validity. The multidimensional factor structure was established using model-data fit indices such as χ²/df , CFI, TLI, SRMR and RMSEA from Structural Equation Modelling (SEM) analysis. All the values of the indices fulfilled the stipulated criteria with χ²/df < 5.00, CFI > 0.90, TLI > 0.90, SRMR < 0.10 and RMSEA < 0.05. SPAQ has also fulfilled the various measurement invariance tests as indicated by the minimal change in CFI index, ΔCFI ≤ 0.01. The structural model for academic achievement and discipline also indicated good model-data fit. This was indicated by the various indices such as CFI, NNFI 0.90 and RMSEA < 0.05 for the three ethnic groups. The findings showed that certain constructs of parents’ actions had contributed significantly to the academic achievement and discipline of particular ethnic groups. The COND (conduciveness) construct contributed positively and the HWK (homework) construct contributed negatively towards academic achievement. Both constructs contributed significantly in each ethnic group. Parents’ aspiration (ASP) was the sole construct which contributed significantly to the Malay’s academic achievement whereas the CONF (conflict) construct was the only construct which contributed negatively and significantly to the Indian’s academic achievement. All the eight constructs did not have any significant contribution to discipline for the Chinese and Indian ethnic groups. Only five constructs were significant in the Malay ethnic group. The aspiration (ASP), conduciveness (COND) and warmth (WARM) constructs had contributed negatively whereas parental involvement in homework (HWK) and religiousity (REL) had positively contributed to the child’s discipline. As such, it is hoped that parents can exploit the findings to reduce their children’s discipline problems and increase their academic achievement. The Education Ministry and schools may utilise these findings for the benefit of all.
Description
Keywords
Parents’ actions on students’ academic achievement , and discipline in Malaysia
Citation