Construct validity and reliability of Malay Language-perception towards smoking questionnaire (BM-PTSQ) among secondary school adolescents

Abstract

Introduction: Multitude studies have shown perception is an integral factor associated with smoking, However, no such tool was available in Malay language. In this study, we established a Bahasa Malaysia version of PTSQ (BM-PTSQ) and tested the validity and reliability among secondary school adolescents. Methods: The English version of PTSQ originally consists of 12 items. It was translated into Bahasa Malaysia and back-translated again into English to check for consistency. After face validity (face-to-face query) was determined among 20 secondary school adolescents, only 10 items were included in the survey. Construct validity was established from 407 school adolescents through random selection in the same locality. More than 60% of the respondents were female, a majority (67.3%) were schooling in rural areas. Then, the reliability of the questionnaire was determined with Cronbach’s alpha. Results: EFA has grouped PTSQ into two components, they are associated with either knowledge or attitude towards smoking. The variance and Cronbach’s alpha for the first and second component were 38.24% and 0.861 (7 items) and 21.62% and 0.661 (3 items) respectively. Conclusion: The PTSQ showed good validity and reliability for measurement of perception in smoking among school adolescents in Malaysia, thus this is a viable measurement tool. More importantly this study shows an urgent need to improve the smoking education among adolescents in Malaysia.

Validity and reliability of the Malay language Perception Towards Smoking Questionnaire (BM-PTSQ) among secondary school adolescents: Further validation using Confirmatory Factor Analysis

“…In contrast, the second domain consisted of three items, which explained 27.9% of the variance. This finding is consistent with the results reported by Lim et al 18 in a study conducted in the same locality. The explained variance in the initial questionnaire exceeded the required minimum of 50% 30 , and the CFA findings supported the results of parallel analysis and EFA.…”

“…In this study, the items in domain 1 and domain 2 are almost identical to those in Lim et al 18 , and the amount of explained variance in both studies is almost the same. However, the present study’s internal consistency for domain 1 was lower than that of Lim et al 18 (Cronbach alpha value of 0.861 versus 0.800).…”

“…Even if the best option has been found from EFA that considers one or more qualitative or quantitative factor(s), the resulting instrument is, at best, a potential measurement tool, and further and more thorough examination of the instrument is still required. Hence, in this study, we performed quantitative content validation and CFA analysis of the BM-PTSQ to confirm the findings of Lim et al 18 from EFA and to affirm this instrument’s validity and reliability among Malaysian secondary school adolescents.…”

“…The BM-PTSQ scale in this study was adopted from Lim et al 18 . The sample size for the pilot test was the estimated sample size required to compute RMSEA for validation of the DBI.…”

“…Lim et al 18 investigated the perception of smoking among teenagers in Malaysia using a 10-item Malay language Perception Towards Smoking Questionnaire (BM-PTSQ) adapted from Grace et al 19 in which each item was measured on a 5-point Likert scale (strongly disagree, disagree, agree, and strongly agree). Explanatory Factor Analysis (EFA) of the BM-PTSQ yielded a two-domain solution with satisfactory internal consistency reliability.…”