Research Article: Psychometric properties of the Malay version of the self-efficacy for exercise scale

Date Published: May 3, 2019

Publisher: Public Library of Science

Author(s): Abdulwali Sabo, Yee Cheng Kueh, Garry Kuan, Dalton Müller Pessôa Filho.


The present study was aimed at validating the Malay-language version of the Self-Efficacy for Exercise Scale (SEE-M) using confirmatory factor analysis (CFA).

Data were collected from undergraduate students at all campuses of the Universiti Sains Malaysia. A total of 1,605 students completed the SEE-M (female: 71.5%, male: 28.5%), with the mean age of 20.3 years (SD = 1.5). Perceived self-efficacy was assessed with the 18-item SEE-M. Standard forward–backward translation was performed to translate the English version of the Efficacy for Exercise Scale (SEE) into the Malay version (SEE-M).

The 2 initial measurement models tested (1-factor and 3-factor models) did not result in a good fit to the data. Subsequent investigation of the CFA results recommended some modifications, including adding correlations between the item residuals within the same latent variable. These modifications resulted in good fit indices for the 1-factor model (RMSEA = .059, CFI = .939, TLI = .922, SRMR = .049) and the 3-factor model (RMSEA = .066, CFI = .924, TLI = .903, SRMR = .051). The final measurement models comprised all 18 SEE-M items, which had significant factor loadings of more than .40. The test-retest results indicated that the SEE-M was stable, with an intra-class correlation of .99. The composite reliability was .886 for the 1-factor model and .670–.854 for the 3-factor model.

The translated version of the SEE-M was valid and reliable for assessing the level of self-efficacy for exercise among university students in Malaysia.

This study examining the psychometric properties of the SEE scale based on CFA was the first to assess 2 proposed models (1-factor and 3-factor models) simultaneously and to translate the original, English-language SEE into Malay.

Partial Text

The primary, dynamic approaches that can be adopted to lower the risk of various chronic diseases (e.g., cardiovascular diseases, non-insulin-dependent diabetes mellitus, osteoporosis, obesity, and some cancers) can be achieved through regular participation in physical activity [1]. Regular exercise is an essential component of an effective, health-promoting lifestyle. For example, a 15-year prospective study verified that recreational physical activity is an independent predictor of a reduced cardiovascular mortality rate among adults. After adjusting for other risk factors (Framingham risk score and central obesity), the high recreational physical activity group was 35% less likely than the low recreational physical activity group to have cardiovascular mortality outcomes [2].

Mplus version 8 was used to analyze the CFA results. The data were pre-screened, and those questionnaires with missing values were excluded from the analysis. The final data analysis included 1,605 completed questionnaires. The MLR Estimator was selected to perform CFA because it is robust to non-normality distribution of data and produces estimates with standard errors, including a mean adjusted chi-square statistic [28].

In this study, we translated the English version of the SEE into Malay and then confirmed the questionnaire’s psychometric properties to provide a tool to measure individuals’ levels of self-efficacy for participation in physical activity among Malay-speaking populations. Perceived self-efficacy is an essential behavior in the causal mechanisms that influence the health promotion achieved through lifestyle modifications [5]. Consequently, the SEE items should accurately reflect the self-efficacy construct and be tested with all ethnic and racial populations.