• Korean Journal of Social Welfare
• /
• v.58 no.2
• /
• pp.87-118
• /
• 2006

The purpose of this paper is to test construct validity of contextual performance of social worker. The participants in this study included 170 social worker in community welfare center nationwide. A narrative data of social work practice experience and MTMM(multitriat-multimethod) data on job performance were collected. I classified meaningful contextual behavior cases by qualitative methods, and examined discriminant validity of contextual and task performance by CFA model. Data analysis showed that social workers' contextual performance has discriminant validity and sub constructs of this concept have internal consistency. However, some construct are not completely independent.

• Korean Management Science Review
• /
• v.20 no.1
• /
• pp.37-49
• /
• 2003

There is an increasing need to use mobile Internet as a research tool as wireless technology has been developing rapidly. However, little research has been conducted to confirm methodological validity of mobile Internet survey. This study examines the possibility of using mobile Internet as a survey tool by comparing survey results of mobile and stationary Internet surveys with the same questionnaire. The results were analyzed from both economic and theoretical perspectives. Both mobile and stationary Internet survey sites were implemented with supports from domestic mobile and stationary Internet portals and telecommunication companies. The results show that there exist several differences between two survey methods. First, many respondents who use mobile Internet gave up at the early stage. However, once people continued to respond, they answered the questions to the end. Second, means and standard deviations of mobile internet respondents were higher than that of stationary Internet. Third, the results of two survey methods were significantly different by comparing construct validity that includes both discriminant validity and convergent validity. Finally, this paper ends with implications and limitations of using mobile Internet as a survey tool.

• Journal of Korean Academy of Nursing
• /
• v.49 no.1
• /
• pp.59-68
• /
• 2019

Purpose: This study evaluated the psychometric properties of the Korean version of Yale Food Addiction Scale for Children (YFAS-C). Methods: Participants were 419 young adolescent students (11~15 years old). The content validity of the expert group was calculated as the content validity index (CVI) after the translation and reverse translation process of the 25 items of the YFAS-C. The multitrait-multimethod matrix (MTMM) method was used to verify the construct validity; the generalized linear model (GLM) was used to evaluate the concurrent and incremental validity. Reliability was calculated as Kuder-Richardson-20 (KR-20) and Spearman-Brown coefficients. Results: The CVI of the 25 items was greater than the item-level CVI .80 and the scale-level CVI .90. The Korean version of YFAS-C had verified convergent validity in emotional eating and external eating and discriminant validity in restrained eating. In addition, it had verified concurrent validity in emotional eating and external eating. Finally the incremental validity of the Korean version of YFAS-C was statistically significant on BMI. Reliability was KR-20 ${\alpha}=.69$ and the Spearman-Brown coefficient was .64. Conclusion: The Korean version of YFAS-C is a valid and reliable scale for measuring the severity of food addiction; it can be a useful scale for preventing obesity by predicting food addiction early.