리커트 척도에서 각 척도점(anchor)에 사용하기 적합한 표현들의 탐색 및 제안
Exploring and proposing conceptually equal-distanced anchor wording for each scale point of Likert scales to measure agreement, accordance, and frequency
0
조회
0
다운로드
1
인용
0
참고문헌
키워드
초록
리커트 척도는 심리학을 포함한 사회과학 등의 분야에서 자주 사용되는 측정 도구이다. 리커트 척도는 각 점수들 간의 간격이 동등하다는 가정 하에 사용되고 있으나 척도점에 사용되는 구체적인 표현들(예: ‘약간’, ‘매우’)이 무엇인지에 따라 응답자들은 각 점수 간의 간격을 동등하게 지각하지 않을 수 있다. 이는 응답값에 체계적 오류를 발생시켜, 그 결과 분석과 추정에 편향을 일으킬 수 있다. 이러한 문제점을 개선하고자 본 연구에서는 ‘동의 또는 해당’과 ‘빈도’를 측정하기 위해 사용하는 리커트 척도에 어떤 표현들이 주로 사용되는지를 조사하고 각 표현이 나타내는 의미를 수량화하여, 각 척도점에 사용하기 적합한 표현들을 선정하였다. 먼저 예비 조사를 통해 최근(2020년 1월 - 2022년 12월) 출판된 연구들에서 사용된 척도점의 표현들을 파악하였다. 그 결과, 최근 연구들에서 자주 사용하는 동의 표현이 32개, 빈도 표현이 26개임을 확인하였다. 다음으로 본 조사에서는 성인 남녀 205명에게 예비 조사에서 살펴본 각 표현이 의미하는 정도를 슬라이드 척도에 평정하게 하였다. 참가자들이 각 표현이 나타내는 의미가 얼마인지 응답한 값을 기반으로 4점, 5점, 7점 척도의 각 척도점에 사용하기 적합한 표현들이 무엇인지를 살펴보았다. 본 연구에서 선정한 각 척도점 표현들을 연구자들이 맥락에 맞게 수정하여 사용하길 제안하며, 각 척도점에 부적합한 표현을 사용하여 발생하는 응답의 체계적 오류를 줄이길 기대한다.
Abstract
The Likert scale is a measurement tool commonly used in various fields including social sciences such as psychology. Although the Likert scale is used with the assumption that the intervals between each scale point are equal, respondents may not perceive the intervals between each scale point equally, depending on the wordings used in scales (e.g., ‘slightly’, ‘very’). This can lead to systematic errors in responses and biases in analysis and estimation of results. In this study, we empirically explored and proposed anchor wordings suitable for use in each point, considering the errors induced by the wordings. Specifically, we investigated what anchor wordings have been commonly used in the Likert scale to measure ‘consent or accordance’ and ‘frequency’. We calculated and proposed what anchor wordings are most suitable for each scale point. In the preliminary survey, we collected and identified wordings of scale points used in recently published studies (January 2020 - December 2022). In the main survey, we recruited participants and had them rate the meaning of each anchor wordings obtained from the preliminary survey using slide scales. The preliminary survey found that 32 anchor wordings of consent and 26 of frequency were frequently used in recent studies. In the main survey, we proposed what anchor wordings are suitable for each scale point on the 4-, 5-, and 7-point scales based on the values that participants responded to what each wordings represent.
참고문헌 (25)
- "매우" (2008)[URL]
- "전적" (2008)[URL]
- Alwin, D. F., "Margins of error: A study of reliability in survey measurement" (2007), John Wiley & Sons
- Alwin, D. F., "The reliability of survey attitude measurement : The influence of question and respondent attributes", Sociological Methods & Research, 20(1), pp.139~181 (1991)[DOI]
- Bassili, J. N., "Response-time measurement in survey research a method for CATI and a new look at nonattitudes", Public Opinion Quarterly, 55(3), pp.331~346 (1991)[DOI]
- Bishop, P. A., "Use and misuse of the Likert item responses and other ordinal measures", International Journal of Exercise Science, 8(3), pp.297~302 (2015)
- Casper, W. C., "Selecting response anchors with equal intervals for summated rating scales", Journal of Applied Psychology, 105(4), pp.390~409 (2020)[DOI]
- Croasmun, J. T., "Using likert-type scales in the social sciences", Journal of Adult Education, 40(1), pp.19~22 (2011)
- Dai, S., "Performance of Polytomous IRT Models With Rating Scale Data : An Investigation Over Sample Size, Instrument Length, and Missing Data", Frontiers in Education, 6, pp.1~18 (2021)[DOI]
- Göb, R., "Ordinal methodology in the analysis of Likert scales", Quality & Quantity, 41, pp.601~626 (2007)[DOI]
- Jamieson, S., "Likert scales : How to(ab)use them", Medical Education, 38, pp.1212~1218 (2004)[DOI]
- Joshi, A., "Likert scale : Explored and explained", British Journal of Applied Science & Technology, 7(4), pp.396~403 (2015)[DOI]
- Knapp, T. R., "Treating ordinal scales as interval scales : An attempt to resolve the controversy", Nursing Research, 39, pp.121~123 (1990)[DOI]
- Kvam, P. D., "A unified theory of discrete and continuous responding", Psychological Review, 130(2), pp.368~400 (2023)[DOI]
- Likert, R., "A technique for the measurement of attitudes" (1932), Columbia University Press
- Maitland, A., "Should I label all scale points or just the end points for attitudinal questions?", Survey Practice, 2(4), pp.1~5 (2009)
- Moors, G., "The effect of labeling and numbering of response scales on the likelihood of response bias", Sociological Methodology, 44(1), pp.369~399 (2014)[DOI]
- Saris, W. E., "Design, evaluation, and analysis of questionnaires for survey research" (2007), Wiley-Interscience
- Schwarz, N., "Rating scales numeric values may change the meaning of scale labels", Public Opinion Quarterly, 55(4), pp.570~582 (1991)
- Spector, P. E., "Choosing response categories for summated rating scales", Journal of Applied Psychology, 61(3), pp.374~375 (1976)[DOI]
- Stevens, S. S., "On the theory of scales of measurement", Science, 103, pp.677~680 (1946)[DOI]
- Symonds, P. M., "On the loss of Reliability in Ratings Due to Coarseness of the Scale", Journal of Experimental Psychology, 7(6), pp.456~461 (1924)[DOI]
- Tilton, J. W., "The measurement of overlapping", Journal of Educational Psychology, 28(9), pp.656~662 (1937)[DOI]
- Tsui, M. S., "Empirical Research on Social Work Supervision : The State of the Art(1970-1995)", Journal of Social Service Research, 23(2), pp.39~54 (1997)[DOI]
- Wu, H., "Can Likert scales be treated as interval scales?-A simulation study", Journal of Social Service Research, 43(4), pp.527~532 (2017)[DOI]