• 대한지역사회영양학회지
• /
• 제14권5호
• /
• pp.600-607
• /
• 2009

The purpose of this study was to examine if Likert scales had been properly utilized in community nutrition research. A total of 527 research articles published in the 32 issues of Korean Journal of Community Nutrition from the volume 5, issue 1 in 2000 to the volume 10, issue 2 in 2005 were screened and 55 articles were found to have utilized one or more Likert scales for the studies. Therefore, 109 Likert scales used in the 55 studies were reviewed regarding the name, statement and response items, reliability and validity check, and analysis method. The scales were mostly referred as Likert scales (60%) or Likert-type scales (27%). Some scales were found to be referred as Likert scales although they were Likert-type scales when judged based on the information given in the respective articles. However some scales couldn't be judged for the rightness of the names because the information given for the scales in the articles was not enough. About 23% of the scales consisted of items less than 6 or more than 30, and therefore found to be inappropriate. The percentage of the scales listing all the statement items in the articles was only 25%. Most of the scales (85%) included 5 response items, and the rest included 4 (7%), 7 (6%), or 3 (2%). The percentages of the scales including appropriate center and end items were only 2% and 22%, respectively. Less than half of the scales (41%) were found to have been checked for reliability and only one scale was reported to have been checked for validity. In some scales (6%), the responses were scored improperly for analysis. The responses to the scales were frequently found to have been analyzed by parametric statistics such as mean, ANOVA, t-test, and Pearson's correlation, which might be a problem depending on the size and distribution of study samples. In conclusion, there is much room for improvement in the use of Likert scales in community nutrition research.

• 품질경영학회지
• /
• 제41권2호
• /
• pp.209-220
• /
• 2013

Purpose: This paper compares and discusses the influence on the quality classification of Kano's questionnaire which is used for the Kano model(Kano et al., 1984), the 3-point Likert-scale newly proposed by Kano and the 5-point Likert-scale presented in this study. Methods: For the comparison, the current study conducts a survey of 631 television viewers. The classification results of the three methods are then compared with those of direct classification which is adopted as a standard for classification of quality attributes. Results: The agreement rates between the results using conventional Kano's questionnaire and the results using direct classification is higher than the results using 3-point and 5-point Likert-scales. In addition, the attributes grouped as must-be or attractive in the direct classification appear to be classified as one-dimensional attributes in the Likert-scales. Conclusion: In comparison with the convensional Kano's questionnaire, the Likert-scale questions highly tend to classify the quatity attributes as one-dimensional. Although the classification results of the 3-point and 5-point Likert-scales are the same, the 5-point Likert-scale has the advantage to classify quality attributes in more detail.

• 품질경영학회지
• /
• 제30권3호
• /
• pp.38-52
• /
• 2002

This paper is concerned with the analysis of structural equation model(SEM) with the ordinal data such as Likert scale. The SEM is misused when the arbitrary scores allocated to the Likert scale are treated as quantitative data. The underlying distribution approaches have been studied to solve this problem, and the partial least squares(PLS) Is also tried. In this paper the quantification methods for the Likert scale are proposed to analyze the SEM. We assume that the Likert scale is an observation of the interval of the continuous underlying distribution, and the respondents have their own patterns in the response of some questions. Normal and beta distributions as the response patterns are considered to quantify the Likert scale. To compare the efficiency of the proposed method the bootstrap simulations are tried.

• 한국과학교육학회지
• /
• 제34권4호
• /
• pp.385-391
• /
• 2014

일반적으로 연구자들은 관찰할 수 없는 심리적인 특성들을 연구할 때 주로 척도를 이용하여 수량화한다. 과학교육 연구에서도 척도를 활용하는 사례가 자주 나타난다. 특히, 정의적 영역의 평가에서는 학생들의 태도나 인식 등의 특성을 파악하고 평가하기 위하여 척도를 많이 사용한다. 그 중에서 사용이 간편한 리커트 척도를 많이 활용한다. 하지만 리커트 척도의 경우 활용 과정에서 여러 오류들이 발생하여 이를 절대시하는 것은 유의해야 한다. 이에 본 연구에서는 흥미와 태도관련 과학교육 연구에서 리커트 척도의 활용에 따른 중심화 경향을 살펴보고 연구 대상 및 연구 내용에 따라 그 경향성이 어떠한지 분석하였다. 연구결과 연구 대상에 따라서는 대상 학교급 즉, 연령이 높아질수록 중심화 경향이 더 크게 나타났으며, 영재학생보다 일반학생의 경우 중심화 경향이 더 크게 나타났다. 연구 내용에서는 자기노출이 비교적 큰 영역에서 중심화 경향 더 나타났다. 그러므로 과학교육 연구에서도 리커트 척도를 활용할 때 중심화 경향의 오류가 발생하는 것을 알 수 있었다. 따라서 과학교육 연구에 리커트 척도를 활용할 때는 사회문화적 환경과 연구 대상 및 연구 내용의 특성을 고려할 필요가 있다. 또한 리커트 척도를 개선하여 적용해야 하며 나아가 조사 결과의 수치에만 의존하여 해석하는 것을 피해야 할 것이다.

• 가정과삶의질연구
• /
• 제21권1호
• /
• pp.13-24
• /
• 2003

The purpose of this article Is to review the problems in using Hazan & Shaver,s adult attachment inventory as spousal attachment inventory. In summary, Hazan & Shaver's forced-choice inventory is required to caution in using for spouse attachment inventory because of the disproportionally higher secure-attachment type ratio(72% - 89%). It was previous reseaches that used multi-item Likert inventory instead of Hazan & shaver's forced-choice inventory for result analysis because of disproportionally higher secure-attachment type ratio. Although Hazan & shaver's forced-choice inventory was mostly used for young unmarried adults, Using Hazan & shaver's forced-choice inventory as spouse attachment inventory for married adults seem to be inappropriately. Hazan & Shaver's multi-item Likert inventory has the problem of poor reliability in insecure-avoidence factor especially. Hazan & Shaver's one item Likert inventory has the problem of social desirability.

• 한국지능시스템학회논문지
• /
• 제15권5호
• /
• pp.533-537
• /
• 2005

리커트 척도는 어떤 의도의 질문에 대한 동의를 수치적으로 나타내는 방법으로서 주로 5점 척도를 많이 사용하나 여러번 질문에서 응답자체가 애매성을 가지고 있다. 본 연구에서는 3점 척도의 응답의 애매성을 5점 척도의 멤버쉽 함수로 나타낸 후 한 항목의 응답의 95% 신뢰구간을 다시 멤버쉽함수로 하여 그 데이터를 퍼지가설검정하였다.

• Journal of the Korean Data and Information Science Society
• /
• 제27권2호
• /
• pp.373-380
• /
• 2016

리커트형 데이터가 순서척도임에도 불구하고 많은 연구자들이 구간척도로 간주하여 모수 방법을 적용하고 있다. 본 연구에서는 리커트형 데이터를 어떻게 분석하는 것이 적절한지 모의 실험을 통하여 알아본다. 다양한 분포를 갖는 5점 리커트형 표본을 추출하여 위치를 비교하고, 순서척도의 경우 위치비교 보다 응답분포를 살펴보는 것이 더 타당하므로 응답분포 검정을 실시한다. 위치를 비교하는 방법으로는 구간척도로 생각하고 분석하는 모수 방법인 t-검정과 순서척도로 생각하는 비모수 방법인 만- 휘트니검정 (M-W검정)을 적용하고, 응답분포를 검정하기 위해서는 카이제곱 검정과 콜모고로프 - 스미르노프검정 (K-S검정)을 실시한다. 네 가지 방법의 효율성을 비교하기 위하여 제 1종 오류 (Type I error)의 비율과 검정력 (power)을 구한다.

• 동아시아식생활학회지
• /
• 제25권2호
• /
• pp.363-371
• /
• 2015

This study investigated the effect and interrelationship of evaluation system in employee's job satisfaction at deluxe hotel kitchen in the Seoul area. The questionnaires developed for this study were distributed to 400 chefs during June to August, 2012. A total of 364 questionnaires were used for analysis (91.0%) and the statistical analysis was completed using PASW (SPSS) 18. The status of operating menu evaluation and menu development was analyzed by dividing the questionnaires into three categories: 'menu idea', 'menu evaluator' and 'menu evaluation system'. The 'menu idea' category showed the highest concern for 'Is the selected idea differentiated from competitors' menu?' (3.10 Likert scale), the 'menu evaluator' category showed the highest concern for 'Do the evaluator's individual factors affect menu evaluation?' ($3.88{\pm}0.96$), and the 'menu evaluation system' showed the highest concern for 'Is the menu commercialized by the menu evaluation system?' (3.32 Likert scale). Overall job satisfaction was studied into three categories: 'human field', 'material field' and 'evaluation field'. Overall satisfaction about one's job was 3.24 Likert scale. For satisfaction in each category, satisfaction for 'human field' was the highest with 3.30 Likert scale compared to 'material field' and 'evaluation field' with 2.76 Likert scale and 3.27 Likert scale, respectively. In conclusion, establishment of a unified evaluation system to improve employee's job satisfaction related to menu development is needed.

• 한국한의학연구원논문집
• /
• 제16권2호
• /
• pp.159-166
• /
• 2010

Objective : The purpose of this study is to compare the Effect of Acupuncture on Local Acupoints and Distal Acupoints for Patients with Whiplash Injury. Method : From January 1st 2009 to September 31th 2009, 10 patient were divided into 2 groups. One group was taken Local Acupoints, and the other group was taken Distal Acupoints. To check pain and treatment satisfaction, visual analog scale(VAS) and five-point likert scale were used before and after treatment. Result : Both acupuncture theraphy showed improvement in VAS and five-point likert scale. And Local Acupoints group showed more effective than Distal Acupoints group on five-point likert scale. Conclusion : This Study suggests that Acupuncture on Local Acupoints and Distal Acupoints can be applicable to improve symptom in Patients with Whiplash Injury.

• 종양간호연구
• /
• 제3권1호
• /
• pp.75-84
• /
• 2003

Purpose: this study was to identify influencing factors of burnout on oncology nurses. Method: the data was collected using a questionnaire from October 18, 2002 to November 20, 2002 in five general hospitals at Seoul. Participants of this study were the nurses who worked in the wards occupied by more than 50 percent of cancer patients. The collected data were analyzed by using the SPSS10.0 program. Result: 1. The burnout of the subjects show a mean score of 2.72 using the 5 point Likert scale. There were statistically significant difference in burnout between the general characteristics; age, marital status, educational background, job experience, position, nursing service period, and attitudes on nursing job. 2 The self-esteem of the subjects show a mean score of 3.72 using the 5 point Likert scale. The spiritual well-being of the subjects show a mean score of 3.83 using the 6 point Likert scale. The social support of the subjects show a mean score of 3.74 using the 5 point Likert scale. The job stress of the subjects show a mean score of 2.16 using the 4 point Likert scale. 3. The result of correlation between burnout and other variable was that relationship between the burnout and job stress(r=-.206, p<.05), self-esteem(r=-.417, p<.01), spiritual well-being(r=-.403, p<.01), social support(r=-.386, p<.01) were significant variables. 4. The result of the Stepwise Multiple Regression was that self-esteem (17.4%), spiritual well-being(7.9%), social support(3.4%) and job stress(2.3%) explained 31% of burn out.