A Study on Measurement Methods for Qualification and Skill Mismatch of Youth

본 연구의 문제의식은 불일치 측정의 신뢰도와 타당도를 높이는데 있다. 학력과잉과 스킬불일치 논의 발전을 위해서는 이를 측정하는 방법에 따른 한계를 명료히 하고, 가능하다면 신뢰도와 타당도 면에서 개선된 측정 방법을 찾는 데 연구의 목적이 있다. 즉, 학력과잉과 스킬불일치 측정의 문제점을 명료히 하고, 대안 또는 최선안을 모색하는데 본 연구의 목적이 있다.1. The Objective of Research Despite growing concerns over social and individual costs deriving from qualification and skills mismatches, the reliability and validity of the method that measures such mismatches has not been thoroughly examined. Considering the fact that typical domestic panel surveys and existing researches often diagnose mismatches based on only one question, it is basically impossible to verify the reliability of measurement for mismatches. This research focuses on clarifying the limitation of the measurement method and identifying an improved method of measurement in terms of reliability and validity. In other words, the objective of the study lies in clarifying the problem related to the measurement of qualification and skill mismatches, and exploring an alternative solution. The research questions are as follows: First, how did the existing research measure qualification and skill mismatches? What are the characteristics, as well as merits and demerits of each method? Second, what elements should be considered in order to measure qualification and skill mismatches in a simple, reliable and valid manner? Third, what questions can be used to reliably and validly measure qualification and skill mismatches? How is the reliability and validity of the instruments? What strengths do the instruments developed through this research have, compared with existing tools of measurement? Fourth, what is the current state of qualification mismatches for youth? What differences are there between the “required qualifications” that employers demand and the “appropriate qualifications” that workers perceive? Fifth, in the process of perceiving skills mismatches, are they put different weight on between general skills and specific skills? Sixth, what is the relationship between qualification and skill mismatches? In what proportions do genuine qualification mismatches and apparent qualification mismatches appear amongst the country''s youth? 2. Research Method FGI was carried out to examine cognitive and emotional process that individuals go through in judging qualification and skills mismatches, and to verify the preliminary survey questions. FGI was performed on 7 young subjects currently in employment. The participants consisted of 3 men and 4 women, 4 of which came from a humanities background and 3 from a science background. All of them have university degrees. After the FGI and consultation meeting of experts, online survey was conducted with 1,200 young employed men and women as its participants. Participants were qualified for the survey if they have qualifications higher than junior college and they are less than 35 years for males and 32 for females, employed by businesses with more than 5 employees, and working for a period between 6 months and 5 years in the same industry. The survey was composed mainly of instruments for measuring mismatches and of several factors for examine validity of the instruments. Based on the analysis of the results of the survey, a small seminar was convened to discuss the validity of interpretation of the results, the structure of the research paper, and so on. 3. Review on Measurement Methods for Qualification and Skill Mismatches A. Review of measurement methods for qualification mismatches Objective method which compares the educational level of respondents to the level set by job analysis has a problem in that job analysis doesn''t offer perfect data. The data from job analysis reflects the average qualification requirements in a certain job group, not the qualification requirements of individual job in a group. Furthermore, job analysis data often cannot reflect the changes in educational level according to technological changes and circumstances in workplace. The subjective measurement method has limits in that it measures subjective awareness by respondents. This method, however, has some merits in that it can reflect job specialty and changes of educational level in the labor market. Apart from bridging the gap between subjectivity and objectivity, subjective awareness is worth examining as it has an effect on job satisfaction and organizational commitment. In conclusion, there appears to be no correct answer as to the methods of measuring mismatch and it is adequate to determine the measurement method depending on the availability of data. This research will suggest a method for measuring qualification mismatches subjectively by carefully deliberating questions of self-report measurement. In particular, the research will inquire into the relationship among various questions, also considering questions that distinguish "qualifications required for the employment of a certain work" from "qualifications that are appropriate for carrying out their work". B. Review of measurement methods for skill mismatches Given the fact that most measurements for skill mismatches are subjective, this study will measure mismatches in a subjective manner. When asked whether to agree to both skills mismatch and skill match respectively in a parallel manner as used in PIAAC, respondents may agree simultaneously to having skill mismatch and skill match. However, selecting only one statement when over-skilling, skill match, and under-skilling are presented simultaneously as used in ESWC, cognitive overload may be imposed on respondents. Therefore it is necessary to simplify information and apply visual configuration to lessen the cognitive burden to read complex content of the survey, properly grasp the meaning and make out differences among the three statements. Although the method used in large-scale domestic surveys is clear and simple in allowing respondents to evaluate their mismatches directly, it is hard to judge whether it measures mismatches reliably and validly because of oversimplified question. In this study, we explored the best methods for measuring skill mismatches subjectively, considering all these problems above in the existing questions. 4. Strengths of the Final Questions for Measuring Mismatches A. Final questions for measuring mismatches As a result of deliberating the reliability of candidate questions in multiple ways, three questions for measuring qualification and skill mismatches have been chosen respectively. For measuring qualification mismatches, three questions including “job criterion question”, “appropriate qualifications question” and “existing question” were chosen. The “job criterion question” involves choosing one’s status from “You can perform the current duty if you have ① lower/ ② similar/ ③ higher qualifications than/to your actual qualifications.” The “appropriate qualifications question” involves choosing educational level that are needed to practically perform duties, and its results are compared with the actual educational level of individuals to examine mismatches. The data from "existing question" with a 5-point scale were converted into a scale of 3 points. By yielding an average out of three questions, less than 1.5 was classified as under-qualification, more than 1.5 and less than 2.5 as qualification match, and more than 2.5 as over-qualification. The reliability of the total score for the 3 questions turned out to be 0.65. As for skill mismatches, three questions including “skill criterion question”, “further education needed”, and “existing question” were selected. The “skill criterion question” involves choosing one''s status from “My skill (technology and ability) is appropriate for the duty that is ① easier than/ ② similar to/ ③ harder than my current duty”. The “further education needed” question involved choosing one from “① I need further training to cope well with my duties,” “② My duties correspond well with my present skills,” and “③ I have the skills to cope with more demanding duties". Finally, for the "existing question", respondents were asked to answer the question of “How do you describe your level of skills compared with your current job contents?” on a scale of 5 points from “much lower than my level” to “much higher than my level” and the results were converted to a 3-point scale for analysis. By yielding an average out of three questions, less than 1.5 was classified as under-skilling, more than 1.5 and less than 2.5 as skill match, and more than 2.5 as over-skilling. The reliability of the total score for the 3 questions turned out to be 0.72. B. Strengths of a new measurement method Compared with the existing single-question and 5-point scale method, the merits which were generated through a new measurement method are as follows. Firstly, when mismatch was measured by a new method, the percentage of mismatch decreased in comparison with that of single-question measurement. In particular, the rate of “under-qualification(skilling)” showed an obvious decrease. This phenomenon is attributed to the fact that, in existing questions, 1 and 2 points are judged as under-qualification(skilling) and 4 and 5 as over-qualification(skilling) while "qualification(skill) match" was allotted only to 3 on a 5-point scale. In contrast, the new methods make it possible to evaluate mismatches in a conservative way with regular intervals for each type. Secondly, as the existing question judges mismatches using only one question, it is fundamentally impossible to consider the reliability of such judgment. Because the new method can verify reliability, it is an improved way in methodological terms. Thirdly, the research compared the validity of the existing measurement method to that of the new method by comparing the degree of relationship with variables whose relevance was identified in previous research. The new method had more validity than the existing one in diagnosing “under-qualification(skilling)” as a result of binary logistic regression analysis. 5. Current Status and Related Factors of Qualification and Skills Mismatches The research examined the current status and related factors of qualification and skill mismatches of employees with less than 5 years’ experience. The major facts are as follows: First, 36.6% of all respondents proved to possess over-qualifications. In contrast, the over-skilling was only 23.8%, which indicates that considerable number of respondents with over-qualifications are not mismatched in terms of skill. The percentage of under-qualifications was 3.7% and that of under-skilling was 6.7%. Second, only qualification mismatches were related to educational level and university majors, whereas skill mismatches didn''t show significant difference related to education level and university majors. The group of college graduates and postgraduate had high ratio of over-qualifications (40.1% and 41.4%, respectively) compared to the group of junior college graduates (29.1%). Respondents with humanities background (40.6%) had high over-qualification ratio compared to those with science background (31.6%). Third, under-qualification group were found to possess a positive profile in comparison with qualification-match and over-qualification group. That is, under-qualification group was shown to have higher levels of job satisfaction and organizational commitment than other groups. The average wage for under-qualification group (average annual salary of KRW 33.36 million) was higher than qualification-match group (KRW 31.64 million) and the over-qualification group (KRW 29.33 million). Job stress, however, was also found to be the highest in the under-qualification group. Fourth, while the under-qualification group showed the most positive profile in the area of qualification mismatches, the skill-match group showed the most positive profile in the area of skill mismatches while under-skilling group and over-skilling group revealed similar characteristics. The rate of job satisfaction for skill-match group was the highest with an average of 3.34 on a 5-point scale while that of under-skilling was 2.92 and over-skilling was 2.95, which were similarly low. Organizational commitment showed a similar pattern, with skill-match group recorded the highest average of 2.89, while the under-skilling group (2.42) and over-skilling group (2.37) scored low. Wages were higher in descending order from skill-match (KRW 31.93 million), under-skilling (KRW 30.11 million) and over-skilling (KRW 27.92 million). The under-skilling group was the highest in job stress. Fifth, about 53.4% of over-qualification group were not over-skilled, and as such they were found to possess only "apparent" over-qualification.” 16.9% of all respondents were both over-qualified and over-skilled, and 19.3% were apparent over-qualified. 48.7% of all respondents showed qualification and skill match. Junior college graduates showed high apparent over-qualification proportions in contrast with genuine over-qualification, and it was clear that the larger the scale of a workplace is, the proportion of apparent over-qualification increased in contrast with genuine over-qualification. In comparison with the group with genuine over-qualification, the group with apparent over-qualification tended to be higher in job satisfaction, wage, and organizational commitment. Such a high proportion of apparent over-qualification can be relevant to the fact that qualifications are so highly valued in our society. The fact that the proportion of apparent over-qualification is higher if the scale of companies is larger indicate that they possess more job-related knowhow concerning human resources than is appropriate for the level of skill. Sixth, it is also noteworthy that a difference exists between the required qualifications that are demanded by employers and the appropriate qualifications that workers think necessary. 71.3% of workers perceive that required qualifications are identical to appropriate qualifications while 24.5% think more highly of required qualifications than appropriate qualifications. What is noteworthy is that the location of colleges and universities had an effect on the different types of mismatches. As for many non-metropolitan-based colleges and universities, required qualifications were higher than appropriate qualifications (26.9% vs. 20.9%). Non-regular workers were higher in required qualifications than in appropriate qualifications, compared with regular workers. It can be interpreted that graduates from provincial colleges and universities and non-regular workers engage in low-level jobs than the level of qualification that is commonly perceived. A similar interpretation can be possible for the fact that the group with required qualifications is consistent with appropriate qualifications was higher in job satisfaction, organizational commitment and annual salary than the group which has higher required qualifications thatn appropriate qualifications. Seventh, it was found that young salaried employees were weighing on work-related specific skills than on general skills in the awareness of skill mismatch. 6. Suggestions and Implications Suggestions for future studies for measuring mismatches include the following: Firstly, as the new measurement method which add scores of each question have merits in reliability and validity compared with existing methods using only one question. It is suggested that this method would be used in future surveys for mismatches. It is also suggested that three questions should be used to enhance reliability. When only two questions are used and if different choice is made on each question, these responses cannot be categorized into the options of over-, under-, and match, and hence should be avoided. Secondly, if only one question is to be used in cases where three questions cannot be used, we recommend the "job criterion question" for qualification mismatches and the "skill criterion question" for skill mismatches. Thirdly, as research discovered that there is a possibility that mismatch can be excessively estimated by performing a survey on a 5-point scale as in the existing question, the use of a 3-point scale is recommended. In addition, the research has drawn implications for a necessity of choosing more reliable measurement instrument, for sorting out curricula and the system of qualifications in order to help individual capabilities to be exposed more precisely, and finally, for further study.요 약 제1장 서론_1 제1절 연구의 필요성 및 목적3 제2절 주요 개념과 연구의 범위7 제3절 연구 내용 및 방법16 제2장 학력 및 스킬 불일치 측정 방법 유형 및 사례_23 제1절 학력불일치 측정 방법25 제2절 스킬불일치 측정 방법34 제3절 본 연구에의 시사점38 제3장 학력 및 스킬 불일치 측정도구 개발의 방향과 절차_43 제1절 측정도구 개발의 기본 방향45 제2절 타당성 검토를 위한 모델50 제3절 문항 개발의 절차58 제4절 본조사 문항 구성66 제4장 학력 및 스킬 불일치 측정 문항 분석_69 제1절 학력 및 스킬 불일치 문항 개요71 제2절 학력불일치 문항 간 일치도 분석78 제3절 스킬불일치 문항 간 일치도 분석85 제4절 최종 문항 선정90 제5장 학력 및 스킬 불일치 현황과 측정 방법 타당도 분석_93 제1절 학력 및 스킬 불일치 집단별 차이95 제2절 불일치에 따른 임금 및 만족도 차이98 제3절 학력불일치 및 스킬불일치 관련 요인104 제4절 불일치의 질적 차이 분석109 제5절 측정 방법 타당도 비교117 제6장 결론_121 제1절 불일치 측정 방법에 관한 결론 및 제안124 제2절 학력과 스킬 불일치 현황 요약 및 논점127 제3절 제안 및 시사점130 SUMMARY_133 참고문헌_147 부 록_159 1. 학력 및 스킬 불일치의 개념, 현황 및 원인과 결과161 2. 조사 설문지19