Abstract

본 연구는 한국청소년정책연구원의 한국아동청소년패널(Korea Children Youth Panel Survey; KCYPS) 중1패널 6차년도 자료에 기계학습 기법 중 벌점회귀모형으로 분류되는 adaptive LASSO를 적용하여 진로결정 관련 변수를 탐색하였다. Adaptive LASSO는 LASSO를 개선한 방법으로 변수 선택과 회귀계수 산출이 동시에 가능하며, 변수 선택 일치성을 충족한다는 장점이 있다. KCYPS의 고등학교 3학년 1,938명이 응답한 326개의 설명변수 중 학교/학력, 지적발달, 사회정서발달, 진로계획 영역(이상 개인발달) 변수 17개, 가정환경, 교육환경, 지역사회환경, 매체환경, 활동/문화 환경 영역(이상 발달환경) 변수 14개의 총 31개의 변수가 진로결정 관련 변수로 선택되었다. 본 연구 결과, 정서문제, 진로정체감, 양육방식, 학교생활적응과 같이 선행연구에서 연구된바 있는 기존 변수들을 확인하였으며, 이와 함께 직업관, 지역사회 인식, 휴대전화 보유여부, 체험활동 및 동아리 활동 참여 유무와 같은 새로운 변수들을 탐색할 수 있었다. 자신의 적성을 인지하고 직업의 세계를 탐색하도록 돕는 진로교육의 중요성이 증대되고 있는 시점에 본 연구는 교육적 개입이 어떤 방향으로 나아가야할지 제시하였으며, 심층 연구가 필요한 일부 변수에 대한 후속 연구 또한 제안하였다.The current study explored variables relating to students’ career decisions, using 7th graders’ 6th wave panel of KCYPS. In particular, adaptive LASSO was employed among penalized regression techniques. Despite its strength in variable selection, LASSO is known to produce inconsistent coefficient estimates. In response to the LASSO shortcoming, adaptive LASSO was proposed as a technique to yield consistent estimates. A total of 326 variables responded by 1,938 12th graders were investigated in the adaptive LASSO model and 31 variables were selected as important after relevance counts. The selected 31 variables included 17 and 14 variables from personal development and development environment sections, respectively. Among the 31 selected variables, those regarding emotional problems, career identity, parents’ parenting style, and school adaptation have been investigated in previous research. Newly found variables included variables relating to sense of occupation, community awareness, cell phone possession, and extracurricular activities. The importance of career education is increasingly stressed, as career education helps students recognize their aptitudes and explore possible occupations. Some of the selected variables have practical implications, as educational intervention is possible with them. Further research topics were also suggested.

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call

Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.