Earticle

The mental health of rural-to-urban migrants in China is a critical issue. The aim of this study was to test the migrants' mental health. The findings drawn from this qualitative study of 769 migrants in Wuhan in 2012 based on the Bayesian structural equation model. Overall, the survey found that leisure plays the greatest positive role in migrants' mental health, as well as work, interpersonal relationships, and health status have a negative role in migrants' mental health. Thus, the government must set relevant regulations to help migrants establish a better life and work values to work energetically.

本研究の目的は、青少年の精神健康に及ぼす親支援、教師との関係、ストレス変数間の構造的影響関係を確認することである。韓国教育開発院で施行した「韓国教育縦断研究2013」の2020年8次調査の高校3年生7,324人パネルの中でアンケートに参加した6,054人を研究対象者に選定し、SPSS 26.0、Amos 24.0を活用して、頻度分析、技術統計、相関関係分析、ベイジアン構造方程式モデル経路分析を行った。第一に、青少年の親支援は精神健康に統計的に有意な正[+]の影響を与えた。第二に、教師との関係はストレスに負[-]の影響を与え、精神健康に正[+]の影響を与えた。第三に、ストレスは青少年の精神健康に負[-]の影響を与えることが確認された。第四に、教師との関係はストレスを媒介変数として精神健康に正[+]の影響を与えた。本研究は、青少年の精神健康問題が深刻になる時点で精神健康に及ぼす予測変数を確認し、研究結果は精神健康改善のため教育現場での政策樹立とプログラム運営に関する資料とすることができる。

The purpose of this study is to confirm the structural relationship between parental support, relationships with teachers, and stress variables on adolescents' mental health. Among the panel of 7,324 third-year high school students in the 8th survey in 2020 of the ‘Korean Education Longitudinal Study 2013’ ??conducted by the Korea Educational Development Institute, 6,054 people who participated in the survey were selected as research subjects. Frequency analysis, descriptive statistics, correlation analysis, and Bayesian structural equation model path analysis were performed using SPSS 26.0 and Amos 24.0. First, adolescents' parental support had a statistically significant positive effect on their mental health. Second, the relationship with teachers had a negative effect on stress and a positive effect on mental health. Third, it was confirmed that stress has a negative effect on the mental health of adolescents. Fourth, relationships with teachers had a positive effect on mental health with stress as a mediating variable. This study identified predictors that affect mental health at the point when adolescents' mental health problems become serious, and the research results can serve as data related to policy establishment and program operation in educational settings to improve mental health.

구조방정식 모형은 관측 변수와 잠재 변수 사이의 상호 연관성을 분석하기 위해 사용하는 방법으로, 측정 방정식과 구조식으로 구성되며, 통계학 뿐 아니라, 사회과학, 마케팅, 심리학, 정치학 등의 다양한 응용 분야에서 널리 사용된다. 최근 연구에서는 데이터의 구조식이 선형이 아닌 경우와 오차항이 정규성 가정을 만족하지 않는 복잡한 경우를 다루고 있으며, 선형성과 정규성에 바탕을 둔 기존의 구조방정식 모형은 이러한 데이터를 분석하는 데 한계를 보인다. 본 논문에서는 이러한 기존의 구조방정식 모형이 가지고 있는 선형성에 대한 가정과 오차항에 대한 정규성 가정을 완화하는 베이지안 준모수 구조방정식 모형에 대해서 고찰한다. 이를 위하여, 베이즈 스플라인 구조방정식 모형과 디리슐레 확률과정 혼합 모형을 결합하여, 비선형 방정식과 오차항이 정규분포가 아닌 경우에 사용할 수 있는 베이지안 준모수 구조방정식 모형을 연구한다. 모의실험을 통해 베이지안 준모수 구조방정식 모형의 성능을 고찰하고, 세계가치관조사(world values survey; WVS)와 한국종합사회조사(Korea general social survey; KGSS)자료를 바탕으로 실증적 비교 분석을 수행한다.

Structural equation model is a method used to analyze the correlation between observed and latent variables, consisting of measurement equation and latent equation, and is widely applied to various fields such as social sciences, marketing, psychology and politics, in addition to statistics. Some of recent studies have dealt with the data where nonlinear structural equations need to be used or the error distributions are unknown and do not have to follow the normality assumption. However, existing structural equation models have limitations in analyzing such data since they assume the linearity and normality. In this paper, we introduce the Bayesian spline structural equation models for estimating the nonlinear correlation between latent variables and propose an improved model that can be applied that the error terms do not satisfy normality assumption via Dirichlet process. The performance of the proposed method is verified by simulation studies and real data applications based on survey data from the world values survey (WVS) and Korea general social survey (KGSS).

최근 4차 산업으로의 산업 패러다임의 변화가 이미 시작되었으며, 이러한 4차 산업 시대에 무형 지식재산인 특허의 중요성은 날로 증대되고 있다. 특허의 기술가치평가는 전문가의 의견에 따라서 산정되기 때문에 많은 비용과 시간이 소모되므로 비전문가들의 주관적인 의견에 기인하여 특허의 질적 수준을 판단하게 된다. 따라서 특허의 질적 수준에 대한 객관적이고 합리적인 평가 체계 개발이 필요하다. 본 논문에서는 특허의 가치평가를 기술성, 권리성, 활용성으로 구분하고 베이지안 구조방정식을 사용하여 특허의 정량화되고 객관적인 평가 모델링에 대해 고려하였다. 특히, 한국발명진흥회에서 수집한 자료를 토대로, 직접적으로 측정되기 어려운 질적 성과들을 모형화하고 평가하는데 탁월한 구조방정식과 사전 정보를 활용함으로써 작은 표본 하에서도 안정적인 모형화가 가능한 베이지안 접근법을 함께 적용하여 특허 평가 모형을 개발하였다.

Recently, the industrial paradigm shift to the fourth industry has already begun, and the importance of patents as intangible intellectual property in the fourth industry era is increasing day by day. Since the technical valuation of a patent is calculated according to the opinion of experts, it is costly and time consuming, and hence, the quality of the patent is judged based on subjective opinions of non-experts. Therefore, it is necessary to develop an objective and rational evaluation system for the qualitative level of patents. In this paper, we classify the valuation of patents into technicality, rights, and usability, and consider the quantitative and objective evaluation modeling of patents using Bayesian structural equation model. In particular, based on the data collected by the Korea Invention Promotion Association, we apply the Bayesian approach, which is capable of stable modeling even under small samples by using prior information, and the structural equation model, which is excellent for modeling and evaluating qualitative performance that is difficult to measure directly, to develop a patent evaluation model.