Public Health Research Foundationストレスチェックリスト・ショートフォームの作成 : 信頼性・妥当性の検討 The Development of Public Health Research Foundation Stress Check List Short Form : Reliability and Validity Study
健常者の日常生活におけるストレス反応の表出を精神面と身体面から多面的に査定でき,かつ簡便に査定できるような自己評定式のストレス反応尺度〔PHRFストレスチェックリスト・ショートフォーム;PHRF-SCL (SF)〕を作成し,その信頼性と妥当性を検証した.18〜64歳までの男女34,556人に50項目の質問紙を実施し,因子分析により24項目を選定した結果,次の4下位尺度を得た:不安・不確実感,疲労・身体反応,自律神経症状,うつ気分・不全感.信頼性はCronbachのα係数を算出することにより検証した.また,因子的妥当性は確認的因子分析により,構成概念妥当性は得点が異なると考えられる群間差の検定とストレッサーとの相関分析により,基準関連妥当性はCMIの得点との相関分析により検討した.その結果,PHRF-SCL (SF)は高い信頼性と妥当性をもつ有用な尺度であることが明らかにされた.
Purpose: There are many different types of scales for measuring stress responses. However, most of available scales have certain problems. Some need considerable time and effort to administer, whereas others measure only psychological aspects of stress. The Public Health Research Foundation Stress Check List Short Form[PHRF-SCL (SF)]was developed to fulfill the need for a more practical scale. Its reliability and validity were examined in this study. Method: The original PHRF-SCL is a self-report scale that consists of 50 items measuring stress responses of young people and adults. Participants were 24,854 males and 9,702 females that responded to the PHRF-SCL. Factor analysis with Promax Rotation was conducted for item selection. Reliability of the test was examined using the Cronbach's α. Construct validity was examined using the following 3 analyses: a) confirmatory factor analysis, b) comparison among groups that were expected to show different stress responses, c) correlation analysis of PHRF-SCL (SF) scores and stressor scores. Criterion related validity (concurrent validity) was examined by conducting correlation analysis between PHRF-SCL (SF) scores and Cornell Medical Index (CMI) scores. Results: Twenty-four items of the PHRF-SCL (SF) were eliminated as a result of factor analysis. It was shown that PHRF-SCL (SF) has 4 factor structure: anxiety/uncertainty (6 items), tiredness/physical (body) responses (6 items), autonomic symptoms (6 items), and depression/feeling of insufficiency (6 items). PHRF-SCL (SF) had a high reliability coefficient (0.71-0.85). Also, PHRF-SCL (SF) scores and stressor scores were significantly correlated. Confirmatory factor analysis indicated high fitness between data and models. An ANOVA and a t-test indicated that a group of participants who had experienced a change of jobs had significantly higher scores on this scale than a control group without such an experience. Also, a group of participants with a stress control strategy had significantly lower scores on the scale than a control group without such a strategy. Moreover, CMI scores were significantly correlated with PHRF-SCL (SF) scores. Conclusion: The results of this study suggest that PHRF-SCL (SF) is a useful scale with high reliability and validity.
心身医学 46(4), 301-308, 2006-04-01