67-4

42/90

362branches to cooperate with our research and administered an anonymous online questionnaire survey to all their 495 employees. We explained to the employees that survey participation was volun-tary and would not affect their personnel evalua-tion. The survey was conducted from August 25 to September 9, 2020, and 427 respondents (mean age: 25.92 years, SD±5.34, 147 men and 280 women) were included in the analysis (response rate: 86.3%) (Table 1).MeasuresTeam psychological safety scaleThe 7-item team psychological safety scale developed by Edmondson9) was translated into Japanese and used with the original author’s consent. Some experts in organisational theory9) performed the first round of the scale’s translation. Subsequently, a translation company conducted back-translation. Further, we reviewed the differ-ences of the translated version from the original version and developed the final scale. The survey was answered using a 7-point scale with options ranging from 1, ‘Strongly Disagree’, to 7, ‘Strongly Agree’.Work engagementTo measure work engagement, we used the 9-item Utrecht work engagement scale (UWES-9), a shortened version of the Utrecht work engagement Gender　Male　FemaleAgeYears of work experienceYears of experience as an instructorOccupation　Instructor　Instructor and administrator　AdministratorJob title　Non-manager　ManagerSD, standard deviation.Table 1　Demographic characteristics of survey participantsnscale (UWES) developed by Schaufeli et al.3). This shortened version of the scale has been used to compare work engagement in various countries. The survey was answered using a 7-point scale with options ranging from 0, ‘Never’ to 6, ‘Always (every day)’. Job resourcesWe used the following items related to job resources that could affect work engagement as control variables: control over work, skill utilisation, job aptitude, rewarding from colleagues, support from superiors, and job satisfac-tion. These items were developed in compliance with the questions provided in the Brief Job Stress Questionnaire20).Personal attributesSubjects were asked to provide various demo-graphic details, such as their gender, age, educa-tional qualification, work location, work hours per week, number of team members, occupation, job title, years of work experience, and years of experi-ence as an instructor. To clarify the number of team members, since team psychological safety is a property of small groups10), we added question items to examine whether the number of team members affected team psychological safety and work engagement. However, since neither of the surveyed companies had small teams as official Percentage14728034.365.6339493979.411.59.131311473.326.7MeanSD25.922.342.355.341.812.36job, support