Robust model for health related quality of life data / Norin Rahayu Shamsuddin … [et al.]

The job diversity of academicians teaching at higher institutions would indirectly affect their health and quality of life. Lecturers who are facing problems with their health might affect their performance and their teaching ability. In this study, the Health-Related Quality of Life (HRQoL) measures were used as instruments to assess the quality of life of lecturers from a public university in Malaysia. The HRQOL measures such as SF-36 consist of eight domains. These eight domains were summarized into two categories, which are physical component summary (PCS) and mental component summary (MCS). This paper addresses the effect of various factors dealt by a group of lecturers on PCS from a certain public university in Malaysia. A cross-sectional study was conducted during the semester, and a total of 193 respondents completed the questionnaires. Multiple linear regression models were used to relate the lecturers' PCS to the identified factors. Robust MM-method was used to estimate parameters in the model since the data consist of outliers. For a comparison, ordinary least square regression model was also included in the analysis. Analysis shows that robust model is preferred as it produces highest adjusted R-square and smallest residual standard error values.