A study was undertaken in relation to obsessive-compulsive disorder (OCD) to determine the impact of this disorder on family members who reside with the sufferer. typically family members begin to accommodate such behaviours and in essence, are drawn into the rituals associated with the disorder. this then causes those family members to perceive a burden which ultimately effects quality of life. Data concerning OCD and it's impact on family members measured the following: Severity of OCD perceived by a family member Accommodation behaviours Perceived bürden Perceived quality of life Subsequently a regression and was conducted to test the relationship between these variables. (a) If a correlation matrix was generated for each of the variables, what indications would you be looking for? (WORTH 4 MARKS) (b) Analysis of the data for the above study showed that the adjusted r-square was 0.49. Explain carefully what this statistic measures. How do you interpret this result? (WORTH 4 MARKS) ( c) In regression analysis what is the difference between Enter and Stepwise method of analysis? Which is the preferred method and why? (WORTH 4 MARKS) (d) Multicollinearity can spoilt estimates in a regression. On the other hand psychological constructs may reasonably be considered to correlate. Explain how you would assess multicollinearity among the independent variables for the study above? What action would you take to remedy it? (WORTH 4 MARKS) (e) Participants in the above study were classified in terms of their perception of the severity of OCD in a family member as follows: Low = 1 medium = 2 High = 3 If low is the reference group show how dummy variables could be created and coded for a regression analysis to determine the relationship between perceived severity of OCD and perceived quality of life. Explain the logic of this coding.
