. Sleep quality is measured numerically by a validated scale called the Pittsburgh Sleep Quality Index (PSQI). The purpose of this study was to investigate what factors influence the sleep quality of nurses who often work various shifts and hours. Shift (0=Night, 1=Day), Age, Hours worked per week, Shifts per week, BMI and whether Children were in the home (1=Yes, 2=No). Nurses were provided a survey monkey link and the data was collected and aggregated into the FINAL Q2 database.

a) Perform an exploratory analysis to investigate the relationship of the variables to PSQI. What variables (age, BMI, shifts worked per week, hours worked per week) are significantly associated with poor sleep quality? NOTE: Higher PSQI is indicative of worse sleep quality.

b) Are there any significant differences in mean PSQI between night/day shift or whether children are present or not in the home? Perform the appropriate statistical test, justify why you used that test and give results.

c) Develop a multivariable linear regression model to predict sleep quality in terms of the significant (p<0.05) factors found above in steps a and b. Assess the accuracy of the model using the model diagnostics learned in class particularly R², overall ANOVA model, and individual elements along with multicollinearity measures such as VIF and tolerance. Finally perform and interpret a residual analysis to look for model bias and accuracy.

d) Are there any policy changes that a hospital could adopt for nurses that may mitigate the effects of poor sleep quality? Discuss how these might improve the sleep quality of nurses.

e) The investigator was relatively new to research and didnt consult a biostatistician before designing the survey monkey instrument. Hence some holes are present in the data. Suggest 3 additional data fields that could have been collected or other changes and justify why/how those extra pieces of data or changes would improve the study.