Assignment Task

A

SingHealth Polyclinics (SHP) wants to know what factors affect a patient’s ability to reach their blood cholesterol level goal, as set by their doctor. It surveyed 1093 patients with high blood cholesterol levels. The responses (n = 1093) were collated in the Report1_Q1 data set, which has 1093 rows (excluding headers) and 5 columns and is provided on Moodle. The variables in the data set are coded as follows:

• PATIENT - patient number, from 1 to 1093;
• ADHERENCE - equal to 1 if patient reported taking the cholesterol medication as instructed by the doctor, 0 otherwise;
• SMOKE- equal to 1 if patient reported smoking regularly, 0 otherwise; 
• MONTHS – number of months the patient had been on the cholesterol medication;
• METGOAL- equal to 1 if patient had met their blood cholesterol level goal, 0 otherwise. 

1. SHP wants to examine whether the factors, ADHERENCE, SMOKE, and MONTHS predict whether a person met the blood cholesterol level goal (METGOAL). What statistical test would you use to analyse these data? Provide a justification for choosing this test.

2. Using a statistical package (e.g., SPSS, jamovi, R), conduct the statistical test you identified above with the response variable METGOAL and the predictor variables ADHERENCE, SMOKE, and MONTHS. Copy-and-paste the relevant tables in your answer. Report the results of the analysis in APA format. At minimum, you should discuss whether the model is a good fit to the data, whether the parameters in the model are significant, and what the results mean.

3. Based on your analysis, write the formula for the probability that a patient would reach their blood cholesterol goal. You must include the actual values of the parameter estimates and define any terms used.

4. What is the probability of a patient reaching their blood cholesterol goal if the patient did not adhere to the medication regiment, was a smoker, and had been on the cholesterol medication for 6 monts? Show all working.

5. How much does the model improve the percentage of correct predictions compared to when no predictors are used? 

B

1. What parametric statistical test would you recommend Joshua use to analyse these data? Provide justifications for choosing this test. 

2. Write the statistical model in mathematical form, and name each of the terms in the model.

3. State all the null and alternate hypotheses that you are testing in mathematical symbols and words.

4. Conduct the statistical test you identified in (2a) using a statistical package. Copy-and-paste the relevant output in your answer. Then, report the findings in APA format. In your response, state explicitly which row of the output you chose to interpret and why. Use an alpha level of .05.

5. Given the results of the analysis, would you recommend any additional analyses? If so, state the analyses and why you recommend conducting them. If not, state why it is not necessary to conduct the additional analyses. Whether you recommend additional analyses, you do not need to conduct the analyses. No additional marks will be awarded for doing so.