Question 1 Description: Prof Gustin Beiber has been teaching Hollywood Studies 101 for few decades and have some interesting hypotheses of his own about student behaviors and their association with academic performance. One in particular is his hypothesis that internal students who attend classes regularly perform better than those who do not (or cannot). Prof Beiber has never officially conducted a proper scientific study to test this hypothesis (or answer the research question whether internal students who attend classes regularly perform better than those who do not or cannot). 

Now Prof Beiber is ready to plan this study but he is neither familiar with different study designs nor does he have any knowledge of statistics to collect and analyze data in order to assess the statistical and practical significance of his findings. (Afterall he teaches Hollywood Studies, not Biostatistics or Epidemiology!).

He advertises and hires a research associate to help him out and this happens to be you. He has managed to write down names of few study designs and he asks you to briefly describe those to him and this link can be established between regular attendance and good academic performance. You did not expect this to be the first task, so you look in your attic/shed/storeroom/under the bed storage/ among the diverse collection of old USBs and portable hard drives and finally in an old ceramic box you find your EPID lectures and tutorial notes. You go through those which helps you tremendously in terms of clearing your head about what different study designs are, their unique features, what sort of participants are suitable and which measures of association and effect can be calculated from each design to link the exposure (Regular Attendance) with the outcome (Academic Performance), how to assess statistical and practical significance etc. These are the aspects you will describe for the designs listed below if you were to help plan this hypothetical study. 

a) Cross-sectional study. 

b) Prospective Cohort study.

c) Retrospective Cohort study.

d) Case Control study. 

e) Quasi-experimental study. 

f) Randomized controlled trial

(No need to worry about the ethics for some parts; this is purely to assess your understanding of different study designs and what they can offer). 

Note: Please be specific about the specified exposure and outcome given for this question. No marks for writing general design features from the lectures. We want you to ‘apply’ your knowledge of study designs in light of the given scenario.

Question 2. A study investigated the link between prostate cancer and Dietary Inflammatory Index Score (DIIS) measured as (High and Low) using a prospective cohort design. A total of 12153 men, all free of prostate cancer at baseline, were recruited for this study. At the end of a five year follow up, 72 men were diagnosed with prostate cancer among 3500 who regularly consumed diet with high inflammatory index while there were 15 cases among those who consumed diet that had low inflammatory index. 

Please enter the above data in a new SPSS file and use ‘weight cases’ function. Carry out a suitable test and provide a complete interpretation of the test as well as the appropriate measure of association along with the statistical significance, based on both p value and 95%CI. (See lab materials for labs 5 & 6 for guidance). 

Question 3

Choose any two variables from your dataset that are suitable to explore Bivariate Association. You can recode and create any two categories as most categorical variables have more than two categories.  

Write null and alternative hypotheses

What are the assumptions of the test you used? Are any violated? Or not? Why.

Conduct the appropriate analyses as per your hypotheses, provide a short summary and meaning/interpretation of your results.

Question 4

Write null and non-directional alternative hypotheses that can be tested with an Independent Samples t test

Briefly describe, test and report whether homogeneity of variance assumption is violated or not & why (no need to worry about the other assumptions here).

Conduct the appropriate analyses as per your hypotheses. Provide complete interpretation of your results including their statistical significance as well as the practical significance.

Question 5

Choose only those who are ‘Very Happy’ with Life. Now test the hypothesis that they come from a population in which the average resting energy expenditure is 1450 calories per day. 

Write null and alternative hypotheses

Carry out the appropriate analyses, no need to list or test any assumptions, write a short summary of your results including their statistical significance

Question 6

a) What is the average statistics anxiety score for those who ‘Exercise Very Regularly’?

b) Recode and create a variable BloodGroup2 that has only two categories;

Category A = All those with Rh-Positive blood groups

Category B = All those with Rh-Negative blood groups

Now compare the average Marathon Completion time between the two categories. 

Which group is faster? 

Which group has the most variability & why?

c) Participants were asked to name one thing that drives them crazy (about themselves or others) (Variable DrivesMeCrazy). You will see that top five responses have been coded as A, B, C, D and E. Use your discretion (and creativity) to assign each code an actual value or description, make up any whatever you think these are/can/or should be, and provide a suitable graph showing five things that drive people crazy.

variables-names-and-description