FMHU5002 Introductory Biostatistics – Data Analysis and Reporting Assignment 1
Sydney School of Public Health Semester 1, 2024
FMHU5002 Introductory Biostatistics
Data Analysis and Reporting Assignment
Due Date and Time: Monday, June 3rd 2024 11:59 PM Sydney Time
Assignment Category: Submitted Work
Assignment Sub-category: Assignment
Weighting: 25%
Plagiarism and Academic Dishonesty Policy
You must complete your assignment alone. Submitting assignments that have been jointly completed
is not acceptable. Copying someone else’s work, using generative AI, or quoting from text without
adequate attribution of the source is plagiarism and is not acceptable. All assignments will be verified
by plagiarism detection software. Serious penalties apply for plagiarism, collusion, or contract
cheating. Information about the University’s policy on academic honesty can be found at the following
site:
https://www.sydney.edu.au/students/academic-integrity.html
Late Penalties and Special Consideration
Unless you have an approved simple extension, special consideration or an academic plan, 1.25 marks
(5% of 25) will be deducted from your assignment mark per day (or part thereof) until Monday, June
12th, 11:59 PM (Sydney Local Time). Assignments submitted past this date without approved special
consideration or an academic plan will not be accepted and will be given a zero (0) mark. For students
seeking simple extensions or special consideration, please use the following site:
https://www.sydney.edu.au/students/special-consideration.html
FMHU5002 Introductory Biostatistics – Data Analysis and Reporting Assignment 2
Sydney School of Public Health Semester 1, 2024
Submitting your Assessment
Submit your assessment as a single file in .docx or .pdf format by 11:59 PM Sydney Local Time on
Monday 3rd June 2024 via Canvas (Assessments overview > Assessment 4: Data analysis and
reporting assignment > Assignment Submission – Click Here > Select the file to upload and then click
“Submit Assignment”). Do not attach a jamovi.omv or a .csv file with your submission.
If you have any administrative questions, please post them on the Canvas Discussion Board.
Go to Discussions > Data Analysis and Reporting Assignment Discussion.
Alternatively contact the Associate Lecturer, Lucy Corbett: sph.epibio@sydney.edu.au
If you have difficulties submitting the assignment around the due time, please email
sph.epibio@sydney.edu.au directly with your assignment attached to avoid late penalties. The
timestamp of your email will be used as evidence of the date and time of your assignment submission.
Please note responses to emails will only occur during business hours on standard working days.
Important Notes:
• The data for the assignment has been simulated for the purposes of an assessment exercise.
As such, the outcomes from these analyses have no practical or clinical meaning.
• This assignment paper (including cover page, instructions, and data dictionary) is five (5)
pages in length. Please ensure you have all pages.
• The variable names and coding of the variables (i.e., the data dictionary) for your dataset are
included at the end of this assignment on page 5.
• Name your submission file with your student number (SID), unit of study code, and “A4” (e.g.,
311275249_FMHU5002_A4.pdf). Ensure all pages are numbered, and that your student
number is included in the header or footer of the document.
• Assignments are marked anonymously, so please do NOT put your name anywhere on the
assignment or submission title.
• Any jamovi output presented must be edited to comply with the recommendations for
presenting results as covered in the Lecture 1 Notes.
FMHU5002 Introductory Biostatistics – Data Analysis and Reporting Assignment 3
Sydney School of Public Health Semester 1, 2024
Assignment Questions
You have been asked to help analyse the results of a cross-sectional study among males aged 18-65
years old which was undertaken within a local health district to examine the impact of weight status
(i.e., overweight / not overweight) on health service usage (i.e., overnight hospitalisation in the past
12 months), and systolic blood pressure (mmHg).
Your collaborators within the local health district have provided you with the dataset (in the file
“FMHU5002 A4 Dataset.csv”, the data dictionary for which is provided on the last page of this
assignment document). The data has already been screened by a colleague, and all values in the
dataset should be considered correct (i.e., you do not need to screen the data for impossible,
implausible, extreme, or missing values).
Question 1 (12 marks)
a. Perform an appropriate analysis to assess whether there is an association between weight
status and overnight hospitalisations. In your response, clearly state the null hypothesis that
is being tested, and provide a brief conclusion including relevant statistical values. Your
response should be no longer than 4 sentences.
b. Perform an appropriate analysis to determine whether there is an association between weight
status and mean systolic blood pressure. In your response, clearly state the null hypothesis
that is being tested, and provide a brief conclusion including relevant statistical values. Your
response should be no longer than 4 sentences.
Question 2 (4 marks)
a. For the analysis performed in Question 1a, a prevalence of hospitalisation among people with
a weight status of overweight that is 1.4 times that of people not overweight would be
considered practically important. Based on your analysis, what can you conclude about the
practical importance of the results?
b. You are informed that a sample size calculation was performed in the design stage of the
study. This calculation determined that, in order to detect a practically important effect with
90% power at the 5% significance level, each study group would require 686 individuals.
However, due to staffing and budgetary constraints, this number was not achieved and no
further data can be collected. What are the implications of this on the detectable effect and
the precision of the estimated effect?
Note: you do NOT need to perform any sample size calculations to answer this question.
FMHU5002 Introductory Biostatistics – Data Analysis and Reporting Assignment 4
Sydney School of Public Health Semester 1, 2024
A colleague has also been analysing the data, and based on input from clinical collaborators, they
have included age as a potential confounder of systolic blood pressure in their model, generating
the output below:
Use the information provided in this output to answer the remaining questions (Question 3 and 4).
Question 3 (7 marks)
a. Provide an interpretation of the analysis based on the output above. In your response, clearly
state the null hypothesis that is being tested, and a brief conclusion including any relevant
statistical values. Your response should be no longer than 5 sentences.
b. Based on the provided output, what would be the predicted value of systolic blood pressure
for an individual who is overweight and is 50 years of age.
c. Based on the provided output, what would be the predicted value of systolic blood pressure
for an individual who is not overweight and is 50 years of age.
FMHU5002 Introductory Biostatistics – Data Analysis and Reporting Assignment 5
Sydney School of Public Health Semester 1, 2024
Question 4 (2 marks)
Your colleague has provided you with the following directed acyclic graph (DAG) to help explain why
they performed the analysis which produced the output shown above.
Assuming this DAG is a true reflection of the relationships between the variables shown, can the
regression estimate shown in the above output be interpreted as an estimate of the causal effect of
weight status on systolic blood pressure? Making reference to the DAG, explain why / why not?
Total = 25 marks
This is the end of the assignment questions.
The data dictionary for the assessment data set is provided below.
Data dictionary
Description Variable
Name
Levels
(if appropriate)
Identification number ID Nominal
Age (years) AGEYRS Continuous, measured in years
Weight status WGHTSTAT Dichotomous:
- Overweight
- Not overweight
Overnight hospitalisation in the last
12 months
HOSP Dichotomous:
- Hospitalised
- Not Hospitalised
Systolic blood pressure SBP Continuous, measured in mmHg