BMS1042: Public Health and Preventive Medicine

Biostatistics tutorial – Week 2: Data variables and health

Tutorial objectives
● Illustrate your data with appropriate graphs, tables and figures
● Interpret the summary of your results and describe the impact on the population
● Summarise data using the appropriate summary statistics i.e. mean/SD or median/IQR (ebook)
Things needed for online tutorials:
● Laptop or tablet with Microsoft Excel toolkit | Graph Pad Prism 9 | SPSS (via MOVE)
(Note: Graph Pad Prism is not required until Week 3 & not available on tablets.)
● Scientific calculator.

Graph Pad Prism eBook – relevant lessons this week:

Regardless of software used:

 Descriptive statistics
 Pivot table (1 variable)
 Pivot table (2 variables)
 Theory behind contingency table %

If using Graph Pad Prism: All “Graphs” lessons

Before the tutorial complete:

● Worksheet
● Graphical & Chart Representations – ebook on Moodle

During the tutorial we will cover:

● Activity 1
● Activity 2
● Small group activity – Identify graphs

After the tutorial, you will complete in your own time:

● Consolidation Activity 3
Refresher: two-way table orientation

 Out of Exposure and Disease: What goes in the columns? What goes in the rows?

 Exposed, Unexposed and Total What’s the order of the exposure categories?

 Disease, No disease and Total What’s the order of the disease categories?

 What letters denote which cells for the RR (Relative Risk) and OR (Odds Ratio) formulas?

 Fill in the blank table below to show the usual way a table is arranged.

IMPORTANT: The formulas will only work if the table is arranged in this way.

Disease or Outcome

Disease No disease

Exposure Exposed

Unexposed

Total
Activity 1: Cross tabulation
A study on a university campus looking at alcohol drinking and security found the following:

Binge drinking
Never Frequent Total
Trouble with Yes 71 398 469
security
No 4992 2737 7729
Total 5063 3135 8198

a) What is the usual orientation of a table? Rearrange the table provided if necessary.

Binge drinking causes trouble with Trouble with security (outcome)

security
Yes No Total
Binge drinking Frequent 398 2737 3135
(exposure)
Never 71 4992 5063
Total 469 7729 8198

b) Calculate the overall frequency for “trouble with security”.

Which direction % are you calculating?

(469/8198) 5.72%

Calculate the frequency for “trouble with security” for those who never binge drink.

71/5063=1.4%

Calculate the frequency for “trouble with security” for those who frequently binge drink.

398/3135=12.7%

c) Calculate the overall frequency for “frequent binge drinking”.

Which direction % are you calculating?

3135/8198=38.24%

Calculate the frequency for “frequent binge drinking” for those who have trouble with security

398/469=84.86%

Calculate the frequency for “frequent binge drinking” for those who DO NOT have trouble with security.

2737/7729=35.41%
What conclusions can you draw from the above percentages?

High binge drinking most of the time correlates to higher chances of trouble with security.
12.7% of people who frequently binge drink have trouble with security compared to 1.4% of
people who never binge drink but have trouble with security.

Which percentage is best to use in this particular case? Why?

the frequency for “trouble with security” for those who never binge drink and calculate the
frequency for “trouble with security” for those who frequently binge drink.

Before getting the computer to draw a graph, it’s important to know what it should look like!

What type of graph should you do to display the data of trouble with security and binge drinking?

Line graph or bar chart

Sketch what the graph should look like.

Now, enter the required data into Microsoft Excel (or Google Sheets), and create the graph yourself.

Binge drinking effect on trouble with security

9000
8000
7000
Number of people

6000
5000
4000
3000
2000
1000
0
Yes No Total
Trouble with security (outcome)

Binge drinking (ex...

Activity 2: Manually draw a histogram and boxplot from provided

The haemoglobin levels in the blood of 45 patients in hospital is below:

9.1 10.1 10.7 10.7 10.9 11.3 11.3 11.4 11.4 11.4 11.6 11.8 12.0 12.1 12.3

12.4 12.7 12.9 13.1 13.2 13.4 13.5 13.5 13.6 13.7 13.8 13.8 14.0 14.2 14.2

14.2 14.6 14.6 14.8 14.8 15.0 15.0 15.0 15.1 15.4 15.6 15.7 16.2 16.3 16.9

Data source: Haemoglobin data from p. 16, Advanced Level Mathematics: Statistics 1. Steve Dobbs and Jane Miller.
Cambridge University Press. 2002

Task 1: Draw a histogram

a) Create a frequency distribution table with 8 intervals. (Hint: Lecture slides or eBook)

● What is a suitable width of each interval?

b) Draw a Histogram of the data by hand.

Frequency distribution table Draw your histogram here:

Haemoglobin (g/100 ml) range Number

9-9.99 1

10-10.99 4

11-11.99 7

12-12.99 6

13-13.99 9

14-14.99 8

15-15.99 7

16-16.99 3
Task 2: Draw a boxplot
The haemoglobin levels in the blood of 45 patients in hospital is below:

9.1 10.1 10.7 10.7 10.9 11.3 11.3 11.4 11.4 11.4 11.6 11.8 12.0 12.1 12.3

12.4 12.7 12.9 13.1 13.2 13.4 13.5 13.5 13.6 13.7 13.8 13.8 14.0 14.2 14.2

14.2 14.6 14.6 14.8 14.8 15.0 15.0 15.0 15.1 15.4 15.6 15.7 16.2 16.3 16.9

Data source: Haemoglobin data from p. 16, Advanced Level Mathematics: Statistics 1. Steve Dobbs and Jane Miller.
Cambridge University Press. 2002

What is the:

● Minimum value? ___________9.1

● Maximum value? 16.9

● Median value? ______13.5______

● Q1? _________11.8___________ How do you find Q1?

● Q3? ___________14.8_________ How do you find Q3?

Copy-and-paste the data into Excel, and use the QUARTILE.INC() command to check your answers.

Manually draw the Box Plot. Clearly label what each section of the box plot is.

8
9
Small group activity: Identify graphs
For each graph below, identify what type of graph (Bar chart, etc) it is.

10
Consolidation Activity 3: Combining epidemiology & biostatistics = AIHW data

Non-fatal % and fatal % for the various “Disease of injury” categories in 2015.

Males Females Persons

Non-fatal Fatal Non-fatal Fatal Non-fatal Fatal
All injury 14.5 % 85.5 % 24.7 % 75.3 % 17.7 % 82.3 %
Drowning 1.6 % 98.4 % 2.6 % 97.4 % 1.8 % 98.2 %
Falls 50.6 % 49.4 % 59.4 % 40.6 % 55.1 % 44.9 %
Fires, burns & scalds 50.7 % 49.3 % 60.8 % 39.2 % 54.6 % 45.4 %
Homicide & violence 24.4 % 75.6 % 18.4 % 81.6 % 22.4 % 77.6 %
Poisoning 0.8 % 99.2 % 1.3 % 98.7 % 0.9 % 99.1 %
Road traffic accidents – 8.5 % 91.5 % 13.8 % 86.2 % 10.3 % 89.7 %
motor vehicle
Road traffic accidents – 19.5 % 80.5 % 23.6 % 76.4 % 19.8 % 80.2 %
motorbike
Road traffic accident – 49.1 % 50.9 % 76.6 % 23.4 % 52.7 % 47.3 %
cyclist
Road traffic accident – 12.3 % 87.7 % 19.2 % 80.8 % 14.5 % 85.5 %
pedestrian
Suicide & self-inflicted 0.5 % 99.5 % 2.0 % 98.0 % 0.9 % 99.1 %
injuries

Source: https://www.aihw.gov.au/reports/burden-of-disease/abds-2015-interactive-data-disease-burden/contents/
fatal-vs-non-fatal-burden

● Which categories have a higher Fatal % for Males? What about for Females?

11
12
● Look at the graphs below, and comment on the distribution of males, females and persons by age.

● How does the age distribution of “drowning” and “falls” compare?

13
● How does the age distribution for “motor vehicle” and “motorbike” compare?

14
15