1. PUBH 601 Concepts and Methods of Biostatistics
3-Graphical summaries
Manar Elhassan, PhD
Department of Public Health
College of Health Sciences Qatar
University
Fall 2022
PUBH 601 Concepts and Methods of Biostatistics. Fall 2022 1
Department of Public Health
2. Objectives of today’s class
• Bar chart and pie chart
• Stem and leaf plot
• Histogram
• Frequency polygon and curve
• Box plot
• Scatter plot
PUBH 601 Concepts and Methods of Biostatistics. Fall 2022 2
Department of Public Health
Recognizing and avoiding misuses of graphical summaries
Graphical summaries
3. Student Learning Outcomes of the Course
At the end of this course, students will be able to:
• Select, construct and interpret appropriate numerical
data used in graphical format.
PUBH 601 Concepts and Methods of Biostatistics. Fall 2022 3
Department of Public Health
4. PUBH 601 Concepts and Methods of Biostatistics. Fall 2022
Department of Public Health 4
Statistical
methodology
Descriptive
Describe the
observations/
data/ sample
Inferential
Assess strength of
evidence
for/against a
hypothesis
Organization and
summarization of
sample data
• Tables
• Summary
measures
• Graphs
5. Once we obtained our sample, we would like to summarize it.
PUBH 601 Concepts and Methods of Biostatistics. Fall 2022 5
Department of Public Health
Depending on the type of the data and the dimension there are
different methods of summarizing the data.
Types
•Numerical
•Categorical
Dimensions
•Univariate
•Bivariate
•Multivariable
6. Three steps to summarize Data
• Classify the sample by types and dimension
• Use appropriate numerical summaries
• Use appropriate visual summaries (graphs or
visuallizations)
PUBH 601 Concepts and Methods of Biostatistics. Fall 2022
7. Bar chart and pie chart
Categorical data
PUBH 601 Concepts and Methods of Biostatistics. Fall 2022 7
Department of Public Health
8. Nominal data
Pie and bar chart
Bars are
separated
by spaces
PUBH 601 Concepts and Methods of Biostatistics. Fall 2022
13. Bar chart
PUBH 601 Concepts and Methods of Biostatistics. Fall 2022
How many variables??
14. When to use a bar chart or a pie chart
Pie Chart Best to use when you are trying to
compare parts of a whole. They do not
show changes over time
Bar Chart Used to compare things between different
groups or to track changes over time
PUBH 601 Concepts and Methods of Biostatistics. Fall 2022
15. The majority of
participants are in the
lower two categories of
the distribution
Ordinal data
PUBH 601 Concepts and Methods of Biostatistics. Fall 2022
16. Stem and leaf plot
PUBH 601 Concepts and Methods of Biostatistics. Fall 2022 16
Department of Public Health
17. Stem and leaf plot (stemplot)
Excellent way to begin an analysis where you can exam the
shape, location and spread of the distribution
PUBH 601 Concepts and Methods of Biostatistics. Fall 2022
original distrbuation
this kind give(stem& leaf)
18. Stem and leaf plot (stemplot)
A dentist created the following stem-and-leaf plot showing the
number of teeth each patient have:
How many child have fewer than 13 teeth?
How many children have exactly 25 teeth?
Go to https://www.socrative.com/
Room name: PUBH
Or install the socrative app in your mobile devices
PUBH 601 Concepts and Methods of Biostatistics. Fall 2022
21. Continuousdata-histogram
1
3
14
19
14 13
5
1
• How may patients had
haemolobin levels less than 10
g/100ml?
• How may patients had
haemolobin levels between 13
and 15 g/100ml?
• What percentage of patients had
at least 14 g/100ml haemolobin
level?
PUBH 601 Concepts and Methods of Biostatistics. Fall 2022
29. Example
Outliers or extreme values can also be assessed graphically with
box-whisker plots.
There are a number of way to assess outliers. A popular one is the
Tukey Fences
Outliers are values
below Q1-1.5 IQR or
above Q3+1.5 IQR.
For DBP:
Q1=67, Q3=80, IQR=13
Q1-1.5 IQR = 47.5
Q3+1.5 IQR= 99.5
PUBH 601 Concepts and Methods of Biostatistics. Fall 2022
30. Box plots are very useful for comparing distributions
PUBH 601 Concepts and Methods of Biostatistics. Fall 2022
Department of Public Health 30
31. Box plots are very useful for comparing distributions
PUBH 601 Concepts and Methods of Biostatistics. Fall 2020
Department of Public Health 31
What is the
dimension of
variables in
this graph?
32. Box plots are very useful for comparing distributions
PUBH 601 Concepts and Methods of Biostatistics. Fall 2020
Department of Public Health 32
What type of
variables are
included in
this graph?
34. Figure 2shows the boxplot for the variable OB2014 on the continents Africa, America, Asia, Europe, and Oceania.
The highest concentration of countries with low OB2014 values is in Africa and Asia, while America, Europe, and
Oceania have the highest values. Note that there is no intersection between the boxplots for Europe and Oceania
and those of Africa and Asia, signifying a possible difference between the proportions of obese adults on these
continents.
PUBH 601 Concepts and Methods of Biostatistics. Fall 2022
How do you
interpret
this graph?
35. PUBH 601 Concepts and Methods of Biostatistics. Fall 2022
Go to: Reading Box Plots
Time: 5 minutes
Activity 3-1: Concept check
42. PUBH 601 Concepts and Methods of Biostatistics. Fall 2020 42
Department of Public Health
Activity 3-2: Learn through simulations
Learning outcome:
• Understand the effect of outliers on numerical and graphical
summaries
• Compare side-by-side boxplots
• Download 3-2 Learn through simulations from BB
• Go to https://istats.shinyapps.io/EDA_quantitative/
• Time: 30 minutes
43. PUBH 601 Concepts and Methods of Biostatistics. Fall 2020 43
Department of Public Health
Activity 3-3: Watch Stata tutorial
Create basic box plots using Stata
44. PUBH 601 Concepts and Methods of Biostatistics. Fall 2020 44
Department of Public Health
Activity 3-4: Practice producing and interpreting
plots using Stata
Dataset: 3-chol.dta found in BB
Demonstration: Change sex from string to numeric:
Data > Create or change data > Other variable-transformation commands > Encode
value labels from string variable
Now practice:
• Distribution of Chol1
• Distribution of gender
• What is the appropriate visual display for Chol1 and gender. Note Chol1
is continuous and gender is binary?
• Create a scatter plot Chol1 and gender
Graphics > Twoway graph (scatter)
45. PUBH 601 Concepts and Methods of Biostatistics. Fall 2020 45
Department of Public Health
Remember the exit slip
Week 3 Quiz
Homework