Descriptive Methods
Summarizing qualitative data
Learning Outcomes
By the end of this lesson you must be able to:
•Summarize categorical data into a frequency table
and cross tabulation table
•Interpret the findings from a categorical frequency
table and a cross tabulation table
•Construct and interpret bar and pie charts
Qualitative Data
• Can easily be described and summarized by;
– frequency distribution tables,
– bar graphs,
– pie charts
– Cross tabulation
Need to Summarize data
• Data can be easily understood when it is summarized into
appropriate tables and then display graphically: Tables and
graphs convey meaning quickly
• Summary tables and graphs are commonly used to convey
statistical results.
• Summary tables and graphs can be used to summarize a single
random variable (most preferred clinic by patients) .
• Summary of tables and graphs are also used to examine the
relationship between two random variables ( e.g. between
smoking and lung cancer patterns).
Frequency
• Frequency refers to the number of times a certain
event/data point appears in a distribution.
• Frequency may be determined by counting the tally
marks for each class or category.
Frequency Distribution
• A frequency distribution is the representation of data in a
table format. The table consists of the frequencies of each
data point or event.
• The frequency distribution should contain the columns for
frequency, relative frequency and cumulative frequency.
– Relative Frequency refers to the proportion of frequency
divided by total frequency.
– Cumulative frequency is the accumulated relative frequencies
(or accumulated frequencies) of the values of the variable as
they increase.
Frequency distribution for Qualitative data
Assume you have the qualitative data like the following (whereby F=Female and
M=male); F M F F F F M F M M M F F M M M F M F M F M F M F F M M F F
These data can be represented in a table in the frequency distribution by tallying
the frequency and counting them,
Sex Tallying Frequency Relative Cumulative
Frequency frequency
M
F
Total
Note: The table should be named(title) and the rows and column labelled. It
should contain the units of measurements used.
Frequency distribution for Qualitative data
Assume you have the qualitative data like the following (whereby F=Female and
M=male); F M F F F F M F M M M F F M M M F M F M F M F M F F M M F F
These data can be represented in a table in the frequency distribution by tallying
the frequency and counting them,
Note: The table should be named(title) and the rows and column labelled. It
should contain the units of measurements used.
Pie chart
Pie Chart
The presentation of qualitative data in a circle which is
apportioned to different variables
Assume we have the following data of patients weights
summarized in a frequency distribution below;
BMI Frequency Relative frequency(%)
Normal 30 15%
Mild 10 5%
Overweight 40 20%
Obese 100 50%
Severe obese 20 10%
Total 200 100%
….Pie charts
Bar chart
• This is a better way of graphically presenting qualitative data than pie chart,
especially when there are many numbers of observations.
• Assume we would like to represent graphically the following qualitative data
of the proportions of contraceptives used in a particular country.
Abstinence 2%, Oral contraceptive 34%, Depo Provera 10%,
Loop 5%, Spermicides 2%, Condoms 50%, Vasectomy 1%
Hysterectomy 2% and Norplant 1%
The pie chart can be used for these data, however it will be too congested and
lose meaning. It is therefore advisable to present many qualitative
observations in a bar chart.
Constructing a bar chart
• To construct a bar chart, draw a horizontal axis (x-axis) to
represent the categories and a vertical axis (y-axis) scaled to show
either the frequency counts or percentages of each category.
• Construct vertical bars for each category to the height of its
frequency count (or percentage) on the y-axis.
• Use space between bars….This imply there’s no continuity of your
dataset which is the essential meaning of qualitative data AND the
width of the bars should be equal to each other.
• Use limited numbers of bars for easy comparison.
Note the following…
• The sum of the heights across the bars must equal the sample size(
or 100%) .
• The bars must be of equal width to avoid distorting a category
importance.
• The order of bars on x- axis and width of bars does not convey any
importance of category.
• Your data may be arranged into categories in either descending or
ascending orders.
…Bar chart
Two categorical variables: Cross tabulation
• Also called a contingency table (two way or 2x2 table)
• summarizes the joint responses of two categorical
variables.
• The table shows the number (and or percentage) of
observations that jointly belong to each combination of
categories of the two categorical variables.
…Cross tabulation
• It summarizes how the frequencies of one
categorical variable are associated with the
categories of another.
• It can further be displayed graphically either as a
stacked bar chart ( also called a component bar
chart ) or multiple bar chart.
Example 2x2
30% of 200 passengers who came from China by Ethiopian airways were
females, and among them 10 had symptoms for corona virus compared to
90 males who were asymptomatic.
What is the;
1)The number of number of patients who were asymptomatic?
2) Percentage of females who are suspected cases?
Example….2x2
Outcome
Yes No Total
??
Row totals
Independent
??
Predictor/
variables
Total
Grand total
Column totals
Example…..
1.Total number of
Corona virus asymptomatic= 140
Sex symptomatic No symptoms Total
2.Percentage of suspected
Male 50 90 140 cases= 10/60 x 100=16.7%
Female 10 50 60
Total 60 140 200
Multiple Bar- Chart
