1

3.1 Lesson 6: Frequency Tables

When you look at a newspaper, a corporate annual report, a government study, or a journal article in your field of
study, you are likely to see statistical data displayed in graphical or table form.
There are two main reasons we display data in graphs and tables:
1. They convey information in an easy to read manner.
2. To highlight some specific piece of information from the results of the study.

The type of graph or table chosen to display data is dependent on the type of data.

A ____________________ ______________ is a table that displays either qualitative or quantitative data so that
it is easier to read than raw data. A basic frequency table has two basic columns:
 The list of categories or intervals(bins)
 The frequency (count) for each category, which is the number of data values in the category.

Frequency Table Example:

The Rocky Mountain Beverage Company wants feedback on its new product, Coral Cola, and sets up a taste test
with 20 people. Each individual is asked to rate the taste of the cola on a 5-point scale:

(bad taste) 1 2 3 4 5 (excellent taste)

The 20 ratings are as follows:

1 3 3 2 3 3 4 3 2 4
2 3 5 3 4 5 3 4 3 1

Construct a frequency table for these data.

Taste Scale Frequency

Total

Sometimes we might be interested in the fraction or percentage of some group or the total number below some
category. We use the relative frequency and the
cumulative frequency for these, respectively.

The ________________ _______________of any category is the proportion or percentage of the data values that
fall in that category:

relative frequency =

Relative frequency may be reported as a fraction, decimal, or percent. (We do not usually reduce the fractions.)

The _________________ ______________ of any category is the number of data values in that category and all
preceding categories.

1
 2

Relative Frequency Table Example:

Use the taste test data to create the relative frequency table and the cumulative frequency columns.

Taste Scale Frequency Relative Frequency Cumulative Frequency

Total

When it is impossible or impractical to have a category for every value in a numerical data set, we bin (or group)
the data into categories (bins), each covering a range of possible data values.

Frequency Table Example using bins:

For the 30 stocks of the Dow Jones Industrial Average, the table below shows the annual revenue (in billions of
dollars), the one-year total return, and the rank on the Fortune 500 list of largest U.S. companies. Create a
frequency table for the revenue.

 The revenue data range from $21.6 billion (McDonald’s) to $351.1 billion (Wal-Mart). We create bins
spanning a range from $0 to $400 billion. This covers the full range of the data, with extra room below the
lowest data value and above the highest data value.
 We give each bin a width of $50 billion so that we can span the $0 to $400 billion range with eight bins.
Also, the width of $50 billion is a convenient number that helps make the table easy to read.
 Because the data values are given to the nearest tenth (of a billion dollars), we also define the bins to the
nearest tenth so that they do not overlap. That is, bins go from $0 to $49.9 billion, from $50.0 to $99.9
billion, and so on.

There is nothing special about the bin groupings I chose. Choose breakdowns that make sense or let technology
choose these breakdowns for you. You will want a minimum of 5 bins (helps with graphing later) and probably a
max of 15-20 bins depending on how many pieces of data are in your dataset.
2
 3

Annual Revenue in billions of dollars Frequency

(number of
companies)
0 – 49.9
50 – 99.9
100 – 149.9
150 – 199.9
200 – 249.9
250 – 299.9
300 – 349.9
350 – 399.9
Total

3.1 Frequency Tables

Lesson 6 ICE

1. A guidebook for San Francisco list 2 one-star restaurants, 12 two-star restaurants, 18 three-star
restaurants, 6 four-star restaurants, and 2 five-star restaurants. Complete the frequency table for these
ratings.

Number of Stars Frequency Rel. Freq (%) Cum. Freq

1
2
3
4
5
Total

3
 4

2. The following data show the body temperatures (°F) of randomly selected subjects. Construct a frequency
table with seven classes: 96.9 – 97.2, 97.3 – 97.6, 97.7 – 98.0, and so on.

Temperature (°F) Frequency Tallies (if you need them)

96.9 – 97.2
97.3 – 97.6
97.7 – 98.0
98.1 – 98.4
98.5 – 98.8
98.9 – 99.2
99.3 – 99.7
Total

3.2 Lesson 7: Picturing Data

In addition to using a frequency table to make data easier to read, sometimes we want more information from
our data on a quick glance. We do this by displaying our data in a graph. A graph quickly shows us the distribution
of the data, or how it’s spread out over all possible values.

The type of graph chosen to display data are dependent on the type of data.
And, all good graphs include the following traits:
1. A descriptive title or caption.
2. A vertical scale and label.
3. A horizontal scale and label.
4. A legend if multiple data sets are displayed on one graph.

Bar graphs, Pareto charts, and pie charts are all used to display
___________________ ___________. They may represent frequency or relative frequency (%). The graphs are
used to highlight different important traits of the data we wish a reader to notice.

A bar graph is one of the simplest ways to picture a distribution. Each bar represents the frequency or relative
frequency of one category: the higher the frequency, the longer the bar. The bars can be either vertical or
horizontal.
4
 5

Bar Graph Example:

Let’s create a bar graph from our taste test example of the last section. Recall, the taste scale look like numbers
but are actually qualitative data and represent bad taste (1) to excellent taste (5).

Taste Frequency
Scale
1 2
2 3
3 9
4 4
5 2
Total

Bar Graph: When creating a bar graph, we leave space between the bars and make sure to use good graphing
skills.

A Pareto chart is just a specialized form of a bar graph. The bars are arranged in decreasing order.

Pareto Example:

5
 6

Pie charts may be used to show frequency or relative frequency but they highlight each category with respect to
the whole where a bar graph highlights each category with respect to every other category. Pie charts are used
almost exclusively for ___________________ __________.
A pie chart is a circle divided so that each wedge represents the relative frequency of a particular category—even
if we are representing the raw counts on the circle. The wedge size is proportional to the relative frequency. The
entire pie represents the total relative frequency of 100%. Notice the example below uses raw counts but each car
type wedge size is with respect to the whole.

Types of Cars on American Roads

4

15
US

Sweden
40
Japan

10 France

Germany

Italy

2
30

A dot plot is a variation on a bar graph in which we use dots rather than bars to represent the frequencies. Each
dot represents one data value. The dot plot may be used to represent _________________ ________ or
_________________ _______.

Dot plot Example:

Now, let’s create a dot plot from our taste test example of the last section.

Dot plot:

6
7

7
 8

Dot plot Example of Numerical Data:

An independent Etsy consultant has tracked the weights of her orders for the first few months to determine if she
should be shipping by weight or by prepaid box size. She has determined that more than 25% of her shipments
must be over 10 ounces for the prepaid boxes to be cost efficient.

Are 25% or more of her shipments above the 10 ounce cut-off?

A graph in which the bars have a natural order and the bar widths have specific meaning, is called a histogram.
The bars in a histogram touch each other because there are no gaps between the categories.

Histogram Example:
Let’s go back to our example from the last lesson. For the 30 stocks of the Dow Jones Industrial Average, the table
below shows the annual revenue (in billions of dollars), the one-year total return, and the rank on the Fortune 500
list of largest U.S. companies. Create a histogram for the revenue. Use the frequency table we created to do so.
Annual Revenue in billions of dollars Frequency (number of
companies)
0 – 49.9 13
50 – 99.9 10
100 – 149.9 3
150 – 199.9 1
200 – 249.9 1
250 – 299.9 0
300 – 349.9 1
350 – 399.9 1
Total 30

8
 9

A line graph represents the same data as we see in a histogram, but instead of using bars to represent the
frequency, we use a dot and connect the dots with a line. It may also be used to represent time data (years,
months,…), when we do, we typically call this a ________________________.

Line Graph Example:

Using the same data we used to make a histogram, make the equivalent line graph.

The stem-and-leaf plot (or stemplot) looks somewhat like a histogram turned sideways. It is used to graphically
depict___________________ _________. It pictures the data of a sample using the actual digits that make up
the data values. Each numerical data is divided into two parts: The leading digit(s) becomes the stem, and the
trailing digit(s) becomes the leaf. The scale MUST be listed on the display.

Stem & Leaf Example:

A city police officer, using radar, checked the speed of cars as they were traveling down the main street in town.
Construct a stem-and-leaf plot for this data:
16 19 24 26 27 31 32 33 33 35
36 36 37 38 39 39 40 41 49 55

Key:

|
|
|
|
|

9
 10

A quick summary of which graphs and tables are used with each data type:

Qualitative (Categorical) Data Quantitative (Numerical) Data

3.2 Picturing Data

Lesson 7 ICE

Problems 1 – 3 describe data sets but do not give actual data. For each, state the type of graphic that you
believe would be most appropriate for displaying the data, if they were available. Explain your choice.

1. Incomes of college graduates with a BS in nursing.

2. The political party affiliations of 1000 survey subjects.

3. Ages of people who died in car crashes last year from the town of Idontknow, Alaska.

4. The Kappa Iota Sigma Fraternity polled its members on the weekend party theme. The vote was as
follows: six for toga, four for hayride, eight for school spirit, and two for masquerade.
A. Summarize these data in a frequency table.

Party Theme Frequency

Toga
Hayride
School Spirit
Masquerade
Total

10
 11

B. Display the vote count in a Pareto chart. Use good graphing skills.

11
 12

5. The accompanying data represent the percentage of recent high school graduates (graduated within 12
months before the given year-end) who enrolled in college in the fall. Choose the appropriate time-series
(line) plot to match the data and comment on any trends.

Year Percent Enrolled Year Percent Enrolled

1989 59.6 1998 65.6
1990 60.1 1999 62.9
1991 62.3 2000 63.3
1992 61.7 2001 61.5
1993 62.4 2002 65.2
1994 61.7 2003 63.9
1995 61.7 2004 66.7
1996 65 2005 68.4
1997 67 1998 65.6

A. B.

C.

12
 13

6. You constructed a frequency table for a list of body temperatures of randomly selected subjects in the last
ICE (3.1 Lesson 6). The histogram below represents the data from that frequency table. Use the histogram
to answer the questions below.

Human Temperature °F
18
16
14
12
Frequency

10
8
6
4
2
0
96.9 – 97.2 97.3 – 97.6 97.7 – 98.0 98.1 – 98.4 98.5 – 98.8 98.9 – 99.2 99.3 – 99.7
Temperature

A. How many subjects are represented in the histogram?

B. How many subjects have temperatures above 98.8°F?

3.3 Graphics in the Media

Lesson 8

The graphs we have covered so far are basic. Graphs can be much more complicated and portray a great deal of
information for multiple variables at once. Be wary, though, good graphs are clear and easy to read and interpret;
graphs with too much information are often used to distract from bad data or data that does not say what the
author wants to show.

Some examples of more complicated graphs are:

 Multiple bar graphs which is a bar graph that shows a comparison between two or more qualitative sets
of data.
 Multiple line graphs show a comparison of numerical data for different categorical sets of data.
 A stack plot shows different sets of qualitative data that represent either numerical or qualitative data in
each set (similar to the two above mentioned graphs) but set with a vertical stack. Color is used to
separate the stacks.
o Data can be stacked in a bar graph
o Data can be stacked in a line graph

Examples of each graph can be seen below. Let’s look at each and answer the associated questions.

13
 14

Example Triple Bar Graph:

1. What variables are represented in the graph and are they numerical or qualitative? Which is/are the
explanatory and response variables?

2. Is there a noticeable trend?

14
 15

Example Multiple Line Graph:

1. What variables are depicted in the graph and are they numerical or categorical? Which is/are the
explanatory and response variables?

2. If you had money to invest on July 7, where should you have invested according to this graph?

15
 16

Example Stack Plot:

1. What variables are depicted in the graph and are they numerical or categorical?

2. Is there a noticeable trend?

3. Why might we prefer a different type of graph here and what type of graph would that be?

16
 17

3.3 Graphics in the Media

Lesson 8 ICE

1. The stack plot in the figure below shows the number of male and female higher education students for
different years. Projections are from the US National Center for Education Statistics. Discuss the trends
revealed on this graphic.

2. The graph below shows home prices in different regions of the US. Note that data were not adjusted for
the effects of inflation. Describe the general trends that apply to the home price data for all regions.

17
 18

3. Consider the display of median salaries of males and females in recent years? What does the graph
display?

18
 19

4. The stack plot below shows Congressional Budget Office data for actual (through 2011) and projected
spending on entitlement programs though 2085 as percentages of the gross domestic product.
A. Interpret the graph.

B. When will entitlement spending overtake tax revenue?

3.4 Lesson 9: Misleading Graphs

Graphs provide us with a method to quickly display data so we can quickly interpret what that data tells us. These
same graphs can be used to mislead—either intentionally or unintentionally.

Some graphs are inherently misleading in the way they are made (pictographs).

Pictograph Example:
In this pictograph, the lengths of the bars or pictures represent the data, but our eyes tend to focus on the areas
of the bars. The right bar is supposed to show that a dollar in 2006 was worth 41% as much as a dollar in 1980. Its
length is indeed 41% of that of the left bar, but its area is much smaller in comparison (about 17% of the area of
the left bar).

19
 20

Pictograph Example:
This is another example of a pictograph.

Why is this graph misleading?

A second very common misleading graph, is that graph where the numerical scale (frequency, percent, percent
change,…) does not start at zero. This type of misleading graph is used to distort the difference in categories or
over time.

Example Different Scale:

Women as a Percentage of All College Students

20
 21

Each of these graphs represent the same data.

1. Visually which one looks like it shows that women were a huge proportion of college students near 2000?

2. What is the total change in percent of women from 1910 to 2000?

It is acceptable to use a graph like graph (a), but when we do not start our scale at zero, we should identify this to
the reader clearly with a notation of a scale break (a zigzag); We do so by starting the scale at zero, show a zigzag
and then jump to 30 to indicate the change in scale.

3.4 Misleading Graphs

Lesson 9 ICE

1. The figure below depicts the amounts of daily oil consumption in the US and Japan.

A. Does the illustration accurately depict the data? Why or why not?

B. Construct a bar graph that represents the data that is not misleading.

21
 22

2. The figure below shows the numbers of US adoptions from China in 2005 and 2010. What is wrong with
this graph? How should it be remade to be fair and objective?

22