Frequency Distributions and

graphs
Frequency Distributions and Their Graph
 

OVERVIEW
Today we are constantly bombarded with visual images.  We spend lots of time
watching television, playing video games, or working at a computer monitor. These
technologies are advancing rapidly, making the images sharper and more attractive to
our eyes. Flat screens, high-resolution monitors, and HD televisions have made
significant improvements over the past few years. These types of upgrades are not
limited to video technology, but have also become an important part of the way
businesses communicate with customers and employees.  Presentations and reports
are expected to include high-quality graphs and charts that effectively transform data
into easy to view information. While the written word is still vital, words become even
more powerful when coupled with an effective visual illustration of data.  Let's take a
look back in history and how displaying data has played a positive role!  Let's read

about Florence Nightingale.  Florence Nightingale introduction.pdf  Download

Florence Nightingale introduction.pdf
 Upon graduation you will find yourself on both ends of the data analysis business. On
the one hand, regardless of what you end up doing for a career, you will almost certainly
be involved in preparing reports and making presentations requiring the use of the
visual descriptive statistical tools. You will be on the “do it” end of the data analysis
process.  Therefore, you need to know how to use these statistical tools.  On the other
hand, you will also find yourself reading reports or listening to presentations that others
have made.  In many instances, you will be required to make important decisions, or
reach conclusions, based on the information in those reports or presentations.  So, you
will be on the “use it” end of the data analysis process.  
Charts and graphs are not just tools used internally by businesses. Many
advertisements will use graphs and charts effectively to convey their message. Virtually
every Newspaper and Magazine contains different graphs, charts, or tables that display
data in an informative way.  You will find yourself as both a producer and a consumer of
the descriptive statistical techniques known as graphs, charts, and tables.

let's get crazy with graphs.mp4  Download let's get crazy with graphs.mp4Play
media comment.
In This Module you will:

 Construct a frequency distribution table including limits, midpoints, relative

frequencies, cumulative frequencies and boundaries
 Construct frequency histograms, relative frequency histograms, and ogives

Explorations and Learning:

 Print your note guide before you get started: Module 5 note guide.docx   Download
Module 5 note guide.docx                                                                

Frequency Distribution Tables

One way to organize data is to create frequency distribution table.  This table tells us
how many data points fit into certain categories or numerical intervals.  Here is an
example of a frequency distribution table describing how much the volunteers spent on
their vehicle.  Notice that the first class does not start with 1.  It is preferred to start your
frequency tables with the lowest data point!

Price Tally Frequenc Relative Cumulative

y frequency frequency

1,999.50 – 4,999.50 IIII    I 6 0.16 6

4,999.50 – 7,999.50 IIII      II 7 0.18 13

7,999.50 – IIII    IIII    13 0.34 26

10,999.50 III

10,999.50 – IIII 5 0.13 31

13,999.50
 3,999.50 – IIII 4 0.11 35
16,999.50

16,999.50 – I 1 0.03 36
19,999.50

 
Notice that 13 volunteers spent approximately between $8000 and $11,000 on their
vehicle.  18% of the people interviewed spent $5000-$8000 on their new ride.  We can
also tell by this table that the sample of this data was 36 people.
The class width of the classes would be calculated by: 4999.50-1999.50 = 3000
 
Let’s learn how to create these graphs!

Link (Links to an external site.)

Frequency Histograms
Frequency Histograms are a nice way to visually display quantitative data!  Take a
look at this histogram which describes the heights of Black Cherry Trees.

 
By looking at this histogram it is very easy to see that most common height of Black
Cherry trees is 75-80 feet.  Notice how the bars touch, not like in a bar graph.  The
reason is because the data is continuous meaning a tree could be 80.5 feet tall,
therefore there are no breaks between the bars.  Learn more regarding Frequency
Histograms by watching this video.
Link (Links to an external site.)

Misleading Histograms
Many graphs can be drawn to be very misleading to the public. Newspapers, Politicians
and Companies do this on accident or on purpose to illustrate evidence to the position
that they are in favor for.  Whether a company is comparing their product to a
competitor, a newspaper is comparing American schools to European schools, or a
politician describing the unemployment rates while he/she was in office; the viewers of
these graphs should have some basic understanding of the graphs so they can ask
intelligent questions regarding the data and/or identify possible bias. 
Please go to the link given below.  From the data pull down menu, choose data that
interests you.  Then use the slider underneath the graph to change the width of the
classes.  What happens when the class sizes are very small?  What happens when the
class sizes are very large?  Approximately how many classes give a reliable graph? 
Histogram Activity:  Link (Links to an external site.)

Histograms on the TI-83/84

You will have to be able to create Histograms and other graphs on your graphing
calculator.  The graphing calculator will be an awesome tool for us throughout this
course.  It will be difficult for you to complete many calculations without one.  Get your
calculator ready and follow along with this tutorial on how to create histograms on your
calculator.  Pause the video if you need extra time to enter data.  You will not be able to
enter all of the data that is on this video, but put in enough so that you can follow along. 
:-)

Link (Links to an external site.)

 

Ogive Graphs
An Ogive (pronounced O’jive) is a line graph that displays the cumulative relative
frequency of each class.  By analyzing this graph you can determine the percentage of
data points above and below a certain entry. Below is a great example describing the
weights of students in a particular high school. 
 From this graph it is easy to see that if you weighed around 150 pounds, 50% of the
students weigh more than you and 50% of the students weigh less than you do.
The Ogive always has the same type of shape.  Below is a picture of the common Ogive
shape.

This plot is commonly referred to as an ogive plot. The name ogive comes from a
particular pointed arch originally present in Arabic architecture and later incorporated in
Gothic cathedrals. Here is a picture of the Notre Dame Cathedral in Paris.  The red
curve illustrates the Ogive curve. If a data distribution is symmetric and mound shaped,
then its ogive plot will look just like the shape of one half of such an arch.
It is time to learn how to create the Ogive Graphs.  Watch the following video for
instruction. 

Link (Links to an external site.)

ACTIVE LEARNING
1. Discussion Post:  Misleading Graphs
2. Module 5 Active Learning Assessment:  Mod 5: Active Learning Assessment

3. Graphing Histograms on your graphing calculator Activity:  Mod 5: Graphing

Histograms on the graphing calculator
4. Module 5 Assessment:  Module 5 Assessment
5. Module 5 Feedback: Mod 5 Feedback
FREQUENCY DISTRIBUTIONS OF DATA

A frequency distribution lists categories (or intervals) and the number of times each occurs
(frequency).
A relative frequency distribution lists categories (or intervals) and the proportion of times each
occurs.

Relative Frequency=class  frequencytotal  frequency=fnRelative Frequency=class  frequencytotal  
frequency=fn

Frequency Distribution for Qualitative Data

Letter Grades

The letter grades in a statistics class are given as follows (raw data):

Letter Grades for 40 statistics students

C C A A C A A A B C A F C D D B A C

A A D C A C C B B B D D D C B C C C

Construct a frequency distribution table.

Construct a relative frequency distribution table.

a) How many students scored a B or better?

b) What percent of students scored a D?
c) What proportion of students scored a C or worse?

SolutionThe following is a frequency distribution table that helps summarize the data
in Table above.

Frequency distribution of the grades

Grade Number of Students, f

A 10

B 8
Frequency distribution of the grades

Grade Number of Students, f

C 14

D 6

F 2

Total n = 40

Relative Frequency Distribution of Letter Grades

Grade Frequency Relative Frequency

A 10 10/40 = 25%

B 8 8/40 = 20%

C 14 14/40 = 35%

D 6 6/40 = 15%

F 2 2/40 = 5%

Total n = 40 100%
a) 18 students scored a B or higher. (A or B)
b) 15% scored a D.
c) 55% scored a C or worse.

Graphical Representation of Qualitative DataBar Chart - consists of rectangular bars

showing the frequencies of each category of a distribution. The bar chart for the frequency
distribution in Table is seen below:

Bar Chart for Letter Grades

Pie Chart - a circle divided into sections reflecting the percentage of frequencies in each
category of the distribution. The Pie Chart for Table is seen below.

Pie Chart for Letter Grades

Frequency Distribution for Quantitative Data
Age of Patients

The following are ages of the 25 patients on the 6th floor of a hospital. Construct a frequency
distribution and a relative frequency distribution for these data: 12, 21, 51, 53, 26, 21, 21, 38, 55,
52, 38, 37, 40, 13, 38, 55, 31, 54, 48, 49, 54, 42, 56, 26, 54.

Age Class Frequency Relative Frequency

10 to under 20 2 2/25 = 8%

20 to under 30 5 5/25 = 20%

30 to under 40 5 5/25 = 20%

40 to under 50 4 4/25 = 16%

50 to under 60 9 9/25 = 36%

Total n = 25 100%

The numbers 10, 20, 30, 40, 50 are lower limits.

The numbers 20, 30, 40, 50, 60 are upper limits.
The difference between two consecutive lower (or upper) limits is called Class Width.
Class Width = 20 - 10 = 10 or 30 – 20 = 10 …
Note that it is more convenient to use classes with intervals widths that are multiples of 5.

The range that captures all the data values is partitioned into five non-overlapping intervals or classes.
The endpoints of each class are called class limits, lower and upper respectively, or class boundaries.

Class MidpointClass midpoints represent the value in the middle of the class intervals.

ClassMidpoint=LowerLimit+UpperLimit2Class Midpoint = {Lower Limit+Upper Limit} over {2}

Age Class Frequency Class Midpoint

10 to under 20 2 (10 + 20)/2 = 15

 Age Class Frequency Class Midpoint

20 to under 30 5 (20 + 30)/2 = 25

30 to under 40 5 (30 + 40)/2 = 35

40 to under 50 4 (40 + 50)/2 = 45

50 to under 60 9 (50 + 60)/2 = 55

Total n = 25

Steps to constructing classes for a frequency distribution

Step 1: Determine the appropriate number of classes
We can use the 2k>n2^k > n rule to determine the number of classes, where k equals the number
of classes and n equals the number of data points.
kk 3 4 5 6 7 8 9

2k2k 8 16 32 64 128 256 512

From previous example, n = 25 so we use 5 classes since 2  = 32 > 25.

5

For n = 30, we see that 2  = 32 > 30, so we still use 5 classes.

5

For n = 95, 2  = 128 > 95, so we use 7 classes.

7

Step 2: Determine the class width.

class width = highest value - lowest valuenumber of classesclass width = highest value - lowest 
valuenumber of classes

Round up the result to the next higher integer to ensure that all of the data will be included in the
frequency table. If the range of the data is large, a multiple of 5 is more convenient to use. From
the given data, highest value = 56, lowest value =12, number of classes = 5. So the class width is
(56 – 12)/5 = 44/5 = 8.8. Rounding this to the next integer gives 9. However, a class width of 10
is used because a multiple of 5 more convenient.
Step 3: Set the lower limit of the first class to the minimum value in the data set or a more
convenient value smaller than the minimum. Then add the class width to the first lower limit to
obtain the lower limit of the second class, then again for the third, and so on. The upper limits are
now obvious.

Years of Service

The following data represents the years of service of a sample of management employees:

35, 28, 32, 33.5, 31, 33, 34, 32, 30, 23.5, 16, 23, 19, 35, 21, 26, 31.5, 11.5, 28, 37.5
28, 22, 19, 23, 29, 29, 28, 19, 13, 15, 12, 30, 15, 15, 26, 20, 15, 19.5, 22, 32

Use the steps described above to construct a frequency distribution for the data.

SolutionStep 1: Finding the number of classes.

From the given data, n = 40. We use the 2  > n rule
k

Since 2 = 32 < 40 and 2  = 64 > 40, we use 6 classes.

5 6

Step 2: Finding the class width

Highest value = 37.5
Lowest Value = 11.5
Number of classes = 6
class width =37.5−11.56=4.33≈5  (rounded up to a convenient value)class width =37.5−11.56=
4.33≈5  (rounded up to a convenient value)

Step 3: Construct the table

Minimum value = 11.5 → Convenient Lower Limit of first class = 10.
Since the class width is 5, we have the following frequency distribution table.
Years Frequency

Total 40

10 to under 15 4

15 to under 20 8

20 to under 25 7

25 to under 30 8
 Years Frequency

30 to under 35 10

35 to under 40 3

Graphical Representation of Quantitative Data

HISTOGRAM: A histogram is a graph that displays interval data by using adjacent vertical bars
to represent the frequencies. There are no spaces between the bars of a histogram unless a class
frequency is zero. The frequency histogram for the Age data in Table is shown below.

FREQUENCY POLYGON: a line plot of class frequencies against their respective

class midpoints. The frequency polygon for the Age data in Table is shown below.
Cumulative Frequency Distribution

A cumulative frequency distribution shows the number (or percent) of observations that are
less than or equal to each class interval. The cumulative frequency for each interval is the sum of
the frequencies for that class and all previous classes. The following table shows the cumulative
frequencies and relative cumulative frequencies for the Age data in Table.

Class Interval Frequency Cumulative Frequency Cumulative Relative Frequency

10 to under 20 2 2 2/25 = 8%

20 to under 30 5 5+2=7 7/25 = 28%

30 to under 40 5 5 + 7 = 12 12/25 = 48%

40 to under 50 4 4 + 12 = 16 16/25 = 64%

50 to under 60 9 9 + 16 = 25 25/25 = 100%

Total n = 25

The table shows that 28% of the patients are under 30 years old while 64% are under 50 years
old.
Cumulative Frequency Graph (Ogive)An Ogive (pronounced oh-jive) is a graphical
representation of a cumulative frequency distribution. On an ogive, cumulative frequencies are
plotted against class upper limits. The ogive for the preceding table is shown below.

It can be seen on the ogive that about 40% of the patients are under 35 years old.

References

NHL 2015-2016 Standings (League). Available online at https://www.nhl.com/standings

(accessed September 2, 2016).

Homework
Lost Games

Canadian teams in the NHL performed poorly during the 2015-2016 season. No Canadian team
made the playoffs that season. The bar graph below shows the number of regulation losses each
team had.
a) Which team lost the highest number of games in regulation?
b) Which team lost the least number of games in regulation?
c) How many games were lost in total by the teams?
d) What is the relative frequency of the games lost by Calgary?
e) Which of the Canadian teams performed best during the 2015-2016 season?

a) Edmonton
b) Ottawa
c) 42+39+38+38+40+35+43 = 275
d) 40/275 = 14.55%
e) Ottawa

Music Type

Some psychologists believe that the type of music you enjoy listening to tells a lot about your
personality. The following pie chart shows the music preference of students at a postsecondary
institution.
a) What percent of students that preferred Pop or Jazz?
b) If 32 students preferred Reggae, how many students were sampled?
c) Suppose 600 students were sampled, how many preferred Country?
d) Suppose 90 students preferred Pop, how many preferred Rock?
e) What is the size of the angle (degrees) representing the Classical sector?

a) 18+9 = 27
b) 32/0.08 = 400
c) 0.12*600 = 72
d) 90/0.18*0.13 = 65
e) 0.11*360 = 39.6 o

Exercise Hours

It is generally believed that physical exercises help prevent excess weight gain, help prevent
diseases, and so on. To understand the state of physical activities of his students, a fitness
instructor collected data on the number of hours per week her students exercise. The data are
summarized in the histogram below.
a) The fitness instructor collected data from how many students?
b) How many class intervals were used?
c) Based on the 2  rule, are the number of classes appropriate? Explain.
k

d) What is the width of each class interval?

e) How many students exercise for less than 4 hours per week?
f) What percent of the students exercise between 4 and 10 hours per week?
g) What is the class midpoint for those that exercise from 6 to under 8 hours?

a) 18+15+22+13+10+8+4 = 90
b) 7
c) Yes. 2  = 128 > 90
7

d) 2
e) 18+15 = 33
f) (22+13+10)/90 = 50%
g) (6 + 8)/2 = 7 hours