www.rejinpaul.

com

LECTURE NOTES

ON

STATISTICS FOR MANAGEMENT

MBA I SEMESTER
IARE-R18

Ms. G. Joseph Mary

Assistant Professor

MASTER OF BUSINESS ADMINISTRATION

INSTITUTE OF AERONAUTICAL ENGINEERING

(Autonomous)
DUNDIGAL, HYDERABAD - 500 043
 www.rejinpaul.com
STATISTICS FOR MANAGEMENT

I Semester: MBA

Course Code Category Hours / Week Credits Maximum Marks

L T P C CIA SEE Total
CMBB05 Core
4 - - 4 30 70 100
Contact Classes: 45 Tutorial Classes: Nil Practical Classes: Nil Total Classes: 45
OBJECTIVES:
The course should enable the students to:
I. Understand the various statistical techniques and solve problems effectively in the statistics.
II. Analyze the different types of skewness and know about the coefficient variations of skewness.
III. Know the application of statistical measures of central tendency and also statistical measures of dispersion.
IV. Emphasize application of ANOVA, other non-parametric test and analyze the recent trends.
V. Apply the time series analysis and also trend analysis of data and also know its importance for solving the problems
arising.
COURSE OUTCOMES:
1. Recognize the significance, limitations, origin and development of statistics.
Acquire the knowledge about different managerial applications of statistics in various fields in modern
2.
times and analyze the use of computers in statistics.
3. Discuss various types of measures of central tendency and measures of dispersion..
4. Analyze the different types of coefficient of skewness and the coefficient of variation.
5. Understand the tabulation and classification of data to draw effective solutions for solving problems.
Demonstrate the diagrammatical and graphical representation of data by using different dimensional
6.
diagrams.
7. Examine the differences between uni-variate, bi variate and multi variate data.
8. Apply different types of small sample tests and techniques of ANOVA.
9. Analyze correlation analysis and different types of coefficient of correlation.
10. Describe the regression analysis, time series analysis and trend analysis of data.

UNIT-I INTRODUCTION TO STATISTICS Classes:08

Overview, origin and development and managerial applications of statistics and branches of the study, statistics and
computers, limitations of statistics.
UNIT -II MEASURES OF CENTRAL TENDENCY Classes:09
Mean, median, mode, geometric mean and harmonic mean, dispersion, range ,quartile deviation, mean deviation; co-efficient
of variation skewness: Karl pearson co-efficient of skewness, bowleys co-efficient of skewness, kelleys co-efficient of
skewness; theory and problems, discussion on direct and indirect methods of solving the problems.
UNIT -III TABULATION OF UNIVARIATE Classes:08
Bi variate and multi variate data, data classification and tabulation, diagrammatic and graphical representation of data.

One dimensional, two dimensional and three dimensional diagrams and graphs.
 www.rejinpaul.com
UNIT -IV SMALL SAMPLE TESTS Classes:10
T-Distribution: properties and applications, testing for one and two means, paired t-test; analysis of variance: one way and two
way ANOVA(with and without interaction),chi-square distribution: test for a specified population variance, test for goodness of
fit, test for independence of attributes; correlation analysis: scatter diagram, positive and negative correlation, limits for
coefficient of correlation, Karl Pearson‘s coefficient of correlation, spearman‘s rank correlation, concept of multiple and partial
correlation.
UNIT -V REGRESSION ANALYSIS Classes: 10
Concept, least square fit of a linear regression, two lines of regression, properties of regression coefficients; Time Series
Analysis: Components, models of time series additive, multiplicative and mixed models; Trend analysis: Free hand curve, semi
averages, moving averages, least square methods; Index numbers: introduction, characteristics and uses of index numbers, types
of index numbers, un weighted price indices, weighted price indices, tests of adequacy and consumer price indexes.
Text Books:
1. Levin R.I., Rubin S. David, ―Statistics for Management‖, Pearson, 7th Edition, 2015.
2. Beri, ―Business Statistics‖, TMH, 1st Edition, 2015.
3. Gupta S.C, ―Fundamentals of Statistics‖, HPH, 6th Edition, 2015.
Reference Books:
1. Levine , Stephan , krehbiel , Berenson, ―Statistics for Managers using Microsoft Excel‖, PHI, 1st Edition, 2015.
2. J. K Sharma, ―Business Statistics‖, Pearson Publications, 2nd Edition, 2015.

Web References:

1. https://aditya30702.files.wordpress.com/2012/07/statistics-for-managers-using-microsoft-excel- gnv64.pdf
2. http://www.nprcet.org/mba/document/First%20Semester/BA7102%20STATISTICS%20FOR%20
MANAGEMENT%20LT%20P%20C%203%201%200%204%20ODD.pdf
E-Text Books:

1. http://bookboon.com/en/statistics-and-mathematics-ebooks
2. http://www.ebay.com/bhp/statistics-for-managers-using-microsoft-excel
 www.rejinpaul.com

UNIT-I

INTRODUCTION TO STATISTICS

DEFINITION OF STATISTICS

Branch of mathematics concerned with collection, classification, analysis, and interpretation of

numerical facts, for drawing inferences on the basis of their quantifiable likelihood (probability).
Statistics can interpret aggregates of data too large to be intelligible by ordinary observation because
such data (unlike individual quantities) tend to behave in regular, predictable manner. It is
subdivided into descriptive statistics and inferential statistics.

HISTORY OF STATISTICS

The Word statistics have been derived from Latin word ―Status‖ or the Italian word ―Statista‖,
meaning of these words is ―Political State‖ or a Government. Shakespeare used a word Statist is his
drama Hamlet (1602). In the past, the statistics was used by rulers. The application of statistics was
very limited but rulers and kings needed information about lands, agriculture, commerce, population
of their states to assess their military potential, their wealth, taxation and other aspects of
government.

Gottfried Achenwall used the word statistik at a German University in 1749 which means that
political science of different countries. In 1771 W. Hooper (Englishman) used the word statistics in
his translation of Elements of Universal Erudition written by Baron B.F Bieford, in his book
statistics has been defined as the science that teaches us what is the political arrangement of all the
modern states of the known world. There is a big gap between the old statistics and the modern
statistics, but old statistics also used as a part of the present statistics.

During the 18th century the English writer have used the word statistics in their works, so statistics
has eveloped gradually during last few centuries. A lot of work has been done in the end of the
nineteenth century.
At the beginning of the 20th century, William S Gosset was developed the methods for
decision making based on small set of data. During the 20th century several statistician are
active in developing new methods, theories and application of statistics. Now these days the
availability of electronics computers is certainly a major factor in the modern development of
statistics.
 www.rejinpaul.com

Descriptive Statistics and Inferential Statistics

Every student of statistics should know about the different branches of statistics to correctly
understand statistics from a more holistic point of view. Often, the kind of job or work one is
involved in hides the other aspects of statistics, but it is very important to know the overall
idea behind statistical analysis to fully appreciate its importance and beauty.
The two main branches of statistics are descriptive statistics and inferential statistics. Both of
these are employed in scientific analysis of data and both are equally important for the
student of statistics.

Descriptive Statistics
Descriptive statistics deals with the presentation and collection of data. This is usually the
first part of a statistical analysis. It is usually not as simple as it sounds, and the statistician
needs to be aware of designing experiments, choosing the right focus group and avoid biases
that are so easy to creep into the experiment.
Different areas of study require different kinds of analysis using descriptive statistics. For
example, a physicist studying turbulence in the laboratory needs the average quantities that
vary over small intervals of time. The nature of this problem requires that physical quantities
be averaged from a host of data collected through the experiment.

MANAGERIAL APPLICATIONS OF STATISTICS

Statistics is a the mathematical science involving the collection, analysis and interpretation of data.
A number of specialties have evolved to apply statistical theory and methods to various disciplines.
Certain topics have "statistical" in their name but relate to manipulations of probability distributions
rather than to statistical analysis.
 Actuarial science is the discipline that applies mathematical and statistical methods
to assess risk in the insurance and finance industries.
 Astrostatistics is the discipline that applies statistical
analysis to the understanding of astronomical data.
 Biostatistics is a branch of biology that studies biological
phenomena and observations by means of statistical analysis, and
includes medical statistics.
 Business analytics is a rapidly developing business process that applies
statistical methods to data sets (often very large) to develop new insights
and understanding of business performance & opportunities
 www.rejinpaul.com

 Chemometrics is the science of relating measurements made on a chemical

system or process to the state of the system via application of mathematical
or statistical methods.
 Demography is the statistical study of all populations. It can be a very
general science that can be applied to any kind of dynamic population, that
is, one that changes over time or space.
 Econometrics is a branch of economics that applies statistical methods to
the empirical study of economic theories and relationships.
 Environmental statistics is the application of statistical methods to
environmental science. Weather, climate, air and water quality are
included, as are studies of plant and animal populations.
 Geostatistics is a branch of geography that deals with the analysis
of data from disciplines such as petroleum geology, hydrogeology,
hydrology, meteorology, oceanography, geochemistry, geography.
 Operations research (or Operational Research) is an interdisciplinary
branch of applied mathematics and formal science that uses methods
such as mathematical modeling, statistics, and algorithms to arrive at
optimal or near optimal solutions to complex problems.
 Population ecology is a sub-field of ecology that deals with the
dynamics of species populations and how these populations interact with
the environment.
 Quality control reviews the factors involved in manufacturing and
production; it can make use of statistical sampling of product items to aid
decisions in process control or in accepting deliveries.

 Quantitative psychology is the science of statistically explaining and

changing mental processes and behaviors in humans.
 Statistical finance, an area of econophysics, is an empirical attempt to
shift finance from its normative roots to a positivist framework using
exemplars from statistical physics with an emphasis on emergent or
collective properties of financial markets.
 Statistical mechanics is the application of probability theory, which includes
mathematical tools for dealing with large populations, to the field of
mechanics, which is concerned with the motion of particles or objects when
subjected to a force.
 www.rejinpaul.com

Statistical physics is one of the fundamental theories of physics, and uses

methods of probability theory in solving physical problems.

STATISTICS AND COMPUTERS

Crunch numbers to the nth degree — and see what happens. When you study computer
science and mathematics, you‗ll use algorithms and computational theory to create
mathematical models or define formulas that solve mathematical problems. In other words,
you'll design new tools that can predict the future.

The Computer Applications option gives students the flexibility to combine a

traditional computer science degree with a non-traditional field. Our state-of-the-
art labs for high- performance computing, networks and artificial intelligence will
give you experience with the tools you‗ll use in the field. Through labs, lectures
and projects, you‗ll also:
1. Investigate the computational limits of the algorithms and data structures
that support complex software systems
2. Develop new applications and tools in multi-disciplinary areas of science and
research
3. Explore opportunities for advanced computer modeling and
simulation

IMPORTANCE OF STATISTICS IN DIFFERENT FIELDS

Statistics plays a vital role in every fields of human activity. Statistics has important role in
determining the existing position of per capita income, unemployment, population growth
rate, housing, schooling medical facilities etc…in a country. Now statistics holds a central
position in almost every field like Industry, Commerce, Trade, Physics, Chemistry,
Economics, Mathematics, Biology, Botany, Psychology, Astronomy etc…, so application of
statistics is very wide. Now we discuss some important fields in which statistics is
commonly applied.

1. Business:
Statistics play an important role in business. A successful businessman
must be very quick and accurate in decision making. He knows that what
his customers wants, he should therefore, know what to produce and sell
and in what quantities. Statistics helps businessman to plan production
 www.rejinpaul.com

according to the taste of the costumers, the quality of the products can
also be checked more efficiently by using statistical methods. So all the
activities of the businessman based on statistical information. He can
make correct decision about the location of business, marketing of the products,
financial resources etc…
2. In Economics:
Statistics play an important role in economics. Economics largely
depends upon statistics. National income accounts are multipurpose
indicators for the economists and administrators. Statistical methods are
used for preparation of these accounts. In economics research statistical
methods are used for collecting and analysis the data and testing
hypothesis. The relationship between supply and demands is studies by
statistical methods, the imports and exports, the inflation rate, the per
capita income are the problems which require good knowledge of
statistics.

4. In Mathematics:
Statistical plays a central role in almost all natural and social sciences.
The methods of natural sciences are most reliable but conclusions draw
from them are only probable, because they are based on incomplete
evidence. Statistical helps in describing these measurements more
precisely. Statistics is branch of applied mathematics. The large number
of statistical methods like probability averages, dispersions, estimation
etc… is used in mathematics and different techniques of pure
mathematics like integration, differentiation and algebra are used in
statistics.
5. In Banking:
Statistics play an important role in banking. The banks make use of
statistics for a number of purposes. The banks work on the principle that
all the people who deposit their money with the banks do not withdraw it
at the same time. The bank earns profits out of these deposits by lending
to others on interest. The bankers use statistical approaches based on
probability to estimate the numbers of depositors and their claims for a
certain day.
 www.rejinpaul.com

6. In State Management (Administration):

Statistics is essential for a country. Different policies of the government
are based on statistics. Statistical data are now widely used in taking all
administrative decisions. Suppose if the government wants to revise the
pay scales of employees in view of an increase in the living cost,
statistical methods will be used to determine the rise in the cost of living.
Preparation of federal and provincial government budgets mainly depends
upon statistics because it helps in estimating the expected expenditures
and revenue from different sources. So statistics are the eyes of
administration of the state.
7. In Accounting and Auditing:
Accounting is impossible without exactness. But for decision making
purpose, so much precision is not essential the decision may be taken
on the basis of approximation, know as statistics. The correction of the
values of current asserts is made on the basis of the purchasing power
of money or the current value of it.In auditing sampling techniques are
commonly used. An auditor determines the sample size of the book to
be audited on the basis of error.
8. In Natural and Social Sciences:
Statistics plays a vital role in almost all the natural and social sciences.
Statistical methods are commonly used for analyzing the experiments
results, testing their significance in Biology, Physics, Chemistry,
Mathematics, Meteorology, Research chambers of commerce,
Sociology, Business, Public Administration, Communication and
Information Technology etc…

9. In Astronomy:
Astronomy is one of the oldest branches of statistical study; it deals
with the measurement of distance, sizes, masses and densities of
heavenly bodies by means of observations. During these
measurements errors are unavoidable so most probable measurements
are founded by using statistical methods.
 www.rejinpaul.com

UNIT-II
MEASURES OF CENTRAL TENDENCY

MEASURES OF CENTRAL TENDENCY:

The term central tendency refers to the "middle" value or perhaps a typical value of the data,
and is measured using the mean, median, or mode. Each of these measures is calculated
differently, and the one that is best to use depends upon the situation.
In the study of a population with respect to one in which we are interested we may get a large
number of observations. It is not possible to grasp any idea about the characteristic when we
look at all the observations. So it is better to get one number for one group. That number must
be a good representative one for all the observations to give a clear picture of that characteristic.
Such representative number can be a central value for all these observations. This central value
is called a measure of central tendency or an average or a measure of locations. There are five
averages. Among them mean, median and mode are called simple averages and the other two
averages geometric mean and harmonic mean are called special averages.

Arithmetic mean or mean

Arithmetic mean or simply the mean of a variable is defined as the sum of the
observations divided by the number of observations. It is denoted by the symbol If the
variable x assumes n values x1, x2 … xn then the mean is given by

The arithmetic mean is the most common measure of central tendency. It is

simply the sum of the numbers divided by the number of numbers. The symbol
"μ" is used for the mean of a population. The symbol "M" is used for the mean of
a sample. The formula for μ is shown below:
μ = ΣX/N
where ΣX is the sum of all the numbers in the population and
N is the number of numbers in the population.
The formula for M is essentially identical:
M = ΣX/N
where ΣX is the sum of all the numbers in the sample and
N is the number of numbers in the sample.
 www.rejinpaul.com

As an example, the mean of the numbers 1, 2, 3, 6, 8 is 20/5 = 4 regardless

of whether the numbers constitute the entire population or just a sample
from the population.

Example 1
Calculate the mean for pH levels of soil 6.8, 6.6, 5.2, 5.6, 5.8

Grouped Data
The mean for grouped data is obtained from the following formula:

Where x = the mid-point of individual class

f = the frequency of individual class
n = the sum of the frequencies or total frequencies in a sample.
Short-cut method

Where

A = any value in x
n = total frequency
c = width of the class interval
Example 2
Given the following frequency distribution, calculate the arithmetic mean
Marks : 64 63 62 61 60 59
Number of Students : 8 18 12 9 7 6
Solution
X F Fx D=x-A Fd
64 8 512 2 16
63 18 1134 1 18
62 12 744 0 0
61 9 549 -1 -9
60 7 420 -2 -14
59 6 354 -3 -18
60 3713 -7
 www.rejinpaul.com

Short-cut method

Here A = 62

Example 3
For the frequency distribution of seed yield of plot given in table, calculate the mean yield
per plot.
Yield per plot in(ing) 64.5- 84.5- 104.5- 124.5-
84.5 104.5 124.5 144.5

No of 3 5 7 20
plots

Solution
Yield ( in g) No of Plots (f) Mid X Fd

64.5-84.5 3 74.5 -1 -3
84.5-104.5 5 94.5 0 0
104.5-124.5 7 114.5 1 7
124.5-144.5 20 134.5 2 40
Total 35 44

A=94.5
The mean yield per plot is
Direct method:

= =119.64 gms

Shortcut method
 www.rejinpaul.com

Merits and demerits of Arithmetic mean

Merits
1. It is rigidly defined.
2. It is easy to understand and easy to calculate.
3. If the number of items is sufficiently large, it is more accurate and more reliable.
4. It is a calculated value and is not based on its position in the series.
5. It is possible to calculate even if some of the details of the data are lacking.
6. Of all averages, it is affected least by fluctuations of sampling.
7. It provides a good basis for comparison.
Demerits
1. It cannot be obtained by inspection nor located through a frequency graph.
2. It cannot be in the study of qualitative phenomena not capable of numerical measurement i.e.
Intelligence, beauty, honesty etc.,
3. It can ignore any single item only at the risk of losing its accuracy.
4. It is affected very much by extreme values.
5. It cannot be calculated for open-end classes.
6. It may lead to fallacious conclusions, if the details of the data from which it is computed are
not given.
 www.rejinpaul.com

Median
The median is the middle most item that divides the group into two equal parts, one part
comprising all values greater, and the other, all values less than that item.
Ungrouped or Raw data
Arrange the given values in the ascending order. If the number of values are odd, median
is the middle value.If the number of values are even, median is the mean of middle two values.
By formula

When n is odd, Median = Md

When n is even, Average of

Example 4
If the weights of sorghum ear heads are 45, 60,48,100,65 gms, calculate the median
Solution
Here n = 5
First arrange it in ascending order
45, 48, 60, 65, 100

Median =

= =60

Example 5
If the sorghum ear- heads are 5,48, 60, 65, 65, 100 gms, calculate the median.
Solution

Here n = 6
 www.rejinpaul.com

Grouped data
In a grouped distribution, values are associated with frequencies. Grouping can be in the
form of a discrete frequency distribution or a continuous frequency distribution. Whatever may
be the type of distribution, cumulative frequencies have to be calculated to know the total
number of items.
Cumulative frequency (cf)
Cumulative frequency of each class is the sum of the frequency of the class and the
frequencies of the pervious classes, ie adding the frequencies successively, so that the last
cumulative frequency gives the total number of items.
Discrete Series
Step1: Find cumulative frequencies.

Step3: See in the cumulative frequencies the value just greater than

Step4: Then the corresponding value of x is median.

Example 6
The following data pertaining to the number of insects per plant. Find median number of insects
per plant.
Number of insects per plant (x) 1 2 3 4 5 6 7 8 9 10 11 12
No. of plants(f) 1 3 5 6 10 13 9 5 3 2 2 1

Solution
Form the cumulative frequency table
x f cf
1 1 1
2 3 4
3 5 9
4 6 15
5 10 25
6 13 38
7 9 47
8 5 52
9 3 55
10 2 57
11 2 59
12 1 60
60
 www.rejinpaul.com

Median = size of

Here the number of observations is even. Therefore median = average of (n/2)th item and
(n/2+1)th item.

= (30th item +31st item) / 2 = (6+6)/2 = 6

Hence the median size is 6 insects per plant.

Continuous Series
The steps given below are followed for the calculation of median in continuous series.
Step1: Find cumulative frequencies.

Step2: Find

Step3: See in the cumulative frequency the value first greater than , Then the corresponding

class interval is called the Median class. Then apply the formula

Median =

where l = Lower limit of the medianal class

m = cumulative frequency preceding the medianal class
c = width of the class
f =frequency in the median class.
n=Total frequency.
 www.rejinpaul.com

Example 7
For the frequency distribution of weights of sorghum ear-heads given in table below. Calculate
the median.
Weights of ear No of ear Less than Cumulative
heads ( in g) heads (f) class frequency (m)
60-80 22 <80 22
80-100 38 <100 60
100-120 45 <120 105
120-140 35 <140 140
140-160 24 <160 164
Total 164
Solution

Median =

It lies between 60 and 105. Corresponding to 60 the less than class is 100 and corresponding to
105 the less than class is 120. Therefore the medianal class is 100-120. Its lower limit is 100.
Here 100, n=164 , f = 45 , c = 20, m =60

Median =

Merits of Median
1. Median is not influenced by extreme values because it is a positional average.
2. Median can be calculated in case of distribution with open-end intervals.
3. Median can be located even if the data are incomplete.

Demerits of Median
1. A slight change in the series may bring drastic change in median value.
2. In case of even number of items or continuous series, median is an estimated value
other than any value in the series.
3. It is not suitable for further mathematical treatment except its use in
calculating mean deviation.
4. It does not take into account all the observations.
 www.rejinpaul.com

Mode
The mode refers to that value in a distribution, which occur most frequently. It is an
actual value, which has the highest concentration of items in and around it. It shows the centre of
concentration of the frequency in around a given value. Therefore, where the purpose is to know
the point of the highest concentration it is preferred. It is, thus, a positional measure.
Its importance is very great in agriculture like to find typical height of a crop variety,
maximum source of irrigation in a region, maximum disease prone paddy variety. Thus the mode
is an important measure in case of qualitative data.

Computation of the mode Ungrouped or Raw Data

For ungrouped data or a series of individual observations, mode is often found by mere
inspection.
Example 8
Find the mode for the following seed weight
2 , 7, 10, 15, 10, 17, 8, 10, 2 gms
Mode = 10
In some cases the mode may be absent while in some cases there may be more than one mode.
Example 9
(1) 12, 10, 15, 24, 30 (no mode)
(2) 7, 10, 15, 12, 7, 14, 24, 10, 7, 20, 10
the modal values are 7 and 10 as both occur 3 times each.

Grouped Data
For Discrete distribution, see the highest frequency and corresponding value of x is mode.
Example:
Find the mode for the following
Weight of sorghum in No. of ear head(f)
gms (x)
50 4
65 6
75 16
80 8
95 7
100 4
 www.rejinpaul.com

Solution
The maximum frequency is 16. The corresponding x value is 75.
 mode = 75 gms.
Continuous distribution
Locate the highest frequency the class corresponding to that frequency is called the modal class.
Then apply the formula.

Mode =

Where = lower limit of the model class

= the frequency of the class preceding the model class
= the frequency of the class succeeding the model class

and c = class interval

Example 10
For the frequency distribution of weights of sorghum ear-heads given in table below. Calculate
the mode
Weights of ear No of ear
heads (g) heads (f)
60-80 22
80-100 38
100-120 45 f
120-140 35
140-160 20
Total 160
Solution

Mode =

Here 100, f = 45, c = 20, m =60, =38, =35

Mode =

= 109.589
 www.rejinpaul.com

Geometric mean
The geometric mean of a series containing n observations is the nth root of the product of the
values. If x1, x2…, xn are observations then

G.M=

Log GM =

GM = Antilog

For grouped data

GM = Antilog

GM is used in studies like bacterial growth, cell division, etc.

Example 11
If the weights of sorghum ear heads are 45, 60, 48,100, 65 gms. Find the Geometric mean for the
following data
Weight of ear Log x
head x (g)
45 1.653
60 1.778
48 1.681
100 2.000
65 1.813
Total 8.925
Solution
Here n = 5
GM = Antilog

= Antilog

= Antilog
 www.rejinpaul.com

Grouped Data

Example 12
Find the Geometric mean for the following
Weight of sorghum (x) No. of ear head(f)
50 4
65 6
75 16
80 8
95 7
100 4
Solution
Weight of No. of ear Log x f x log x
sorghum (x) head(f)
50 5 1.699 8.495
63 10 10.799 17.99
65 5 1.813 9.065
130 15 2.114 31.71
135 15 2.130 31.95
Total 50 9.555 99.21

Here n= 50

GM = Antilog

= Antilog

= Antilog 1.9842 = 96.43

 www.rejinpaul.com

Continuous distribution
Example 13
For the frequency distribution of weights of sorghum ear-heads given in table below.
Calculate the Geometric mean
Weights of ear No of ear
heads ( in g) heads (f)
60-80 22
80-100 38
100-120 45
120-140 35
140-160 20
Total 160
\

Solution
Weights of ear No of ear Mid x Log x f log x
heads ( in g) heads (f)
60-80 22 70 1.845 40
59
80-100 38 90 1.954 74.25
100-120 45 110 2.041 91.85
120-140 35 130 2.114 73.99
140-160 20 150 2.176 43.52
Total 160 324.2

Here n = 160

GM = Antilog

= Antilog

= Antilog
= 106.23
 www.rejinpaul.com

Harmonic mean (H.M)

Harmonic mean of a set of observations is defined as the reciprocal of the arithmetic
average of the reciprocal of the given values. If x1, x2…..xn are n observations,

For a frequency distribution

H.M is used when we are dealing with speed, rates, etc.

Example 13
From the given data 5, 10,17,24,30 calculate H.M.
X

5 0.2000
10 0.1000
17 0.0588
24 0.0417
30 0.4338
= 11.526

Example 14
Number of tomatoes per plant are given below. Calculate the harmonic mean.
Number of tomatoes per plant 20 21 22 23 24 25
Number of plants 4 2 7 1 3 1
Solution
Number of No of
tomatoes per plants(f)
plant (x)
20 4 0.0500 0.2000
21 2 0.0476 0.0952
22 7 0.0454 0.3178
23 1 0.0435 0.0435
24 3 0.0417 0.1251
25 1 0.0400 0.0400
18 0.8216
 www.rejinpaul.com

Merits of H.M
1. It is rigidly defined.
2. It is defined on all observations.
3. It is amenable to further algebraic treatment.
4. It is the most suitable average when it is desired to give greater weight to smaller observations
and less weight to the larger ones.
Demerits of H.M
1. It is not easily understood.
2. It is difficult to compute.
3. It is only a summary figure and may not be the actual item in the series
4. It gives greater importance to small items and is therefore, useful only when small items have
to be given greater weightage.
5. It is rarely used in grouped data.

Percentiles
The percentile values divide the distribution into 100 parts each containing 1 percent of
the cases. The xth percentile is that value below which x percent of values in the distribution fall.
It may be noted that the median is the 50th percentile.
For raw data, first arrange the n observations in increasing order. Then the x th percentile
is given by

For a frequency distribution the xth percentile is given by

Where
= lower limit of the percentile calss which contains the xth percentile value (x. n /100)
= cumulative frequency uotp
= frequency of the percentile class

C= class interval
N= total number of observations
 www.rejinpaul.com

Percentile for Raw Data or Ungrouped Data

Example 15
The following are the paddy yields (kg/plot) from 14 plots:
30,32,35,38,40.42,48,49,52,55,58,60,62,and 65 ( after arranging in ascending order). The
computation of 25th percentile (Q1) and 75th percentile (Q3) are given below:

= 3rd item + (4th item – 3rd item)

= 35 + (38-35)

= 35 + 3 = 37.25 kg

= 11th item + (12th item – 11th item)

= 55 +(58-55)

= 55 + 3 = 55.75 kg
 www.rejinpaul.com

Example 16
The frequency distribution of weights of 190 sorghum ear-heads are given below. Compute 25th
percentile and 75th percentile.
Weight of ear- No of ear
heads (in g) heads
40-60 6
60-80 28
80-100 35
100-120 55
120-140 30
140-160 15
160-180 12
180-200 9
Total 190

Solution
Weight of ear- No of ear heads Less than class Cumulative
heads (in g) frequency
40-60 6 < 60 6
60-80 28 < 80 34
80-100 35 <100 69 47.5
100-120 55 <120 124
142.5
120-140 30 <140 154
140-160 15 <160 169
160-180 12 <180 181
180-200 9 <200 190
Total 190

or P25, first find out , and for , and proceed as in the case of median.

For P25, we have = 47.5.

The value 47.5 lies between 34 and 69. Therefore, the percentile class is 80-100. Hence,

= 80 +7.71 or 87.71 g.
 www.rejinpaul.com

Quartiles
The quartiles divide the distribution in four parts. There are three quartiles. The second
quartile divides the distribution into two halves and therefore is the same as the median. The first
(lower).quartile (Q1) marks off the first one-fourth, the third (upper) quartile (Q3) marks off the
three-fourth. It may be noted that the second quartile is the value of the median and 50th
percentile.

Raw or ungrouped data

First arrange the given data in the increasing order and use the formula for Q1 and Q3
then quartile deviation, Q.D is given by

Where item and item

Example 18
Compute quartiles for the data given below (grains/panicles) 25, 18, 30, 8, 15, 5, 10, 35, 40, 45

Solution
5, 8, 10, 15, 18, 25, 30, 35, 40, 45

= (2.75)th item

= 2nd item + (3rd item – 2nd item)

= 8+ (10-8)

= 8+ x 2
 www.rejinpaul.com

= 8+1.5
= 9.5

= 3 x (2.75) th item
= (8.75)th item

= 8th item + (9th item – 8th item)

= 35+ (40-35)

= 35+1.25
= 36.25

Discrete Series
Step1: Find cumulative frequencies.

Step2: Find

Step3: See in the cumulative frequencies, the value just greater than , then the

corresponding value of x is Q1

Step4: Find

Step5: See in the cumulative frequencies, the value just greater than ,then the

corresponding value of x is Q3

Example 19
Compute quartiles for the data given bellow (insects/plant).
X 5 8 12 15 19 24 30
f 4 3 2 4 5 2 4
 www.rejinpaul.com

Solution
x f cf
5 4 4
8 3 7
12 2 9
15 4 13
19 5 18
24 2 20

=18.75th item Q1= 8; Q3=24

Continuous series
Step1: Find cumulative frequencies

Step2: Find

Step3: See in the cumulative frequencies, the value just greater than , then the

corresponding class interval is called first quartile class.

Step4: Find See in the cumulative frequencies the value just greater than then the

corresponding class interval is called 3rd quartile class. Then apply the respective formulae

Where l1 = lower limit of the first quartile class

f1 = frequency of the first quartile class

c1 = width of the first quartile class
m1 = c.f. preceding the first quartile class
 www.rejinpaul.com

l3 = 1ower limit of the 3rd quartile class f3 =

frequency of the 3rd quartile class c3 = width of
the 3rd quartile class
m3 = c.f. preceding the 3rd quartile class

Table 1 shows the number of touchdown (TD) passes thrown by each of the 31 teams in the
National Football League in the 2000 season. The mean number of touchdown passes thrown is
20.4516 as shown below.
μ = ΣX/N
= 634/31
= 20.4516

Table 1. Number of touchdown passes.

37 33 33 32 29 28 28 23 22 22 22 21 21 21 20 20 19 19 18 18 18 18 16 15 14 14 14 12 12 9 6

Although the arithmetic mean is not the only "mean" (there is also a geometric mean), it is by far
the most commonly used. Therefore, if the term "mean" is used without specifying whether it is
the arithmetic mean, the geometric mean, or some other mean, it is assumed to refer to the
arithmetic mean.

Median
The median is also a frequently used measure of central tendency. The median is the
midpoint of a distribution: the same number of scores is above the median as below it.
For the data in Table 1, there are 31 scores. The 16th highest score (which equals 20) is
the median because there are 15 scores below the 16th score and 15 scores above the
16th score. The median can also be thought of as the 50th percentile.

Computation of the Median

When there is an odd number of numbers, the median is simply the middle number. For
example, the median of 2, 4, and 7 is 4. When there is an even number of numbers, the
median is the mean of the two middle numbers. Thus, the median of the numbers 2, 4, 7,
12 is (4+7)/2 = 5.5. When there are numbers with the same values, then the formula for
the third definition of the 50th percentile should be used.
 www.rejinpaul.com
Mode
The mode is the most frequently occurring value. For the data in Table 1, the mode is
18 since more teams (4) had 18 touchdown passes than any other number of
touchdown passes. With continuous data such as response time measured to many
decimals, the frequency of each value is one since no two scores will be exactly the
same (see discussion of continuous variables).
Therefore the mode of continuous data is normally computed from a grouped frequency
distribution. Table 2 shows a grouped frequency distribution for the target response time
data. Since the interval with the highest frequency is 600-700, the mode is the middle of
that interval (650).

GEOMETRIC MEAN
Geometric Mean is a special type of average where we multiply the numbers together and then
take a square root (for two numbers), cube root (for three numbers) etc.
Example: What is the Geometric Mean of 2 and 18?
 First we multiply them: 2 × 18 = 36
 Then (as there are two numbers) take the square root: √36 = 6

In one line:
Geometric Mean of 2 and 18 = √(2 × 18) = 6
It is like the area is the same!

Example: What is the Geometric Mean of 10, 51.2 and 8?

 First we multiply them: 10 × 51.2 × 8 = 4096

 Then (as there are three numbers) take the cube root: 3√4096 = 16

In one line:
Geometric Mean = 3√(10 × 51.2 × 8) = 16

It is like the volume is the same:

 www.rejinpaul.com

Example: What is the Geometric Mean of 1,3,9,27 and 81?

 First we multiply them: 1 × 3 × 9 × 27 × 81 = 59049

 Then (as there are 5 numbers) take the 5th root: 5√59049 = 9

In one line:
Geometric Mean = 5√(1 × 3 × 9 × 27 × 81) = 9

I can't show you a nice picture of this, but it is still true that:
1 × 3 × 9 × 27 × 81 = 9 × 9 × 9 × 9 × 9

Harmonic Mean
A kind of average. To find the harmonic mean of a set of n numbers, add the reciprocals of the
numbers in the set, divide the sum by n, then take the reciprocal of the result. The harmonic
mean of {a1, a2, a3, a4, . . ., an} is given below.

RANGE
The difference between the lowest and highest values.
In {4, 6, 9, 3, 7} the lowest value is 3, and the highest is 9, so the range is 9 − 3 = 6.
Range can also mean all the output values of a function.
 www.rejinpaul.com
Quartile Deviation :
In a distribution, partial variance between the upper quartile and lower quartile is known as
'quartile deviation'. Quartile Deviation is often regarded as semi inter quartile range.

Formula : (Upper quartile- lower quartile) / 2

upper quartile = 400, lower quartile = 200 then
Quartile deviation (QD) = (400-200)/2 = 200/2
=100.
Mean Deviation
The mean of the distances of each value from their mean.
Yes, we use "mean" twice: Find the mean ... use it to work out distances ... then find the mean of
those distances!

Three steps:
 1. Find the mean of all values
 2. Find the distance of each value from that mean (subtract the mean from each value,
ignore minus signs)
 3. Then find the mean of those distances Example: the Mean Deviation of 3, 6, 6, 7, 8, 11,
15, 16
Step 1: Find the mean:
Mean = 3 + 6 + 6 + 7 + 8 + 11 + 15 + 16 = 72 = 9
8 8
Step 2: Find the distance of each value from that mean:
Value Distance from 9
3 6
6 3

6 3
7 2
8 1
11 2
15 6
16 7
 www.rejinpaul.com

Step 3. Find the mean of those distances:

6+3+3+2+1+2+6+7 30
Mean Deviation = ----------------------------------- = ----- = 3.75
8 8

So, the mean = 9, and the mean deviation = 3.75

It tells us how far, on average, all values are from the middle.

In that example the values are, on average, 3.75 away from the middle.
For deviation just think distance Formula
The formula is:
Σ|x - μ|
Mean Deviation =
N
Let's learn more about those symbols!
Firstly:
 μ is the mean (in our example μ = 9)
 x is each value (such as 3 or 16)
 N is the number of values (in our example N = 8)
 www.rejinpaul.com

Standard Deviation
The Standard Deviation is a measure of how spread out numbers are.
Its symbol is σ (the greek letter sigma)
The formula is easy: it is the square root of the Variance. So now you ask, "What is the
Variance?"
Variance:
The Variance is defined as. The average of the squared differences from the Mean.

Examples of Standard Deviation:

This tutorial is about some examples of standard deviation using all methods which are discussed
in the previous tutorial.
Example:
Calculate the standard deviation for the following sample data using all methods: 2, 4, 8, 6, 10,
and 12.
Method-I: Actual Mean Method
Marks f X F(X) (X−X¯)2 f(X−X)2

1−3 40 2 80 4 160
3−5 30 4 120 0 0
5−7 20 6 120 4 80
7−9 10 8 80 16 160
Total 100 400 400

X¯=∑f(X ) / ∑f =400/ 100=4

Method-II: Taking assumed mean as 2

Marks f X D=(X−2) fD fD2
1−3 40 2 0 0 0
3−5 30 4 2 60 120

5−7 20 6 4 80 320
7−9 10 8 6 60 160
Total 100 200 800
 www.rejinpaul.com

SKEWNESS

Lack of symmetry is called Skewness. If a distribution is not symmetrical then it is called

skewed distribution. So, mean, median and mode are different in values and one tail becomes
longer than other. The skewness may be positive or negative.

Positively skewed distribution:

If the frequency curve has longer tail to right the distribution is known as positively skewed
distribution and Mean > Median > Mode.

Negatively skewed distribution:

If the frequency curve has longer tail to left the distribution is known as negatively skewed
distribution and Mean < Median < Mode.

Measure of Skewness:
The difference between the mean and mode gives as absolute measure of skewness. If we divide
this difference by standard deviation we obtain a relative measure of skewness known as
coefficient and denoted by SK.

Karl Pearson coefficient of Skewness

SK=Mean−Mode/S.D
 www.rejinpaul.com
Sometimes the mode is difficult to find. So we use another formula

SK=3(Mean−Median)/S.D
Bowley‗s coefficient of Skewness
SK=Q1+Q3−2Median/Q3−Q1

Kelly‗s Measure of Skewness is one of several ways to measure skewness in a data distribution.
Bowley‗s skewness is based on the middle 50 percent of the observations in a data set. It leaves
25 percent of the observations in each tail of the distribution. Kelly suggested that leaving out
fifty percent of data to calculate skewness was too extreme. He created a measure to find
skewness with more data. Kelly‗s measure is based on P90 (the 90th percentile) and P10 (the 10th
percentile). Only twenty percent of observations (ten percent in each tail) are excluded from the
measure.

Kelly’s Measure Formula.

Kelley‗s measure of skewness is given in terms of percentiles and deciles(D). Kelley‗s absolute
measure of skewness (Sk)is:

Sk=P90 + P10 – 2*P50 = D9 + D1-2*D5.

Kelly‗s Measure of Skewness gives you the same information about skewness as the other three
types of skewness measures

A measure of skewness = 0 means that the distribution is symmetrical.

A measure of skewness > 0 means a positive skewness.
A measure of skewness < means a negative skewness.
www.rejinpaul.com
 www.rejinpaul.com

UNIT –III

TABULATION OF UNIVARIATE

Univariate data:Univariate means "one variable" (one type of data)

Example: Travel Time (minutes): 15, 29, 8, 42, 35, 21, 18, 42, 26
The variable is Travel Time
Bivariate data:Bivariate means "two variables", in other words there are two types of data With
bivariate data you have two sets of related data that you want to compare: Example:
An ice cream shop keeps track of how much ice cream they sell versus the temperature on that
day.
The two variables are Ice Cream Sales and Temperature.
Here are their figures for the last 12 days:

Ice Cream Sales vs Temperature

Temperature °C Ice Cream Sales

14.2° $215
16.4° $325
11.9° $185
15.2° $332

18.5° $406
22.1° $522
19.4° $412
25.1° $614
23.4° $544
18.1° $421
22.6° $445
17.2° $408
 www.rejinpaul.com

And here is the same data as a Scatter Plot:

Now we can easily see that warmer weather and more ice cream sales are linked, but the
relationship is not perfect.

Multivariate data

Multivariate Data Analysis refers to any statistical technique used to analyze data that arises
from more than one variable. This essentially models reality where each situation, product, or
decision involvesmore than a single variable.
 www.rejinpaul.com

Univariate Data
Bivariate Data

 involving a single variable  involving two variables

 does not deal with causes or  deals with causes or relationships
relationships
 the major purpose of univariate analysis is to  the major purpose of bivariate analysis is to
describe explain

 central tendency - mean, mode, median  analysis of two variables

 dispersion - range, variance, max, min, simultaneously
quartiles, standard deviation.  correlations
 frequency distributions  comparisons, relationships, causes,
 bar graph, histogram, pie chart, line explanations
graph, box-and-whisker plot  tables where one variable is contingent on
the values of the other variable.
 independent and dependent variables
Sample question: Is there a relationship between Sample question: Is there a relationship between
the number of females in Computer Programming the number of females in Computer Programming
and their scores in Mathematics? and their scores in Mathematics?

DIAGRAMATIC AND GRAPHIACAL; REPRESENTATION OF DATA

Although tabulation is very good technique to present the data, but diagrams are an advanced
technique to represent data. As a layman, one cannot understand the tabulated data easily but
with only a single glance at the diagram, one gets complete picture of the data presented.
According to M.J. Moroney, ―diagrams register a meaningful impression almost before we think.
Importance or utility of Diagrams
 Diagrams give a very clear picture of data. Even a layman can understand it very easily
and in a short time.
 We can make comparison between different samples very easily. We don‗t have to use
any statistical technique further to compare.
 This technique can be used universally at any place and at any time. This technique is
used almost in all the subjects and other various fields.
 www.rejinpaul.com
 Diagrams have impressive value also. Tabulated data has not much impression as
compared to Diagrams. A common man is impressed easily by good diagrams.
 This technique can be used for numerical type of statistical analysis, e.g. to locate Mean,
Mode, Median or other statistical values.
 It does not save only time and energy but also is economical. Not much money is needed
to prepare even good diagrams.
 These give us much more information as compared to tabulation. Technique of tabulation
has its own limits.
 This data is easily remembered. Diagrams which we see leave their lasting impression
much more than other data techniques.
 Data can be condensed with diagrams. A simple diagram can present what even cannot
be presented by 10000 words.

General Guidelines for Diagrammatic presentation

 The diagram should be properly drawn at the outset. The pith and substance of the subject
matter must be made clear under a broad heading which properly conveys the purpose of
a diagram.
 The size of the scale should neither be too big nor too small. If it is too big, it may look
ugly. If it is too small, it may not convey the meaning. In each diagram, the size of the
paper must be taken note-of. It will help to determine the size of the diagram.
 For clarifying certain ambiguities some notes should be added at the foot of the diagram.
This shall provide the visual insight of the diagram.
 Diagrams should be absolutely neat and clean. There should be no vagueness or
overwriting on the diagram.
 Simplicity refers to love at first sight. It means that the diagram should convey the
meaning clearly and easily.
 Scale must be presented along with the diagram.
 It must be Self-Explanatory. It must indicate nature, place and source of data presented.
 Different shades, colors can be used to make diagrams more easily understandable.
 Vertical diagram should be preferred to Horizontal diagrams.
 It must be accurate. Accuracy must not be done away with to make it attractive or
impressive.
 www.rejinpaul.com
Limitations of Diagrammatic Presentation

 Diagrams do not present the small differences properly.

 These can easily be misused.
 Only artist can draw multi-dimensional diagrams.
 In statistical analysis, diagrams are of no use.
 Diagrams are just supplement to tabulation.
 Only a limited set of data can be presented in the form of diagram.
 Diagrammatic presentation of data is a more time consuming process.
 Diagrams present preliminary conclusions.
 Diagrammatic presentation of data shows only on estimate of the actual behavior of the
variables.

Types of Diagrams
(a) Line Diagrams

In these diagrams only line is drawn to represent one variable. These lines may be vertical or
horizontal. The lines are drawn such that their length is the proportion to value of the terms or
items so that comparison may be done easily.
 www.rejinpaul.com

(b) Simple Bar Diagram

Like line diagrams these figures are also used where only single dimension i.e. length can
present the data. Procedure is almost the same, only one thickness of lines is measured. These
can also be drawn either vertically or horizontally. Breadth of these lines or bars should be equal.
Similarly distance between these bars should be equal. The breadth and distance between them
should be taken according to space available on the paper.

Imagine you just did a survey of your friends to find which kind of movie they liked best:

Table: Favorite Type of Movie

Comedy Action Romance Drama SciFi

4 5 6 1 4

We can show that on a bar graph like this:

 www.rejinpaul.com

It is a really good way to show relative sizes: we can see which types of movie are most liked,
and which are least liked, at a glance.

We can use bar graphs to show the relative sizes of many things, such as what type of car people
have, how many customers a shop has on different days and so on.

(c) Multiple Bar Diagrams

The diagram is used, when we have to make comparison between more than two variables. The
number of variables may be 2, 3 or 4 or more. In case of 2 variables, pair of bars is drawn.
Similarly, in case of 3 variables, we draw triple bars. The bars are drawn on the same
proportionate basis as in case of simple bars. The same shade is given to the same item.
 www.rejinpaul.com

Draw a multiple bar chart to represent the import and export of Canada (values in $) for the years
1991 to 1995.

Years Imports Exports

1991 7930 4260
1992 8850 5225
1993 9780 6150
1994 11720 7340
1995 12150 8145

chart showing the import and export of Canada from 1991 – 1995.
 www.rejinpaul.com

(d) Sub-divided Bar Diagram

The data which is presented by multiple bar diagram can be presented by this diagram. In this
case we add different variables for a period and draw it on a single bar as shown in the following
examples. The components must be kept in same order in each bar. This diagram is more
efficient if number of components is less i.e. 3 to 5.

(e) Percentage Bar Diagram

Like sub-divide bar diagram, in this case also data of one particular period or variable is put on
single bar, but in terms of percentages. Components are kept in the same order in each bar for

Easy comparison.

(f) Duo-directional Bar Diagram

In this case the diagram is on both the sides of base line i.e. to left and right or to above or below
sides.
(g) Broken Bar Diagram
This diagram is used when value of some variable is very high or low as compared to others. In
this case the bars with bigger terms or items may be shown broken.
 www.rejinpaul.com

One dimentional diagrams:

A diagram in which the size of only one dimension i.e. length is fixed in proportion to the value
of the data is called one dimensional diagram. Such diagrams are also popularly called bar
diagrams. These diagrams can be drawn in both vertical and horizontal manner. The related
different bar diagrams differ from each other only in respect of their length dimension, while
they remain the same in respect of their other two dimensions i.e. breadth and thickness. The size
of breadth of each of such diagrams is determined taking into consideration the number of
diagrams to be drawn, and the size of the paper at one‗s end. They may take the form of a line, or
a thread if they are to be drawn in large numbers on the surface of a paper. However, their
breadth should neither be too large nor too small for that in both the cases they look ugly. The
dimension of thickness does not look prominent in such diagrams. The examples of such
diagrams are given on the next page.

Techniques of drawing bar diagrams

The following are the techniques of drawing the bar diagrams :
(i) First, draw the base line, preferably horizontally, and divide it into a number of equal parts
keeping in view the number of diagrams to be drawn.
(ii) Then, draw the scale line preferably vertically, and divide into a number of equal parts
keeping in view the maximum value to be represented.
(iii) Then, fix the width of the bar uniformly keeping in view the number of bars to be drawn
and the gaps to be provided in between each two of them.
(iv) Then, fix the size of gaps to be provided between each of the two bars uniformly.
(v) Then, fix the lengths of the different bars in proportion of the value of the data.
(vi) Then, draw the different bars in accordance with their length, and width thus fixed, and
arranged in order o their length, or time of occurrence.
(vii) Then, decorate the bars with similar or different colours, or shades according to the
similarly or dissimilarity in the nature of the data respectively.
(viii) Give a description of the data in short at the bottom of respective bars.
(xiv) Put the respective figures at the top of each bar to read out the exact value at a glance
without looking at the scale.
 www.rejinpaul.com

Advantages
The chief advantages of a bar diagram can be outlined as under:
1. It is very simple to draw and read as well.
2. It is the only form of diagram which can represent a large number of data on a piece of paper.
3. It can be drawn both vertically and horizontally.
4. It gives a better look and facilitates comparison.

Disadvantages
1. It cannot exhibit a large number of aspects of the data.
2. The which of the bars are fixed arbitrarily by a drawer.

Two-Dimensional

A shape that only has two dimensions (such as width and height) and no thickness. Squares,
Circles, Triangles, etc are two dimensional objects.
Also known as "2D".
 www.rejinpaul.com

3.Three-Dimensional

An object that has height, width and depth, like any object in the real world. Example:
your body is three-dimensional.
Also known as "3D".

Pie Chart: a special chart that uses "pie slices" to show relative sizes of data.
Imagine you survey your friends to find the kind of movie they like best:
Table: Favorite Type of Movie

Comedy Action Romance Drama SciFi

4 5 6 1 4

You can show the data by this Pie Chart:

It is a really good way to show relative sizes: it is easy to see which movie types are most liked,
and which are least liked, at a glance.
 www.rejinpaul.com

UNIT-IV

SMALL SAMPLE TESTS

statistical Test for Population Mean (Small Sample)

In this section wil ladjust our statistical test for the population mean to apply to small sample
situations. Fortunately (sic!), this will be easy (in fact, once you understand one statistical test,
additional tests are easy since they all follow a similar procedure.

The only difference in performing a "small sample" statistical test for the mean as opposed to a
"large sample test" is that we do not use the normal distribution as prescribed by the Central
Limit theorem, but instead a more conservative distribution called the T-Distribution. The
Central Limit theorem applies best when sample sizes are large so that we need to make some
adjustment in computing probabilities for small sample sizes. The appropriate function in Excel
is the TDIST function, defined as follows:
TDIST(T, N-1, TAILS), where

 Tis the value for which we want to compute the probability

 N is the sample size (and N-1 is frequently called the "degrees of freedom")
 TAILS is either 1 (for a 1-tail test) or 2 (for a 2-tail test). Since we again consider 2-
tailed tests only we always use 2

With that new Excel function our test procedure for a sample mean, small sample size, is as
follows:
Statistical Test for the Mean (small sample size N < 30):
Fix an error level you are comfortable with (something like 10%, 5%, or 1% is most
common). Denote that "comfortable error level" by the letter "A".
Then setup the test as follows:
Null Hypothesis H0:
mean = M, i.e. The mean is a known number M Alternative Hypothesis Ha:
mean ≠ M, i.e. mean is different from M (2-tailed test)
 www.rejinpaul.com

Test Statistics:
Select a random sample of size N, compute its sample mean X and the standard deviation
S. Then compute the corresponding t-score as follows:
T= (X - M) / ( S / sqrt(N) )
Rejection Region (Conclusion)

Compute p = 2*(1 - P(t > |T|)) = TDIST(ABS(T), N-1, 2)

If the probability p computed in the above step is less than A (the error level you were
comfortable with inititially, you reject the null hypothesis H 0 and accept the alternative
hypothesis. Otherwise you declare your test inconclusive.

Comments:
 The null and alternative hypothesis for this test are the same as before
 The calculation of the test statistics is the same as before, but the result is called T instead
of Z (oh well -:)
 The TDIST function is similar to the NORMSDIST function, but it does not work for
negative values of T (a limitation of Excel), and it automatically gives a "tail" probability.
Thus, the computation of the p-value had to be adjusted accordingly.
 The ABS function in the above formulas stands for the "absolute value" function. (In
other words, just drop any minus signs ... -:)

Example 1: A group of secondary education student teachers were given 2 1/2 days of training
in interpersonal communication group work. The effect of such a training session on the
dogmatic nature of the student teachers was measured y the difference of scores on the "Rokeach
Dogmatism test given before and after the training session. The difference "post minus pre
score" was recorded as follows:

-16, -5, 4, 19, -40, -16, -29, 15, -2, 0, 5, -23, -3, 16, -8, 9, -14, -33, -64, -33

Can we conclude from this evidence that the training session makes student teachers less
dogmatic (at the 5% level of significance) ?
This is of course the same example as before, where we incorrectly used the normal distribution
 www.rejinpaul.com

to compute the probability in the last step. This time, we will do it correctly, which is fortunately

almost identical to the previous case (except that we use TDIST instead of NORMDIST):
 Null Hypothesis: there is no difference in dogmatism, i.e. mean = 0
 Alternative Hypothesis: dogmatism is different, i.e. mean not equal to 0
 Test statistics: sample mean = -10.9, standard deviation = 21.33, sample size = 20.
Compute

T = (-10.9 - 0) / (21.33 / sqrt(20) ) = -2.28

 Rejection Region: We use Excel to compute p = TDIST(2.28, 19, 2) = 0.034, or 3.4%.

That probability is less than 0.05 so we reject the null hypothesis.

Note that in the previous section we (incorrectly) computed the probability p to be 2.2%, now it
is 3.4%. The difference is small, but can be significant in special situations. Thus, to be safe:

 if N > 30 use the Z-Test based on the standard normal distribution NORMSDIST as in
the previous section
 if N < 30 use the T-Test based on the T-Distribution TDIST as in this section

Example 2: Suppose GAP, the clothing store, wants to introduce their line of clothing for
women to another country. But their clothing sizes are based on the assumption that the
average size of a woman is 162 cm. To determine whether they can simply ship the clothes
to the new country they select 5 women at random in the target country and determine
their heights as follows: 149, 165, 150, 158, 153

should they adjust their line of clothing or they ship them without change? Make sure to decide
at the 0.05-level. By now statistical testing is second-nature (I hope -:)
 Null Hypothesis: mean height in new country is the same as in old country, i.e. M = 162
 Alt. Hypothesis: mean height in new country is different from old country, i..e. M not
equal to 162 (either too small or too tall would be bad for GAP)
 Test Statistics: we can compute the sample mean = 155 and the sample standard
deviation = 6.59 while the sample size is clearly N = 5.
 www.rejinpaul.com

 Therefore
T = (155 - 162) / ( 6.59 / sqrt(5) ) = -2.37

 Rejection Region: We use Excel to compute p = TDIST(2.37, 4, 2) = 0.077. Thus, if we

did decide to reject the null hypothesis the probability of that decision being wrong is
7.7%. That is larger than 0.05, thus we declare the test inconclusive.

t- Distribution.

Properties of the t-Distribution

If a population is essentially normal, then the distribution is:

This is the equation for the Student t-Distribution, or simply t-Distribution, for all samples of
size n less than 30.
 www.rejinpaul.com

To find the Rejection Region, we can use the t-Distribution Table. This table merely requires
knowledge of the sample size, which allows us to calculate the Degrees of Freedom, and the

Significance Level.
As illustrated above, the t-distribution has many properties which differentiate it from the
standard normal or z-distribution.
The distribution shares the bell curve of the z, but reflects the variablility that is inherent
with smaller sample sizes.
The shape of the t-distribution is dependent on the sample size n .
The standard deviation is greater than 1.
As the sample size n increases, the shape of the curve approaches the standard deviation.

PAIRED T TEST
 Paired sample t-test is a statistical technique that is used to compare two population
means in the case of two samples that are correlated. Paired sample t-test is used in
‗before-after‗ studies, or when the samples are the matched pairs, or when it is a case-
control study. For example, if we give training to a company employee and we want to
know whether or not the training had any impact on the efficiency of the employee, we
could use the paired sample test. We collect data from the employee on a seven scale
rating, before the training and after the training. By using the paired sample t-test, we
can statistically conclude whether or not training has improved the efficiency of the
employee. In medicine, by using the paired sample t-test, we can figure out whether or
not a particular medicine will cure the illness.

Steps:
1. Set up hypothesis: We set up two hypotheses. The first is the null hypothesis, which
assumes that the mean of two paired samples are equal. The second hypothesis will
be an alternative hypothesis, which assumes that the means of two paired samples are
not equal.
2. Select the level of significance: After making the hypothesis, we choose the level
of significance. In most of the cases, significance level is 5%, (in medicine, the
significance level is set at 1%).

3. Calculate the parameter: To calculate the parameter we will use the following
 www.rejinpaul.com
formula:

Where d bar is the mean difference between two samples, s² is the sample variance, n is
the sample size and t is a paired sample t-test with n-1 degrees of freedom. An alternate
formula for paired sample t-test is:

4. Testing of hypothesis or decision making: After calculating the parameter, we will

compare the calculated value with the table value. If the calculated value is greater than
the table value, then we will reject the null hypothesis for the paired sample t-test. If the
calculated value is less than the table value, then we will accept the null hypothesis and
say that there is no significant mean difference between the two paired samples.

Assumptions:
1. Only the matched pairs can be used to perform the test.
2. Normal distributions are assumed.
3. The variance of two samples is equal.
4. Cases must be independent of each other.

Administration, Analysis and Reporting

Statistics Solutions consists of a team of professional methodologists and statisticians that
can assist the student or professional researcher in administering the survey instrument,
collecting the data, conducting the analyses and explaining the results.
 www.rejinpaul.com

Let x = the difference in weight 3 months after the program starts. The null hypothesis is:
H0: μ = 0; i.e. any differences in weight is due to chance
We can make the following calculations using the difference column D:
s.e. = std dev / = 6.33 / = 1.6343534
tobs = (x̄ – μ) /s.e. = (10.93 – 0) /1.63 = 6.6896995
tcrit = TINV(α, df) = TINV(.05, 14) = 2.1447867
Since tobs > tcrit we reject the null hypothesis and conclude with 95% confidence that the
difference in weight before and after the program is not due solely to chance.
Alternatively we can use a type 1 TTEST to perform the analysis as follows:
p-value = TTEST(B4:B18, C4:C18, 2, 1) = 1.028E-05 < .05 = α
and so once again we reject the null hypothesis.

Chi-Square Test for Independence

This lesson explains how to conduct a chi-square test for independence. The test is applied
when you have two categorical variables from a single population. It is used to determine
whether there is a significant association between the two variables.
For example, in an election survey, voters might be classified by gender (male or female) and
voting preference (Democrat, Republican, or Independent). We could use a chi-square test for
independence to determine whether gender is related to voting preference. The sample problem
at the end of the lesson considers this example.

When to Use Chi-Square Test for Independence

The test procedure described in this lesson is appropriate when the following conditions are met:
 The sampling method is simple random sampling.
 The variables under study are each categorical.
 If sample data are displayed in a contingency table, the expected frequency count for
each cell of the table is at least 5.
 www.rejinpaul.com
This approach consists of four steps: (1) state the hypotheses, (2) formulate an analysis plan, (3)
analyze sample data, and (4) interpret results.

State the Hypotheses

Suppose that Variable A has r levels, and Variable B has c levels. The null hypothesis states that
knowing the level of Variable A does not help you predict the level of Variable B. That is, the
variables are independent.

H0: Variable A and Variable B are independent.

Ha: Variable A and Variable B are not independent.

The alternative hypothesis is that knowing the level of Variable A can help you predict the level
of Variable B.

Note: Support for the alternative hypothesis suggests that the variables are related; but the
relationship is not necessarily causal, in the sense that one variable "causes" the other.

Formulate an Analysis Plan

The analysis plan describes how to use sample data to accept or reject the null hypothesis. The
plan should specify the following elements.

 Significance level. Often, researchers choose significance levels equal to 0.01, 0.05, or
0.10; but any value between 0 and 1 can be used.

 Test method. Use the chi-square test for independence to determine whether
there is a significant relationship between two categorical variables.
 www.rejinpaul.com
Analyze Sample Data
Using sample data, find the degrees of freedom, expected frequencies, test statistic, and the P- value
associated with the test statistic. The approach described in this section is illustrated in the sample
problem at the end of this lesson.

 Degrees of freedom. The degrees of freedom (DF) is equal to:

DF = (r - 1) * (c - 1)

where r is the number of levels for one catagorical variable, and c is the number of levels
for the other categorical variable.

 Expected frequencies. The expected frequency counts are computed separately for each
level of one categorical variable at each level of the other categorical variable. Compute r
* c expected frequencies, according to the following formula.

Er,c = (nr * nc) / n

where Er,c is the expected frequency count for level r of Variable A and level c of
Variable B, nr is the total number of sample observations at level r of Variable A, n c is
the total number of sample observations at level c of Variable B, and n is the total
sample size.

 Test statistic. The test statistic is a chi-square random variable (Χ2) defined by the
following equation.

Χ2 = Σ [ (Or,c - Er,c)2 / Er,c ]

where Or,c is the observed frequency count at level r of Variable A and level c of
Variable B, and Er,c is the expected frequency count at level r of Variable A and level c
of Variable B.
 www.rejinpaul.com
 P-value. The P-value is the probability of observing a sample statistic as extreme as the
test statistic. Since the test statistic is a chi-square, use the Chi-Square Distribution
Calculator to assess the probability associated with the test statistic. Use the degrees of
freedom computed above.

Interpret Results

If the sample findings are unlikely, given the null hypothesis, the researcher rejects the null
hypothesis. Typically, this involves comparing the P-value to the significance level, and rejecting
the null hypothesis when the P-value is less than the significance level.

Test Your Understanding

Problem

A public opinion poll surveyed a simple random sample of 1000 voters. Respondents were classified by
gender (male or female) and by voting preference (Republican, Democrat, or Independent). Results are
shown in the contingency table below.

Voting Preferences
Row
Republican Democrat total
Independent
Male 200 150 50 400

Female 250 300 50 600

Column 450 450 100 1000

total

Is there a gender gap? Do the men's voting preferences differ significantly from the women's
preferences? Use a 0.05 level of significance.

Solution
 www.rejinpaul.com

The solution to this problem takes four steps:

(1) state the hypotheses,
(2) formulate an analysis plan,
(3) analyze sample data, and
(4) interpret results.
We work through those steps below:

State the hypotheses. The first step is to state the null hypothesis and an alternative hypothesis.

H0: Gender and voting preferences are independent.

Ha: Gender and voting preferences are not independent.

Formulate an analysis plan. For this analysis, the significance level is 0.05. Using sample data,
we will conduct a chi-square test for independence.
 Analyze sample data. Applying the chi-square test for independence to sample data, we
compute the degrees of freedom, the expected frequency counts, and the chi-square test
statistic. Based on the chi-square statistic and the degrees of freedom, we determine the
P-value.

DF = (r - 1) * (c - 1) = (2 - 1) * (3 - 1) = 2

Er,c = (nr * nc) / n

E1,1 = (400 * 450) / 1000 = 180000/1000 = 180

E1,2 = (400 * 450) / 1000 = 180000/1000 = 180
E1,3 = (400 * 100) / 1000 = 40000/1000 = 40
E2,1 = (600 * 450) / 1000 = 270000/1000 = 270
E2,2 = (600 * 450) / 1000 = 270000/1000 = 270
E2,3 = (600 * 100) / 1000 = 60000/1000 = 60
 www.rejinpaul.com

Χ2 = Σ [ (Or,c - Er,c)2 / Er,c ]

Χ2 = (200 - 180)2/180 + (150 - 180)2/180 + (50 - 40)2/40

+ (250 - 270)2/270 + (300 - 270)2/270 + (50 - 60)2/60

Χ2 = 400/180 + 900/180 + 100/40 + 400/270 + 900/270 + 100/60

Χ2 = 2.22 + 5.00 + 2.50 + 1.48 + 3.33 + 1.67 = 16.2

where DF is the degrees of freedom, r is the number of levels of gender, c is the number
of levels of the voting preference, nr is the number of observations from level r of gender,
nc is the number of observations from level c of voting preference, n is the number of
observations in the sample, Er,c is the expected frequency count when gender is level r
and voting preference is level c, and Or,c is the observed frequency count when gender is
level r voting preference is level c.

The P-value is the probability that a chi-square statistic having 2 degrees of freedom is
more extreme than 16.2.

We use the Chi-Square Distribution Calculator to find P(Χ2 > 16.2) = 0.0003.

 Interpret results. Since the P-value (0.0003) is less than the significance level (0.05), we
cannot accept the null hypothesis. Thus, we conclude that there is a relationship between
gender and voting preference.

Note: If you use this approach on an exam, you may also want to mention why this approach is
appropriate. Specifically, the approach is appropriate because the sampling method was simple
random sampling, the variables under study were categorical, and the expected frequency count
was at least 5 in each cell of the contingency table.
 www.rejinpaul.com
Example: "Which holiday would you rather have?"

Beach Cruise

Men 209 280

Women 225 248

Does Gender affect Preferred Holiday?

If Gender (Man or Woman) does affect Preferred Holiday we say they are dependent.
By doing some special calculations (explained later), we come up with a "p" value:
p value is 0.13

Now, p < 0.05 is the usual test for dependence. In this case p is greater than 0.05, so we believe
the variables are independent (ie not linked together).
In other words Men and Women probably do not have a different preference for Beach Holidays
or Cruises.
Understanding "p" Value
"p" is the probability the variables are independent.

Imagine for the previous example you had tried to fool the test by choosing a random sample of
Men each time:

Men: Men:
Beach 209, Cruise Beach 225, Cruise
280 248
 www.rejinpaul.com

Is it likely you would get such different results surveying Men each time?
Well the "p" value of 0.132 says that it really could happen every so often.
Surveys are random after all. We expect slightly different results each time, right?

So most people want to see a p value less than 0.05 before they are happy to say the results show
the groups have a different response.

Let's see another example:

Example: "Which pet would you rather have?"

Cat Dog

Men 207 282

Women 231 242

By doing the calculations (shown later), we come up with:

P value is 0.043

In this case p < 0.05, so this result is thought of as being "significant" meaning we think the
variables are not independent.

In other words, because 0.043 < 0.05 we think that Gender is linked to Pet Preference (Men and
Women have different preferences for Cats and Dogs).

Just out of interest, notice that the numbers in our two examples are similar, but the resulting p-
values are very different: 0.132 and 0.043. This shows how sensitive the test is!

Why p<0.05 ?

It is just a choice! Using p<0.05 is common, but we could have chosen p<0.01 to be even more
sure that the groups behave differently, or any value really.
 www.rejinpaul.com

Calculating P-Value

So how do we calculate this p-value? We use the Chi-Square Test!

Chi-Square Test

Note: Chi Sounds like "Hi" but with a K, so say Chi-Square like "Ki square"
And Chi is the greek letter Χ, so we can also write it Χ2
Important points before we get started:

 This test only works for categorical data (data in categories), such as Gender {Men,
Women} or color {Red, Yellow, Green, Blue} etc, but not numerical data such as
height or weight.
 The numbers must be large enough. Each entry must be 5 or more. In our example
we have values such as 209, 282, etc, so we are good to go.

Our first step is to state our hypotheses:

Hypothesis: A statement that might be true, which can then be tested.

The two hypotheses are.
 Gender and preference for cats or dogs are independent.
 Gender and preference for cats or dogs are not independent.

Subtract expected from actual, square it, then divide by expected:

Now add up those values:

1.099 + 0.918 + 1.136 + 0.949 = 4.102

Chi-Square is 4.102
 www.rejinpaul.com

From Chi-Square to p

To get from Chi-Square to p-value is a difficult calculation, so either look it up in a table, or use
the Chi-Square Calculator.

But first you will need a "Degree of Freedom" (DF)

Calculate Degrees of Freedom
Multiply (rows − 1) by (columns − 1)
Example: DF = (2 − 1)(2 − 1) = 1×1 = 1
Result
The result is:
p = 0.04283

Done!
Chi-Square Formula
This is the formula for Chi-Square:

= The observed (actual) value

E = The expected value

Introduction to Correlation and Regression Analysis

In this section we will first discuss correlation analysis, which is used to quantify the association
between two continuous variables (e.g., between an independent and a dependent variable or
between two independent variables). Regression analysis is a related technique to assess the
relationship between an outcome variable and one or more risk factors or confounding variables.
The outcome variable is also called the response or dependent variable and the risk factors and
confounders are called the predictors, or explanatory or independent variables. In regression
analysis, the dependent variable is denoted "y" and the independent variables are denoted by "x".
 www.rejinpaul.com

[NOTE: The term "predictor" can be misleading if it is interpreted as the ability to predict
even beyond the limits of the data. Also, the term "explanatory variable" might give an
impression of a causal effect in a situation in which inferences should be limited to
identifying associations. The terms "independent" and "dependent" variable are less
subject to these interpretations as they do not strongly imply cause and effect.

Correlation Analysis
In correlation analysis, we estimate a sample correlation coefficient, more specifically the
Pearson Product Moment correlation coefficient. The sample correlation coefficient, denoted r,
ranges between -1 and +1 and quantifies the direction and strength of the linear association
between the two variables. The correlation between two variables can be positive (i.e., higher
levels of one variable are associated with higher levels of the other) or negative (i.e., higher
levels of one variable are associated with lower levels of the other).

The sign of the correlation coefficient indicates the direction of the association. The magnitude
of the correlation coefficient indicates the strength of the association.
For example, a correlation of r = 0.9 suggests a strong, positive association between two
variables, whereas a correlation of r = -0.2 suggest a weak, negative association. A correlation
close to zero suggests no linear association between two continuous variables.

LISA: [I find this description confusing. You say that the correlation coefficient is a measure of
the "strength of association", but if you think about it, isn't the slope a better measure of
association? We use risk ratios and odds ratios to quantify the strength of association, i.e., when
an exposure is present it has how many times more likely the outcome is. The analogous
quantity in correlation is the slope, i.e., for a given increment in the independent variable, how
many times is the dependent variable going to increase?
And "r" (or perhaps better R-squared) is a measure of how much of the variability in the
dependent variable can be accounted for by differences in the independent variable. The
analogous measure for a dichotomous variable and a dichotomous outcome would be the
attributable proportion, i.e., the proportion of Y that can be attributed to the presence of the
exposure.]
 www.rejinpaul.com

It is important to note that there may be a non-linear association between two continuous
variables, but computation of a correlation coefficient does not detect this. Therefore, it is always
important to evaluate the data carefully before computing a correlation coefficient. Graphical
displays are particularly useful to explore associations between variables.

The figure below shows four hypothetical scenarios in which one continuous variable is plotted
along the X-axis and the other along the Y-axis.

 Scenario 1 depicts a strong positive association (r=0.9), similar to what we might see
for the correlation between infant birth weight and birth length.
 Scenario 2 depicts a weaker association (r=0,2) that we might expect to see between
age and body mass index (which tends to increase with age).
 Scenario 3 might depict the lack of association (r approximately 0) between the extent of
media exposure in adolescence and age at which adolescents initiate sexual activity.
 Scenario 4 might depict the strong negative association (r= -0.9) generally observed
between the number of hours of aerobic exercise per week and percent body fat.

Example - Correlation of Gestational Age and Birth Weight

A small study is conducted involving 17 infants to investigate the association between
gestational age at birth, measured in weeks, and birth weight, measured in grams.
 www.rejinpaul.com

We wish to estimate the association between gestational age and infant birth weight. In this
example, birth weight is the dependent variable and gestational age is the independent variable.
Thus y=birth weight and x=gestational age. The data are displayed in a scatter diagram in the
figure below.
 www.rejinpaul.com

Each point represents an (x,y) pair (in this case the gestational age, measured in weeks, and the
birth weight, measured in grams). Note that the independent variable is on the horizontal axis (or
X-axis), and the dependent variable is on the vertical axis (or Y-axis). The scatter plot shows a
positive or direct association between gestational age and birth weight. Infants with shorter
gestational ages are more likely to be born with lower weights and infants with longer gestational
ages are more likely to be born with higher weights.

The formula for the sample correlation coefficient is

where Cov(x,y) is the covariance of x and y defined as

are the sample variances of x and y, defined as

The variances of x and y measure the variability of the x scores and y scores around their
respective sample means (

, considered separately). The covariance measures the variability of the (x,y) pairs
around the mean of x and mean of y, considered simultaneously.

To compute the sample correlation coefficient, we need to compute the variance of gestational
age, the variance of birth weight and also the covariance of gestational age and birth weight.

We first summarize the gestational age data. The mean gestational age is:
 www.rejinpaul.com

To compute the variance of gestational age, we need to sum the squared deviations (or
differences) between each observed gestational age and the mean gestational age. The
computations are summarized below.

The variance of gestational age is:

Next, we summarize the birth weight data. The mean birth weight is:

The variance of birth weight is computed just as we did for gestational age as shown in the table
below.
 www.rejinpaul.com

The variance of birth weight is:

Next we compute the covariance,

To compute the covariance of gestational age and birth weight, we need to multiply the deviation
from the mean gestational age by the deviation from the mean birth weight for each participant
(i.e.,

The computations are summarized below. Notice that we simply copy the deviations from the
mean gestational age and birth weight from the two tables above into the table below and
multiply.
 www.rejinpaul.com

The covariance of gestational age and birth weight is:

We now compute the sample correlation coefficient:

Not surprisingly, the sample correlation coefficient indicates a strong positive correlation.
As we noted, sample correlation coefficients range from -1 to +1. In practice, meaningful
correlations (i.e., correlations that are clinically or practically important) can be as small as 0.4
(or -0.4) for positive (or negative) associations. There are also statistical tests to determine
whether an observed correlation is statistically significant or not (i.e., statistically significantly
different from zero). Procedures to test whether an observed sample correlation is suggestive of a
statistically significant correlation are described in detail in Kleinbaum, Kupper and Muller. 1
 www.rejinpaul.com

Spearman Rank Correlation :

The Pearson correlation coefficient between the ranked variables has been termed as the
Spearman correlation coefficient. It is also referred as 'grade correlation'.

Formula :
R = 1 - ( (6 X σd2) / (n3 - n) )

Partial correlation analysis involves studying the linear relationship between two
variables after excluding the effect of one or more independent factors.
Simple correlation does not prove to be an all-encompassing technique especially under the
above circumstances. In order to get a correct picture of the relationship between two
variables, we should first eliminate the influence of other variables.

For example, study of partial correlation between price and demand would involve studying the
relationship between price and demand excluding the effect of money supply, exports, etc.

Partial correlation analysis involves studying the linear relationship between two
variables after excluding the effect of one or more independent factors.
Simple correlation does not prove to be an all-encompassing technique especially under the
above circumstances. In order to get a correct picture of the relationship between two
variables, we should first eliminate the influence of other variables.
For example, study of partial correlation between price and demand would involve studying the
relationship between price and demand excluding the effect of money supply, exports, etc.

Partial correlation analysis involves studying the linear relationship between two
variables after excluding the effect of one or more independent factors.
Simple correlation does not prove to be an all-encompassing technique especially under the
above circumstances. In order to get a correct picture of the relationship between two
variables, we should first eliminate the influence of other variables.
For example, study of partial correlation between price and demand would involve studying the
relationship between price and demand excluding the effect of money supply, exports, etc.
 www.rejinpaul.com

Multiple Correlation
Another technique used to overcome the drawbacks of simple correlation is multiple regression
analysis.

Here, we study the effects of all the independent variables simultaneously on a dependent
variable. For example, the correlation co-efficient between the yield of paddy (X1) and the other
variables, viz. type of seedlings (X2), manure (X3), rainfall (X4), humidity (X5) is the multiple
correlation co-efficient R1.2345 . This co-efficient takes value between 0 and +1.
The limitations of multiple correlation are similar to those of partial correlation. If multiple and
partial correlation are studied together, a very useful analysis of the relationship between the
different variables is possible.
 www.rejinpaul.com

UNIT-V

REGRESSION ANALYSIS

Regression Analysis
Introduction
As you develop Cause & Effect diagrams based on data, you may wish to examine the degree of
correlation between variables. A statistical measurement of correlation can be calculated using
the least squares method to quantify the strength of the relationship between two variables. The
output of that calculation is the Correlation Coefficient, or (r), which ranges between -1 and 1.
A value of 1 indicates perfect positive correlation - as one variable increases, the second
increases in a linear fashion. Likewise, a value of -1 indicates perfect negative correlation - as
one variable increases, the second decreases. A value of zero indicates zero correlation.

Before calculating the Correlation Coefficient, the first step is to construct a scatter diagram.
Most spreadsheets, including Excel, can handle this task. Looking at the scatter diagram will
give you a broad understanding of the correlation. Following is a scatter plot chart example
based on an automobile manufacturer.

In this case, the process improvement team is analyzing door closing efforts to understand what
the causes could be. The Y-axis represents the width of the gap between the sealing flange of a
car door and the sealing flange on the body - a measure of how tight the door is set to the body.
The fishbone diagram indicated that variability in the seal gap could be a cause of variability in
door closing efforts.
 www.rejinpaul.com

In this case, you can see a pattern in the data indicating a negative correlation (negative slope)
between the two variables. In fact, the Correlation Coefficient is -0.78, indicating a strong
inverse or negative relationship.

MoreSteam Note: It is important to note that Correlation is not Causation - two variables can
be very strongly correlated, but both can be caused by a third variable. For example, consider
two variables: A) how much my grass grows per week, and B) the average depth of the local
reservoir. Both variables could be highly correlated because both are dependent upon a third
variable - how much it rains.

In our car door example, it makes sense that the tighter the gap between the sheet metal sealing
surfaces (before adding weatherstrips and trim), the harder it is to close the door. So a
rudimentary understanding of mechanics would support the hypothesis that there is a causal
relationship. Other industrial processes are not always as obvious as these simple examples, and
determination of causal relationships may require more extensive experimentation (Design of
Experiments).

Simple Regression Analysis

While Correlation Analysis assumes no causal relationship between variables, Regression
Analysis assumes that one variable is dependent upon: A) another single independent variable
(Simple Regression) , or B) multiple independent variables (Multiple Regression).

Regression plots a line of best fit to the data using the least-squares method. You can see an
example below of linear regression using the same car door scatter plot:
 www.rejinpaul.com
You can see that the data is clustered closely around the line, and that the line has a downward
slope. There is strong negative correlation expressed by two related statistics: the r value, as
stated before is, -0.78 the r2 value is therefore 0.61. R2, called the Coefficient of Determination,
expresses how much of the variability in the dependent variable is explained by variability in the
independent variable. You may find that a non-linear equation such as an exponential or power
function may provide a better fit and yield a higher r 2 than a linear equation.

These statistical calculations can be made using Excel, or by using any of several statistical
analysis software packages. MoreSteam provides links to statistical software downloads,
including free software.

Multiple Regression Analysis

Multiple Regression Analysis uses a similar methodology as Simple Regression, but includes
more than one independent variable. Econometric models are a good example, where the
dependent variable of GNP may be analyzed in terms of multiple independent variables, such as
interest rates, productivity growth, government spending, savings rates, consumer confidence,
etc.

Many times historical data is used in multiple regression in an attempt to identify the most
significant inputs to a process. The benefit of this type of analysis is that it can be done very
quickly and relatively simply. However, there are several potential pitfalls:

 The data may be inconsistent due to different measurement systems, calibration drift,
different operators, or recording errors.
 The range of the variables may be very limited, and can give a false indication of low
correlation. For example, a process may have temperature controls because temperature
has been found in the past to have an impact on the output. Using historical temperature
data may therefore indicate low significance because the range of temperature is already
controlled in tight tolerance.
 www.rejinpaul.com
 There may be a time lag that influences the relationship - for example, temperature
may be much more critical at an early point in the process than at a later point, or vice-
versa. There also may be inventory effects that must be taken into account to make sure
that all measurements are taken at a consistent point in the process.

Once again, it is critical to remember that correlation is not causality. As stated by Box, Hunter
and Hunter: "Broadly speaking, to find out what happens when you change something, it is
necessary to change it. To safely infer causality the experimenter cannot rely on natural
happenings to choose the design for him; he must choose the design for himself and, in
1
particular, must introduce randomization to break the links with possible lurking variables".

Returning to our example of door closing efforts, you will recall that the door seal gap had an r2
of 0.61. Using multiple regression, and adding the additional variable "door weatherstrip
durometer" (softness), the r2 rises to 0.66. So the durometer of the door weatherstrip added some
explaining power, but minimal. Analyzed individually, durometer had much lower correlation
with door closing efforts - only 0.41.
This analysis was based on historical data, so as previously noted, the regression analysis only
tells us what did have an impact on door efforts, not what could have an impact. If the range of
durometer measurements was greater, we might have seen a stronger relationship with door
closing efforts, and more variability in the output.

Trend Analysis
There are no proven "automatic" techniques to identify trend components in the time series data;
however, as long as the trend is monotonous (consistently increasing or decreasing) that part of
data analysis is typically not very difficult. If the time series data contain considerable error, then
the first step in the process of trend identification is smoothing.

Smoothing. Smoothing always involves some form of local averaging of data such that the
nonsystematic components of individual observations cancel each other out. The most common
technique is moving average smoothing which replaces each element of the series by either the
simple or weighted average of n surrounding elements, where n is the width of the smoothing
"window" (see Box & Jenkins, 1976; Velleman & Hoaglin, 1981). Medians can be used instead
of means. The main advantage of median as compared to moving average smoothing is that its
results are less biased by outliers (within the smoothing window). Thus, if there are outliers in
 www.rejinpaul.com
the data (e.g., due to measurement errors), median smoothing typically produces smoother or at
least more "reliable" curves than moving average based on the same window width. The main
disadvantage of median smoothing is that in the absence of clear outliers it may produce more
"jagged" curves than moving average and it does not allow for weighting.

In the relatively less common cases (in time series data), when the measurement error is very
large, the distance weighted least squares smoothing or negative exponentially weighted
smoothing techniques can be used. All those methods will filter out the noise and convert the
data into a smooth curve that is relatively unbiased by outliers (see the respective sections on
each of those methods for more details). Series with relatively few and systematically distributed
points can be smoothed with bicubic splines.

Fitting a function. Many monotonous time series data can be adequately approximated by a
linear function; if there is a clear monotonous nonlinear component, the data first need to be
transformed to remove the nonlinearity. Usually a logarithmic, exponential, or (less often)
polynomial function can be used.

Additive models
The models that we have considered in earlier sections have been additive models, and there has
been an implicit assumption that the different components affected the time series additively.

For monthly data, an additive model assumes that the difference between the January and July
values is approximately the same each year. In other words, the amplitude of the seasonal effect
is the same each year.

The model similarly assumes that the residuals are roughly the same size throughout the series --
they are a random component that adds on to the other components in the same way at all parts
of the series.
 www.rejinpaul.com
Multiplicative models
In many time series involving quantities (e.g. money, wheat production, ...), the absolute
differences in the values are of less interest and importance than the percentage changes.

For example, in seasonal data, it might be more useful to model that the July value is the same
proportion higher than the January value in each year, rather than assuming that their difference
is constant. Assuming that the seasonal and other effects act proportionally on the series is
equivalent to a multiplicative model,

Fortunately, multiplicative models are equally easy to fit to data as additive models! The trick to
fitting a multiplicative model is to take logarithms of both sides of the model,

After taking logarithms (either natural logarithms or to base 10), the four components of the time
series again act additively.

What is an additive model?

A data model in which the effects of individual factors are differentiated and added together to
model the data. They occur in several Minitab commands:

 An additive model is optional for Decomposition procedures and for Winters' method.
 An additive model is optional for two-way ANOVA procedures. Choose this option to
omit the interaction term from the model.
What is a multiplicative model?
This model assumes that as the data increase, so does the seasonal pattern. Most time series plots
exhibit such a pattern. In this model, the trend and seasonal components are multiplied and then
added to the error component.
 www.rejinpaul.com
Should I use an additive model or a multiplicative model?

Choose the multiplicative model when the magnitude of the seasonal pattern in the data depends
on the magnitude of the data. In other words, the magnitude of the seasonal pattern increases as
the data values increase, and decreases as the data values decrease.

Choose the additive model when the magnitude of the seasonal pattern in the data does not
depend on the magnitude of the data. In other words, the magnitude of the seasonal pattern does
not change as the series goes up or down.

If the pattern in the data is not very obvious, and you have trouble choosing between the additive
and multiplicative procedures, you can try both and choose the one with smaller accuracy
measures.

INDEX NUMBERS

Introduction:
Index numbers are meant to study the change in the effects of such factors which cannot be
measured directly. According to Bowley, ―Index numbers are used to measure the changes in
some quantity which we cannot observe directly‖. For example, changes in business activity in a
country are not capable of direct measurement but it is possible to study relative changes in
business activity by studying the variations in the values of some such factors which affect
business activity, and which are capable of direct measurement.

Index numbers are commonly used statistical device for measuring the combined fluctuations in
a group related variables. If we wish to compare the price level of consumer items today with
that prevalent ten years ago, we are not interested in comparing the prices of only one item, but
in comparing some sort of average price levels. We may wish to compare the present agricultural
production or industrial production with that at the time of independence. Here again, we have to
consider all items of production and each item may have undergone a different fractional
increase (or even a decrease). How do we obtain a composite measure? This composite measure
is provided by index numbers which may be defined as a device for combining the variations that
have come in group of related variables over a period of time, with a view to obtain a figure that
represents the ‗net‗ result of the change in the constitute variables.
 www.rejinpaul.com

Index numbers may be classified in terms of the variables that they are intended to measure. In
business, different groups of variables in the measurement of which index number techniques are
commonly used are (i) price, (ii) quantity, (iii) value and (iv) business activity. Thus, we have
index of wholesale prices, index of consumer prices, index of industrial output, index of value of
exports and index of business activity, etc. Here we shall be mainly interested in index numbers
of prices showing changes with respect to time, although methods described can be applied to
other cases. In general, the present level of prices is compared with the level of prices in the past.
The present period is called the current period and some period in the past is called the base
period.

Index Numbers:

Index numbers are statistical measures designed to show changes in a variable or group of
related variables with respect to time, geographic location or other characteristics such as
income, profession, etc. A collection of index numbers for different years, locations, etc., is
sometimes called an index series.

Simple Index Number:

A simple index number is a number that measures a relative change in a single variable with
respect to a base.

Composite Index Number:

A composite index number is a number that measures an average relative changes in a group of
relative variables with respect to a base.

Types of Index Numbers:

Following types of index numbers are usually used:

Price index Numbers:

 www.rejinpaul.com

Price index numbers measure the relative changes in prices of a commodities between two
periods. Prices can be either retail or wholesale.

Quantity Index Numbers:

These index numbers are considered to measure changes in the physical quantity of goods
produced, consumed or sold of an item or a group of items.

Uses

This index number is a useful number that helps us quantify changes in our field. It is easier to
see one value than a thousand different values for each item in our field.

Take the stock market, for example. It is comprised of thousands of different public companies.
We could, of course, look at the stock value of each of these companies to see how the
companies are doing as a whole, or we can look at just one number, the stock index, to get a
general feel for how the companies are doing.

The same goes for the cost of goods. We could look at the cost of each item and compare it to its
cost from last year. But that would mean looking at the cost of millions of items. Or we could
look at the cost of goods index, just one number, to see whether prices have increased or
decreased over the past year.

We can say that the index number is one simple number that we can look at to give us a general
overview of what is happening in our field. Let's take a look at two real world index numbers.

Line of Best Fit (Least Square Method)

A line of best fit is a straight line that is the best approximation of the given set of data.
It is used to study the nature of the relation between two variables. (We're only considering the
two-dimensional case, here.)
 www.rejinpaul.com
A line of best fit can be roughly determined using an eyeball method by drawing a straight line
on a scatter plot so that the number of points above the line and below the line is about equal
(and the line passes through as many points as possible).

A more accurate way of finding the line of best fit is the least square method.
Use the following steps to find the equation of line of best fit for a set of ordered pairs
(x1,y1),(x2,y2),...(xn,yn)(x1,y1),(x2,y2),...(xn,yn).

Step 1: Calculate the mean of the xx-values and the mean of the yy-values.
X¯¯¯=∑i=1nxinY¯¯¯=∑i=1nyinX¯=∑i=1nxinY¯=∑i=1nyin

Step 2: The following formula gives the slope of the line of best fit:
m=∑i=1n(xi−X¯¯¯)(yi−Y¯¯¯)∑i=1n(xi−X¯¯¯)2m=∑i=1n(xi−X¯)(yi−Y¯)∑i=
1n(xi−X¯)2

Step 3: Compute the yy-intercept of the line by using the formula:

b=Y¯¯¯−mX¯¯¯b=Y¯−mX¯

Step 4: Use the slope mm and the yy-intercept bb to form the equation of the line.

Example:

Use the least square method to determine the equation of line of best fit for the data. Then plot
the line.

x 8 22 111 6 5 44 121 9 6 11
x 8 1 6 5 2 9 6
y 3 101 33 6 8 121 11 4 9 141
y 3 0 6 8 2 4 9 4
 www.rejinpaul.com

Solution:
Plot the points on a coordinate plane.

Calculate the means of the xx-values and the yy-values.

X¯¯¯=8+2+11+6+5+4+12+9+6+110=6.4Y¯¯¯=3+10+3+6+8+12+1+4+9+1410=7X¯=8+2+11+6
+5+4+12+9+6+11 0=6.4Y¯=3+10+3+6+8+12+1+4+9+1410=7

Now calculate xi−X¯¯¯xi−X¯, yi−Y¯¯¯yi−Y¯, (xi−X¯¯¯)(yi−Y¯¯¯)(xi−X¯)(yi−Y¯), and

(xi−X¯¯¯)2(xi−X¯)2 for each ii.
 www.rejinpaul.com

ii xix i yiy i x i−X¯ y i−Y¯ (xi−X¯)(yi−Y¯) (xi−X¯¯¯)2

11 88 33 1.61.6 −4−4 −6.4−6.4 2.562.56

22 22 10 −4.4−4.4 33 −13.2−13.2 19.3619.3

33 11 33 4.64.6 −4−4 −18.4−18.4 21.1621.1

44 66 66 −0.4−0.4 −1−1 0.40.4 0.160.16

55 55 88 −1.4−1.4 11 −1.4−1.4 1.961.96

66 44 12 −2.4−2.4 55 −12−12 5.765.76

77 12 11 5.65.6 −6−6 −33.6−33.6 31.3631.3

88 99 44 2.62.6 −3−3 −7.8−7.8 6.766.76

99 66 99 −0.4−0.4 22 −0.8−0.8 0.160.16

10 11 14
−5.4−5.4 77 −37.8−37.8 29.1629.1

Calculate the slope.

m=∑i=1n(xi−X¯¯¯)(yi−Y¯¯¯)∑i=1n(xi−X¯¯¯)2=−131118.4≈−1.1m=∑i=1n(xi−X¯)(yi−Y¯)∑i=1n(xi−
X¯)2=−131 118.4≈−1.1

Calculate the yy-intercept.

Use the formula to compute the yy-intercept.

b=Y¯¯¯−mX¯¯¯ =7−(−1.1×6.4) =7+7.04 ≈14.0b=Y¯−mX¯ =7−(−1.1×6.4) =7+7.04 ≈14.0
Use the slope and yy-intercept to form the equation of the line of best fit.
The slope of the line is −1.1−1.1 and the yy-intercept is 14.014.0.
Therefore, the equation is y=−1.1x+14.0y=−1.1x+14.0.
Draw the line on the scatter plot.