RESEARCH DESIGNS:

EXPLORATORY & DESCRIPTIVE

What Is Research Design?

Blueprint

Plan

Guide

Framework
 Research Design Descriptors

Question Data
Perceptual Crystallization Collection
Awareness
Method

Purpose of Experimental
Study Descriptors Effects

Research Time
Environment Dimension
Topical Scope
 Three Tenets of Research Designs

1. Convert the research question and stated

assumptions /hypotheses into operational variables
that can be measured;
2. Specify the process that would be followed to
complete the above task, as efficiently and
economically as possible; and
3. Specify the ‘control mechanism(s)’ that would be
used to ensure that the effect of other variables that
could impact the outcome of the study has been
minimized/negated.
 Classification of Research Designs
Research Design

Exploratory Conclusive

Descriptive Causal

Cross - Sectional Longitudinal

Single Cross- Multi Cross-

Sectional Sectional
 Types of Research Designs

Exploratory research designs: are the simplest, most

flexible and most loosely structured designs. As the
name suggests, the basic objective of the study is to
explore and obtain clarity on the problem situation.

Descriptive research designs: are more structured and

formal in nature. As the name implies the objective of
these studies is to provide a comprehensive and
detailed explanation of the phenomena under study.
 Exploratory Research Designs

Secondary resource analysis: Secondary sources of

data give information –in terms of details of
previously collected findings in facts and figures –
which has been authenticated and published.
Case method: it is intricately designed and reveals a
comprehensive and complete presentation of facts, as
they occur, in a single entity.This could be an
individual, an organisation or an entire country.
 Exploratory Research Designs
Expert opinion survey: valuable insights obtained
from experts which might be based on their experience
in the field or based on academic work done on the
concept.

Focus group discussions: a carefully selected

representative sub set of the larger respondent gather
to discuss together, in a short time frame, the
subject/topic to be investigated.
 Descriptive Research Designs
• Cross-sectional research designs: two criteria
• carried out at a single moment in time, therefore the
applicability is temporal specific
• Conducted on a sub-section of the respondent
population
 Descriptive Research Designs

• Longitudinal studies: three criteria

• The study involves selection of a representative

group as a panel.
• There are repeated measurement of the researched
variable on this panel over fixed intervals of time.
• Once selected the panel composition needs to stay
constant over the study period.
 Experimental Research Designs

An experiment is generally used to infer a

causality. In an experiment, a researcher
actively manipulates one or more causal
variables and measures their effects on the
dependent variable of interest.
 Concepts Used in Experiments

• Independent variables: Independent variables are

also known as explanatory variables or treatments.
The levels of these variables are manipulated
(changed) by researchers to measure their effect on
the dependent variable.
• Test units: Test units are those entities on which
treatments are applied.
• Dependent variables: These variables measures the
effect of treatments (independent variable) on the test
units.
 Concepts Used in Experiments
Experiment: An experiment is executed when the
researcher manipulates one or more independent
variables and measures their effect on the dependent
variables while controlling the effect of the extraneous
variables.
Extraneous variables: These are the variables other
than the independent variables which influence the
response of test units to treatments.
Examples: Store size, government policies,
temperature, food intake, geographical location, etc.
 Validity in Experimentation

Internal validity: Internal validity tries to examine

whether the observed effect on a dependent variable is
actually caused by the treatments (independent
variables) in question.
External validity: External validity refers to the
generalization of the results of an experiment. The
concern is whether the result of an experiment can be
generalized beyond the experimental situations.
 Factors Affecting Internal Validity
• History
• Maturation
• Testing
• Instrumentation
• Statistical regression
• Selection bias
 Factors Affecting External Validity
• The environment at the time of test may be different from
the environment of the real world where these results are to
be generalized.
• Population used for experimentation of the test may not be
similar to the population where the results of the experiments
are to be applied.
• Results obtained in a 5-6 week test may not hold in an
application of 12 months.
• Treatment at the time of the test may be different from the
treatment of the real world.
 Environments of Conducting Experiments

• Laboratory Environment - In a laboratory

experiment, the researcher conducts the experiment
in an artificial environment constructed exclusively
for the experiment.

• Field Environment - The field experiment is

conducted in actual market conditions. There is no
attempt to change the real-life nature of the
environment.
 Degree of Question Crystallization

Exploratory Study Formal Study

• Loose structure •Precise procedures
• Expand •Begins with
understanding hypotheses
• Provide insight •Answers research
• Develop hypotheses questions
 Approaches for Exploratory Investigations

• Participant observation
• Film, photographs
• Projective techniques
• Psychological testing
• Case studies
• Ethnography
• Expert interviews
• Document analysis
 Desired Outcomes of Exploratory Studies
Established range and scope of possible management
decisions

Established major dimensions of research task

Defined a set of subsidiary questions that can guide

research design

Develop hypotheses about possible causes of

management dilemma

Learn which hypotheses can be safely ignored

Conclude additional research is not needed or not

feasible
Commonly Used Exploratory Techniques

Secondary
Data Experience
Analysis Surveys

Focus
Groups
 Experience Surveys
• What is being done?
• What has been tried in the past with or without
success?
• How have things changed?
• Who is involved in the decisions?
• What problem areas can be seen?
• Whom can we count on to assist or participate in
the research?
 Focus Groups

• Group discussion
• 6-10 participants
• Moderator-led
• 90 minutes-2 hours
 Data Collection Method

Monitoring Communication
 The Time Dimension

Cross-sectional Longitudinal

The Topical Scope

Statistical Study Case Study
Breadth Depth
Population inferences Detail
Quantitative Qualitative
Generalizable findings Multiple sources of
information
 The Research Environment

Field conditions

Lab conditions

Simulations
 Purpose of the Study

Reporting Descriptive

Casual - Causal -
Explanatory Predictive
 Descriptive Studies

Who?

How much? What?

When? Where?
 Descriptive Studies

Descriptions of population characteristics

Estimates of frequency of characteristics

Discovery of associations among variables

 Types of Asymmetrical Causal Relationships

Relationship Nature of Examples

Type Relationship
Stimulus- An event or change • A change in work rules leads to a
response results in a higher level of worker output.
response from • A change in government
some object. economic policy restricts corporate
financial decisions.
• A price increase results in fewer
unit sales.
Property- An existing • Age and attitudes about saving.
disposition property causes a • Gender attitudes toward social
disposition. issues.
• Social class and opinions about
taxation.
 Types of Asymmetrical Causal Relationships
Relationship Nature of Examples
Type Relationship
Disposition- A disposition • Opinions about a brand and
behavior causes a specific its purchase.
behavior. • Job satisfaction and work
output.
• Moral values and tax
cheating.
Property- An existing • Stage of the family life cycle
behavior property causes and purchases of furniture.
a specific • Social class and family
behavior. savings patterns.
• Age and sports participation.
 Research Design Descriptors
Category Options
The degree to which the research • Exploratory study
question has been crystallized • Formal study
The method of data collection • Monitoring
• Communication Study
The power of the researcher • Experimental
to produce effects in the • Ex-post facto
variables under study
The purpose of the study • Reporting
• Descriptive
• Causal-Explanatory
• Causal-Predictive
 Research Design Descriptors
Category Options
The time dimension • Cross-sectional
• Longitudinal
The topical scope—breadth and • Case
depth—of the study • Statistical study
The research environment • Field setting
• Laboratory research
• Simulation
The participants’ perceptional • Actual routine
awareness of the research activity • Modified routine
 Statistical Designs
Statistical designs allow for statistical
control and analysis of external variables.
• Completely Randomized Design
• Randomized Block Design
• Latin Square Design
• Factorial Design
 Completely Randomized Design in a One-way ANOVA

• In this design, there is one dependent variable and

one independent variable.
• The dependent variable is metric (interval/ratio
scale) whereas the independent variable is
categorical (nominal scale).
• A sample is drawn at random from each category of
the independent variable.

15-35
 The total variation in the data set is called the total sum of
squares (TSS) and is computed as:

Where,
xij = the jth observation of the ith sample
T•• = Grand total of all the data
k = Number of treatments (samples)
n = Number of observations in each sample

15-36
 The variation between the sample means which is attributed to
specific sources or causes is referred to as the treatment sum
of squares (TrSS).

Where, Ti• = Total of observations for the ith treatment.

The variation within the sample, which is attributed to chance,

is referred to as the error sum of squares (SSE).

15-37
 In order to test the null hypothesis,
H0 : μA = μB = μC = μD
against the alternative hypothesis
H1 : At least two means are not equal
(Treatment means are not equal)
The results could be summarized in the table below called One-way
ANOVA.
k=Number of treatments (samples), n =Number of observations in each sample

15-38
 • For a given level of significance, α, the computed F statistic
is compared with the table value of F with k – 1 degrees of
freedom in the numerator and k (n – 1) degrees of the
freedom for the denominator. If the computed F value is
greater than the tabulated F value, the null hypothesis is
rejected.

Case of Unequal Sample Size from Treatments

• If there are ni observations corresponding to ith treatment,
the computing formula for the sum of squares would look
like:

15-39
 SSE = TSS – TrSS
Where, N = n1 + n2 + . . . . + nk
The total number of degrees of freedom in the case is N – 1,
and the degrees of freedom are k – 1 for the treatments and N
– k for the error.
The ANOVA table can be set up with the above information
and if computed F is greater than the table value of F, null
hypothesis could be rejected.

15-40
 Example 4.2 Alpha Industries Limited has deputed four different
batches of its employees to four different training organizations
(A, B, C and D) for the same training program, which aims to
train them in improving their decision-making skills. Each batch
consists of five employees with similar qualifications and work
experience. After the training program, the company conducted a
common examination to test their level of additional knowledge
gained through the training program. The percentage scores of
the employees of the batches are summarized in Table 4.8.

15-41
 Table 4.8 Percentage Scores of Employees of Batches in the Examination

Training organization
A B C D
80 70 65 90
90 60 50 89
96 55 58 85
85 85 55 95
70 90 40 80

Perform ANOVA to check whether there is a significant difference between the

training organizations in terms improving the decision-making skills of the
employees by assuming a significance level of 0.05.

15-42
 Table 4.9 Column Totals of Table 4.8
Training organization
A B C D
80 70 65 90
90 60 50 89
96 55 58 85
85 85 55 95
70 90 40 80
Yj 421 360 268 439

15-43
 Null hypothesis, Ho : T1 = T2 = T3 = T4

Alternate hypothesis, H1:

Treatment means are not equal for at least one pair of
treatment means.

15-44
 5 4

Y ..
  
i 1 j 1
Yij  1488

Y .1
 421,Y .2  360,Y .3  268 and Y .4  439

421
Y 1   84.2
5
360
Y 2   72
5
268
Y 3   53.6
5
439
Y 4   87.8
5
(Y  Y  Y  Y )

1 2 3 4
Y
rc
1488
Y   74.4
15-45 20
 an estimate of population variance from the variance between
the means or columns.

r  (Y  Y )
2

 
2 j

(C  1)
5[(84.2  74.4)  (72  74.4)  (53.6  74.4)  (87.8  74.4) ]
2 2 2 2

 
2

(C  1)
(5  714)
   1190
2

An estimate of the population variance from the variance

within the samples or columns is obtained by averaging the
15-46 four sample variances:
  (Yi1  Y1 )
2
(80  84.2) 2  (90  84.2) 2  (96  84.2) 2  (85  84.2) 2  (70  84.2) 2
S 2

1
r 1 5 1
393.2
S12   98.3
4
 (Yi 2  Y2 )
2
(70  72) 2  (60  72) 2  (55  72) 2  (85  72) 2  (90  72) 2
S 22  
r 1 5 1
930
S 22   232.5
4
 (Yi 3  Y3 )
2
(65  53.6) 2  (50  53.6) 2  (58  53.6) 2  (55  53.6) 2  ( 40  53.6) 2
S 32  
r 1 5 1
349.2
S 32   87.3
4
 (Yi 4  Y4 )
2
(90  87.8) 2  (89  87.8) 2  (85  87.8) 2  (95  87.8) 2  (80  87.8) 2
S 42  
r 1 5 1
126.8
S 42   31.7
4
15-47
 S12  S 22  S 32  S 42 98.3  232.5  87.3  31.7
2  
4 4
449.8
2   112.45
4
variance between sample means
F
variance within samples
1190
F  10.58
112.45

15-48
 Degrees
Source of Sum of Mean sum of
of F ratio
variation squares squares (MSS)
freedom
Between 1190/112.43 =
3 3570.0 3570/3 = 1190
treatments 10.58
Within
16 1798.8 1798.8/16 = 112.43
treatments
Total 19 5368.8

In Table 4.10, the value of the calculated F ratio is 10.58. The value of F
ratio from the statistical table for a significance level of 0.05 and degrees
of freedom (3,16) is 3.24. Then FC (10.58) > FT = (3.24) at 0.05 and df = (3,16)
15-49
 Randomized Block Design in Two-way ANOVA

• The main limitation of the completely randomized design is

that all extraneous variables were assumed to be constant
over all the treatment groups which may not be true.
• There may be extraneous variables influencing the
dependent variable.
• In the randomized block design it is possible to separate the
influence of one extraneous variable on a particular
dependent variable, thereby providing a clear picture of the
impact of treatment on test units.

15-50
 • Assume that the sales of a product is influenced not only by the price
level (low, medium and high) but also by the size of the stores. The
price level is assumed to be independent variable whereas the size of
the stores are extraneous variable and could be treated as different
blocks.
• The hypothesis to be tested are the following:
I Price
H0 : μA = μB = μC (Assuming three price level)
H1 : At least the two means are not same.
II (Blocks or Stores)
H0 : ν1 = ν2 = ν3 (Assuming three different sizes of stores)
H1 : At least two means are not same.

15-51
 • The total sum of squares (TSS) and treatment sum of
squares (TrSS) could be computed as discussed in
completely randomized design. Now, we would also have
another component called block sum of squares (SSB)
which is due to different store sizes and is computed as:

Where, T•j = Total of the values in the jth block.

The error sum of squares would be computed as:
SSE = TSS – TrSS – SSB
15-52
 The above results could be presented in a two-way ANOVA table as
below:

For a given level of significance, the computed F corresponding to

treatments and blocks could be compared with the corresponding
tabulated values to accept of reject the two hypothesis.
15-53
 Example 4.4 A business school recently conducted a management
meet titled ‘Synergy’ in which 'stock exchange simulation game’
was an event. Students from 15 different business schools
participated in the meet. Five different teams (T1, T2, T3, T4 and
T5)have been formed from these 15 schools to participate in the
simulation game. Each team consisted of a team leader (L1, L2,
L3, L4 and L5) and four members. The details of the game were as
follows: Each team was given 10 lacs rupees.

Daily quotations were displayed in an online network system.

Rumors, political news and industrial news were broadcasted over
a mike. Based on these information, the teams effected purchases
and sales. At the end of the game, the net worth of each team was
assessed.
Then, team leaders were rotated to serve in all other teams as
15-54 team leaders and the game was repeated four more times.
 It is clear that the experimentation is a randomized complete block
design. The net worth of the teams (in lacs of rupees) after the
experiment are, as summarized in Table 4.16.

Table 4.16 Results of Simulation Game

Team Leader

L1 L2 L3 L4 L5

T1 40 30 55 25 35

T2 45 60 10 22 33
Team T3 38 55 40 55 28
T4 30 27 32 56 17

T5 45 34 20 34 37
15-55
 Perform the randomized complete block design to check whether
there is a significant difference between the team leaders (L1, L2, L3,
L4 and L5) as well as between different teams (T1, T2, T3, T4 and T5) in
terms of final net worth by assuming a significance level of 0.05.

The data of this example is reproduced with row, column and grand
means, as shown in Table 4.17.
Table 4.17 Results of Simulation Game
Team leader
L1 L2 L3 L4 L5
T1 40 30 55 25 35 37
T2 45 60 10 22 33 34
Team T3 38 55 40 55 28 43.2
T4 30 27 32 56 17 32.4
T5 45 34 20 34 37 34
39.6 41.2 31.4 38.4 30 36.12

15-56 For the design, the following calculations are done based on the Table 4.17:
 The column mean for each Team leader is given by

 Yij
Y.j  i

r
The row mean for each Team is given by

 Yi
Y i. 
j j

c
The grand mean is given by

 Y i. Y . j
Y 
r c
15-57
 The subscripted dots signify that more than one factor is being
considered.

The results are shown in the table

Output with 5 teams leaders and 5 teams (with Row, Column, and Grand
Means)

Table 4.17 Results of Simulation Game

Team leader
L1 L2 L3 L4 L5
T1 40 30 55 25 35 37
T2 45 60 10 22 33 34
Team T3 38 55 40 55 28 43.2
T4 30 27 32 56 17 32.4
T5 45 34 20 34 37 34
39.6 41.2 31.4 38.4 30 36.12

15-58 For the design, the following calculations are done based on the Table 4.17:
Total Sum of the Squares SST    (Yij  Y ) 2
(40  36.12) 2  15 (30  36.12) 2  37.5
(45  36.12) 2  78.8 (60  36.12) 2  570.3
(38  36.12) 2  3.5 (55  36.12) 2  356.5
(30  36.12) 2  37.5 (27  36.12) 2  83.2
(45  36.12) 2  78.9 (34  36.12) 2  4.5
213.7 1052
(55  36.12) 2  356.5
(10  36.12) 2  682.3
(40  36.12) 2  15.1
(32  36.12) 2  17
(20  36.12) 2  259.9
1330.8
 (25  36.12) 2  123.7 (35  36.12) 2  1.3
(22  36.12) 2  199.4 (33  36.12) 2  9.7
(55  36.12) 2  356.5 (28  36.12) 2  65.9
(56  36.12) 2  395.2 (17  36.12) 2  365.6
(34  36.12) 2  4.5 (37  36.12) 2  1.3
1079.3 443.8

SST  213.7  1052  1330.8  1079.3  443.8

SST = 4119.6

15-60
 SSA  r  (Y . j  Y ) 2 (between-column variations)

SSA  5[(39.6  36.12) 2  (41.2  36.12) 2  (31.4  36.12) 2  (38.4  36.12) 2  (30  36.12) 2 ]

SSA = 514.25

SSB  c  (Y i.  Y ) 2
(between-row variations)

SSB  5[(37  36.12) 2  (34  36.12) 2  (43.2  36.12) 2  (32.4  36.12) 2  (34  36.12) 2 ]

SSB = 368.05

SSE = SST - SSA - SSB =4119.6 – 514.25 – 368.05 = 3237.3

15-61
 df of SSA = c -1 = 5 – 1 = 4
df of SSB = r – 1 = 5 – 1 = 4
df of SSE =( r-1) (c – 1) = 4*4 = 16
df of SST = rc - 1 = 25 – 1 = 24
SSA 514.25
MSA    128.6
c 1 4
SSB 368.05
MSB    92
r 1 4
SSE 3237.3
MSE    202.33
(c  1)( r  1) 16

15-62
 MSA 128.6
  0.64 F ratio for factor A (leader)
MSE 202.33
MSB 92
  0.45 F ratio for factor B (team)
MSE 202.33

15-63
Source of Degrees. of Sum of Mean sum of
F ratio
variation freedom squares squares (MSS)

(between
4 514.24 514.24/4 = 128.56 128.56/202.24 = 0.64
columns)

(Between
4 368.64 368.64/4 = 92.16 92.16/202.24 = 0.45
rows)

Error 16 3235.76 3235.76/16 = 202.24

Total 24 4118.64
 In Table 4.18, the value of the calculated F ratio of the treatment is
0.64, whereas its table value with the significance level of 0.05 and
degrees of freedom (4,16) is 3.01. The calculated F ratio for the block
is 0.45, whereas its table value with the significance level of 0.05 and
degrees of freedom (4, 16) is 3.01.
Component-treatment (team leader): For this component,
FC [0.64] < FT = [3.01] at 0.05 and df =(4,16)
Hence, the null hypothesis, Ho should be accepted.
Inference: This means that there is no significant difference in
terms of net worth between the team leaders.
Component-block (team): For this component,
FC [0.45] < FT = [3.01] at 0.05 and df = (4,16)
Hence, the null hypothesis, Ho should be accepted.
Inference: This means there is no significant difference in
15-65 terms of net worth between different teams.
 Factorial Design

• In factorial design, the dependent variable is the

interval or the ratio scale and there are two or more
independent variables which are nominal scale.
• It is possible to examine the interaction between the
variables. If there are two independent variables,
each having three cells, there would be a total of
nine interactions.
• The hypothesis to be tested would be the following:

15-66
 Row-wise:
H0 : Average of rows are equal.
H1 : Average of rows are not equal
Column-wise:
H0 : Average of columns are equal.
H1 : Average of columns are not equal
Interaction:
H0 : Average of all interactions between rows and columns
are equal.
H1 : Average of all interactions between rows and columns
are not equal
15-67
 • Total sum of squares (TSS), sum of squares due to rows (SSR) & sum
of squares due to columns (SSC) can be computed as explained
earlier. We also need to compute sum of squares due to interaction
(SSI).

• Sum of squares due to error (SSE) can be computed as :

15-68
SSE = TSS – SSR – SSC – SSI
 Latin Square Design

The Latin square design has three important characteristics:

1. The number of categories must be equal for the two
extraneous (control) variables.
2. The number of experimental (treatment) groups should
equal to the numbers of categories in the control variables.
3. Each experimental (treatment) group must appear only once
in every row and column.

15-69
 • Assuming that we are interested in studying the impact of
the price categorized as low (A), medium (B) and high (C).
Two extraneous variables, namely, the store size and the
type of packaging, could also influence sales.
• The layout of the Latin square design for this problem could
be as shown in the table below:

15-70
 • The hypothesis to be tested is:
H0 : All the treatments price levels have an equal effect on sales.
H1 : All the price levels do not have equal effect on sales.
We need to compute the following:

15-71
 • The ANOVA table for Latin square design could be set up
as shown in the next slide.
15-72
 For a given level of significance, the computed F statistic
corresponding to treatment is compared with the tabulated
value to accept or reject the null hypothesis.
15-73
Example 4.6. The production manager of a firm is keen in analyzing the
hourly production volume of a product. He feels that operators (A, B, C,
D and E), machines (M1, M2, M3, M4 and M5) and suppliers of raw
materials (S1, S2, S3, S4 and S5) will affect the production volume of the
product. Hence, he designed a Latin square design, as shown in Table
4.23. The corresponding observations are shown in Table 4.24.

Table 4.23 Experimental Combinations in Latin Square Design (Example 4.6)

Supplier
S1 S2 S3 S4 S5
M1 A B C D E
M2 B C D E A
Machine M3 C D E A B
M4 D E A B C
M5 E A B C D
Table 4.24 Hourly Production Volume as per Latin Square Design (Example 4.6)

Supplier

S1 S2 S3 S4 S5

M1 12 (A) 8 (B) 9 (C) 15 (D) 18 (E)

M2 14 (B) 7 (C) 13 (D) 10 (E) 12 (A)

Machine M3 7 (C) 15 (D) 12 (E) 8 (A) 6 (B)

M4 16 (D) 17 (E) 13 (A) 11 (B) 12 (C)

M5 9 (E) 14 (A) 8 (B) 10 (C) 8 (D)

 Perform statistical analysis to test the following at a
significance level of 0.05

• Whether there is significant difference between operators in

terms of production volume of the product;
• Whether there is significant difference between machines in
terms of production volume of the product;
• Whether there is significant difference between suppliers in
terms of production volume of the product.

15-76
 Hypothesis with respect to treatment (operator):
Null hypothesis, Ho: 01 = O2 = 03 = 04 = 05
Alternate hypothesis, H1: Treatment means are not equal for at least
one pair of treatment means.

Hypothesis with respect to rows (machine):

Null hypothesis, H o : M1 = M2 = M3 = M4 = M5
Alternate Hypothesis, H1: Row means are not equal for at least one
pair of row means.

Hypothesis with respect to columns (supplier):

Null hypothesis, H0: S1 = S2 = S3 = S4 = S5
Alternate hypothesis, H1 : Column means are not equal for at least
one pair of column means.

15-77
 The column mean for each supplier is given by

 Yij
Y.j  i

r
The row mean for each machine is given by

 Yi
Y i. 
j j

c
The grand mean is given by

 Y i. Y . j
Y 
r c
15-78
 The subscripted dots signify that more than one factor is being considered.

The results are shown in the table

Output with 5 suppliers and 5 machines (with Row, Column, and Grand Means)
Table 4.25 Hourly Production Volume in Latin Square Design {Example 4.6)
Supplier

S1 S2 S3 S4 S5 Y 1. 

M1 12 (A) 8 (B) 9 (C) 15 (D) 18 (E) 12.4

M2 14 (B) 7 (C) 13 (D) 10 (E) 12 (A) 11.2

Machine M3 7 (C) 15 (D) 12 (E) 8 (A) 6 (B) 9.6

M4 16 (D) 17 (E) 13 (A) 11 (B) 12 (C) 13.8

M5 9 (E) 14 (A) 8 (B) 10 (C) 8 (D) 9.8

Y .1  11.6 12.2 11 10.8 11.2 11.4

15-79
 SST    (Yij  Y ) 2

SST = 271.8

SSA  r  (Y . j  Y ) 2 (between-column suppliers variations)

SSA = 6.2

SSB  c  (Y i.  Y ) 2 (between-row machines variations)

SSB = 63

15-80
 The data of Table 4.24 is rearranged as shown in Table 4.26, to compute the
sum of squares of 'operator'.

Table 4.26 Rearranged Hourly Production Volume in Latin Square

Design (Example 4.6)

Operator

A B C D E

12 8 9 15 18

12 14 7 13 10

8 6 7 15 12

13 11 12 16 17

Y .1  14 8 10 8 9

11.8 9.4 9 13.4 13.2

15-81
 SSC  r  (Y . j  Y ) 2
(between-column operators variations)

SSC = 85.8

SSE = SST - SSA – SSB - SSC

SSE =271.8 – 6.2 – 63 – 85.8 = 116.8

15-82
 df of SSA = c -1 = 5 – 1 = 4
df of SSB = r – 1 = 5 – 1 = 4
df of SSC = c –1 = 5 – 1 = 4
df of SSE =( r-1) (c – 1) = 4*4 = 16
df of SST = rc - 1 = 25 – 1 = 24

SSA 6.2
MSA    1 .5
c 1 4
SSB 63
MSB    15.8
r 1 4
SSC 85.8
MSC    21.5
c 1 4
SSE 116.8
MSE    7 .3
(c  1)( r  1) 16
15-83
 MSA 1.5
  0.21 F ratio for Suppliers
MSE 7.3
MSB 15.8
  2.2 F ratio for Machines
MSE 7. 3
MSC 21.5
  2.9 F ratio for Operators
MSE 7.3

The ANOVA results of this problem are summarised in Table 4.27.

15-84
 Table 4.27 Results in Latin Square Design (Example 4.6)

Degrees
Source of Sum of Mean sum of
of F ratio
variation squares squares (MSS)
freedom

MSC=SSC/4 MSC/MSE
Between
4 85.76
operators
85.76/4 = 21.44 21.44/7.3 = 2.9
MSB=SSB/4 MSB/MSE
Between
4 62.96
machines
62.96/4 = 15.74 15.74/7.3 = 2.2
MSA=SSA/4 MSA/MSE
Between suppliers 4 6.i6
6.16/4 = 1.54 1.54/7.3 = 0.21

Error 16 116.88 116.88/16 = 7.3

Total 24 SST=271.8

15-85
 In Table 4.27, the value of the calculated F ratio for 'operator' is 2.9,
whereas its table value with the significance level of 0.05 and degrees of
freedom (4,16) is 3.01. The calculated F ratio for 'machine' is 2.2,
whereas its table value with the significance level of 0.05 and degrees of
freedom (4,16) is 3.01. The calculated F ratio for 'supplier' is 0.21,
whereas its table value with the significance level of 0.05 and degrees of
freedom (4,16) is 3.01.

Component-treatment (operator): For this component, Fc[2.9] < FT

[3.01] at 0.05 and df (4,16). Hence, the null hypothesis, Ho should be
accepted.
Inference: This means that there is no significant difference between
operators in terms of hourly production volume.

15-86
 Component-rows (machines): For this component, Fc[2.2] <
FT [3.01] at 0.05 and df (4,16). Hence, the null hypothesis,
(Ho) should be accepted.
Inference: This means that there is no significant
difference between machines in terms of hourly production
volume.

Component-columns (supplier): For this component,

Fc[0.21] < FT [3.01] at 0.05 and df (4,16). Hence, the null
hypothesis, (Ho) should be accepted.
Inference: This means there is no significant difference
between suppliers in terms
15-87 of hourly production volume.