4.

IMPORTANT ASPECTS IN RESEARCH

METHODS
1. Steps Involved in Hypothesis Testing
Formulate H0 and H1
Select Appropriate Test
Choose Level of Significance
Collect Data and Calculate Test Statistic

Determine Probability
Associated with Test
Statistic
Compare with Level of
Significance,

Determine Critical Value

of Test Statistic TSCR

Determine if TSCR falls

into (Non) Rejection
Region

Reject or Do not Reject H0

Draw Marketing Research Conclusion

2. A Classification of Univariate Techniques

Univariate Techniques

Non-numeric Data

Metric Data
Two or More
Samples

One Sample
* t test
* Z test

Independent
* Two- Group test
* Z test
* One-Way
ANOVA

One Sample

Frequency
Chi-Square
K-S
Runs
Binomial

Two or More
Samples

Related
* Paired
t test

Independent
* Chi-Square
* Mann-Whitney
* Median
* K-S
* K-W ANOVA

Related
* Sign
* Wilcoxon
* McNemar
* Chi-Square

3. A Classification of Multivariate Techniques

Multivariate Techniques
Dependence
Technique
One Dependent
Variable
* Cross- Tabulation
* Analysis of
Variance and
Covariance
* Multiple Regression
* Conjoint Analysis

Interdependence
Technique
More Than One
Dependent
Variable

* Multivariate
Analysis of
Variance and
Covariance
* Canonical Correlation
* Multiple Discriminant
Analysis

Variable
Interdependence
* Factor Analysis

Interobject
Similarity
* Cluster Analysis
* Multidimensional Scaling

STATISTICAL TESTS
SUBGROUPS OR
SAMPLES
one

LEVEL
SCALING
Nominal

TEST

Two or more

Nominal

Random sample,
Independent sample

One

Ordinal

K-S

Random sample in
natural order in data

One( Large
Sample)

Metric
(Interval or
Ratio)

Z- Test
for one
mean

Random sample
n30

One (small sample)

Metric
(Interval or
Ratio)
Metric
(Interval or
Ratio)

t-test for
one
mean
Z-test
for one
mean

Random sample
n<30

Two ( small
sample)

Metric
(Interval or
Ratio)

One way
ANOVA
test

Random sample

One (Large
sample)

Metric
(Interval or
Ratio)

Z-test
for one
mean

Random sample
n30

Two ( large
sample)

SPECIAL
REQUIREMENT
Random sample

Random sample
n30

EXAMPLE
Are observed differences
in the numbers
responding to three
different promotions
likely/not likely due to
chance?
Are differences in the
number of men and
women responding to a
promotion likely/not
likely due to chance
Is the observed
distribution of women
preferring an ordered set
of make-up colors(light
to dark) likely/not likely
due to chance?
Is the observed
difference between a
sample estimate of the
mean and some set
standard or expected
value of the mean
likely/not likely due to
chance?
Small sample

Is the observed
differences between the
means for two subgroup
(mean income for men
and women) likely/not
likely due to chance?
Is the observed variation
between the means for
three or more subgroup
(mean expenditure on
entertainment for high,
moderate and low
income people)
likely/not likely due to
chance?
Is the observed variation
between a sample
estimate proportion
(percentage who say they
will buy) and some set
standard or expected

SUBGROUPS OR
SAMPLES

Two ( Large
sample)

LEVEL
SCALING

Metric
(Interval or
Ratio)

TEST

Z-test
for two
mean

SPECIAL
REQUIREMENT

Random sample
n30

EXAMPLE
value likely/not likely
due to chance?
Is the observed variation
between estimated
percentage for two
groups (percentage of
men and women who
have college degrees)
likely/not likely due to
chance?

HYPOTHESIS TESTS RELATED TO DIFFERENCES

Sample

Application

Level of Scaling

Distribution

Nonmetric

Test/Comment
K-S and 2 ,
Runs test for randomness,

One sample

Binomial test for goodness of fit

for dichotomous variables
Means

Metric

t test, if variance is unknown

z-test if variance is known

Proportions

Metric

Distributions

Nonmetric

z-test
K-S two sample test for
examining the equivalence of two

Two
independent

distributions
Means

Metric

samples

F test for equality of variances

Proportions

Metric
Nonmetric

Paired

Two group t test

Means

Metric

Proportions

Nonmetric

Z test
2
Paired t test
McNemar test for binary
variables, 2

samples
Ranking/medians

Non metric

Wilcokson matched pair signed

test is more powerful than the
sign test