8/8/2019 Market Resarch 1
1/23
U NIT II- Research Design
- Primary & Secondary Research
- Basic Methods of Data Collection
8/8/2019 Market Resarch 1
2/23
R ESEARCH D ESIGNTypes of Research Design
8/8/2019 Market Resarch 1
3/23
TYPES OFRESEARCH DESIGN
Exploratory Research Conclusive / Causal Research
a. Search of secondary data a. Descriptive research
b. Survey of knowledgeable person - Case Studyc. Case Study - Statistical Study
b. Experimentation
Usually qualitativedata collected, hence
often calledQualitative research
Usually quantitativedata collected, hence
often calledQuantitative research
8/8/2019 Market Resarch 1
4/23
COMPARISIONRESEARCH DESIGN EXAMPLE
Estimated % change in sales volume for 1% increase in price for four productsbased on three methods of data collection
Product FieldExperimentation LaboratoryExperimentation DescriptiveExperimentation
A -1.57 -1.25 -0.33
B -1.27 -0.64 0.71
C -1.58 -0.76 -1.86
D -1.74 1.13 0.35
If there is no constraint of time and cost then Fieldexperimentation is the best choice
8/8/2019 Market Resarch 1
5/23
T YPES OF D ATA
Primary data:
I nformation developed orgathered by the researcherspecifically for the researchproject at hand.
I nformation gathered andused first time.
Secondary data:
I nformation that has beenpreviously gathered for someother research work.
I nformation that has beenpreviously used and is notfresh or first time for current
research project.
8/8/2019 Market Resarch 1
6/23
M EASUREMENT & S CALINGTypes of Scales
8/8/2019 Market Resarch 1
7/23
M EASUREMENT THE CONCEPT
Measuring what?Person or persons
attribute?
How the attributeis measured?
Values range
We don't measure specificobjects, persons, etc.,
BUTwe measure attributes
(CONCEPTS) or featuresthat define them
What variousattributes of person?
More concrete!Education, age, gender,
height, weight
Abstract!Loyalty, satisfaction,
trust, personality,
CorrespondenceRule of Scale
Certain value onscale correspondto true value of
concept
Which ismore difficult
to define &measure?
8/8/2019 Market Resarch 1
8/23
M EASURING THE ATTRIBUTES
More concretee.g. Weight
More abstracte.g. behaviour
Attributeof person
(or concept)
ScaleMeasuring
differentpossiblevalues of attribute
How much satisfied you are with thebehaviour of sales person?1-Very satisfied2-Satisfied3-Neither satisfied nor dissatisfied4-Dissatisfied5-Very dissatisfied
What is Scale? A device providing range of values that correspond to different values in a
concept being measured
Range of valuesMeasuring the
attributeof
particularconcept of person
8/8/2019 Market Resarch 1
9/23
M EASUREMENT PROCESS
Define conceptsto be measured
ConceptPoor showroom experiencedoes not result in purchase
Defineattributes of the concept
Select scale of measurement
(data type)
Developquestions / items
Designquestionnaire
Pretest & refinequestionnaire
Step 1
Step 2
Step 3
Step 4
Step 5
Step 6
AttributesShowroom experience (ambience, behaviour
and knowledge of sales executive)
ScaleLikert, ordinal or nominal (satisfaction scale,
agreement scale, etc)
QuestionsHow satisfied are you with sales executive?
(Rate using 5-point satisfaction scale)
QuestionnaireHave questions in a structured manner
Finalize the QuestionnaireThis would be used to collect data
finally
Example
8/8/2019 Market Resarch 1
10/23
B ASIC M EASUREMENT CONCEPTS
How much satisfied you are with the behaviour of sales person?How much satisfied you are with the knowledge of sales person?How much satisfied you are with the ambience of showroom?
Concept A generalized idea
that representsomething of
meaning
Researcher should know?
What to measureHow to measure
ConstructIt is a concept that is
measured by usingmultiple indicators
Other terms usedfor indicators?
Variable, attributes,aspects
Does one or more than oneattribute provide more
complete & accurateaccount of concept?
1-Verydissatisfied 2-Verydissatisfied
3-Neither
satisfied nor dissatisfied
3-Satisfied 5-Verysatisfied
These are indicators, attributes,
aspects or variable
8/8/2019 Market Resarch 1
11/23
M EASUREMENT AND S CALING DEFINED
M easurementStandardized process of assigning
numbers or other symbols to certaincharacteristics of objects of interests
according to pre-specified rulesCharacteristics for Standardization
One-to-one correspondence between thesymbol and the characteristic in theobject that is being measured
Rules for assignment should be invariantover time and the objects beingmeasured
ScalingProcess of creating a continuum on
which objects are located according tothe amount of the measuredcharacteristic that the object
possesses
8/8/2019 Market Resarch 1
12/23
T YPES OF SCALES
Nominal Scales
Ordinal Scales
Interval Scales
Ratio Scales
8/8/2019 Market Resarch 1
13/23
T YPES OF SCALES NOMINAL
Numbers identify and classify objects
No ordering or spacing are implied
Only possible arithmetic operation is a count of each category(frequency, %, mode)
Examples:Players in a football team
Colors of traffic light
Gender
8/8/2019 Market Resarch 1
14/23
T YPES OF SCALES O RDINAL
Objects are ranked in order
Indicates the relative position of objects but not the magnitude of difference
Arithmetic operations are limited to statistics such as median or mode
Examples:Result of 100 meter dashRanking of largest fast food companies (most to leastpreferred, where preference is the concept)
8/8/2019 Market Resarch 1
15/23
T YPES OF SCALES I NTERVAL
Object is measured on a continuum
Arbitrary zero point
Differences between objects can be compared
Entire range of statistical operations can be employed (mean,
correlation, ANOVA, regression)Examples:
TemperatureAttitudes
8/8/2019 Market Resarch 1
16/23
T YPES OF SCALES I NTERVAL
Thermometer
Fahrenheit units
on the outer scale
Celsius units onthe inner scale
On the Fahrenheitscale, the freezing
point of water is 32Fahrenheit (F) and the
boiling point 212Fplacing the boiling
and freezing points of water exactly 180F
apart.
On the Celsiusscale, the
freezing andboiling pointsof water are100 degrees
apart.
8/8/2019 Market Resarch 1
17/23
T YPES OF SCALES R ATIO
Interval scale with fixed zero pointRatio of scale values can be computed
Examples:Weight, lengthAgeStore salesMarket shares
8/8/2019 Market Resarch 1
18/23
T YPES OF SCALES R ATIO
ExampleAntique watch X sold for $50
Authentic watch Y sold for $75
Antique watch Z sold for $100
Antique watch B did not sell andthere were no takers for free
Researcher should know?What is the - nominal conclusion?
- ordinal conclusion?- interval conclusion?- ratio conclusion?
8/8/2019 Market Resarch 1
19/23
T YPES OF SCALES R ATIO
ExampleAntique watch X sold for $50
Authentic watch Y sold for $75
Antique watch Z sold for $100
Antique watch B did not sell andthere were no takers for free
Researcher should know?What is the - nominal conclusion?
- ordinal conclusion?- interval conclusion?- ratio conclusion?
8/8/2019 Market Resarch 1
20/23
T YPES OF SCALES P ROPERTIES
Type of Scale Numerical Operation Descriptive Statistics
Nominal Counting Frequency in each category,percentage in each category, mode
Ordinal Rank Ordering M edian, range, percentile ranking
Interval Arithmetic Operationson Intervals betweennumbers
M ean, standard deviation, variance
Ratio Arithmetic Operationson actual quantities
Geometric mean, coefficient of variation
8/8/2019 Market Resarch 1
21/23
CRITERIA FOR SCALE SELECTION
Understanding of the questionsDiscriminatory power of scale descriptorsBalanced versus unbalanced scales
Forced or nonforced choice scalesDesired measure of central tendency anddispersion
8/8/2019 Market Resarch 1
22/23
T YPES OF SCALES L IKERT S CALE
A likert scale isan ordinal scale format that,
asks respondents
to indicate the extent to whichthey agree or disagree
with a series of mental or behavioral belief
statements about a given object
8/8/2019 Market Resarch 1
23/23
T YPES OF SCALESL IKERT SCALE E XAMPLE