You are on page 1of 17

Marketing Research

Chapter 5: Sampling Theory &


Sampling Techniques
Samples and Sampling: Definitions

Population / Universe: The totality of the


people or objects we wish to study

Sample: A subgroup of the elements of the


population / universe selected for
participation in the study
Samples and Sampling: Definitions

Target Group: The collection of elements


that have the information that the researcher
is looking for

Sampling Element: The respondents you are


trying to reach

Sampling Unit: The basic unit that has the


elements of the population to be sampled (e.g.
households, families, companies, individuals)
Sampling Design: 6 Steps

1. Define the Target Group and Sampling Unit


2. Define the Sampling Frame
3. Decide the Sampling Method
-Probablility or Non Probability-
4. Decide the Size of the Sample
5. Define the Sampling Plan
6. Select the Sample
1. Define the Target Group & Sampling Unit

Example: Determine the Population, Target Group, Sampling Unit and Sample
For a study on Mobile Phone Usage

Population: All Cambodians


Sample: Subgroup of Cambodians
between 16-64 Years of age

Sampling Unit: Individuals because


a mobile is a personal product

Target Group: All Cambodians between


16-64 Years of age, Males & Females
1. Define the Target Group & Sampling Unit

Example: Determine the Population, Target Group, Sampling Unit and Sample
For a study on Consumer Refrigerators

Population: All Cambodians who live


in areas with electricity or own
generators Sample: Subgroup of All Male
Heads of Household between 18-64
Years of age who live in areas with
electricity or own generators

Sampling Unit: Head of the


Household or Household Decision
Maker because a refrigerator is a
Household product

Target Group: All Male Heads of Household


between 18-64 Years of age who live in areas with
electricity or own generators
1. Define the Target Group & Sampling Unit

Example: Determine the Population, Target Group, Sampling Unit and Sample
For a study on Bicycle Manufacturers

Population: All Bicycle


Manufacturers
In Cambodia Sample: Subgroup of Bicycle
manufacturers in urban centers

Sampling Unit: The company

Target Group: All Bicycle Manufacturers in urban


centres (business studies usually focus on urban
centers)
2. Define the Sampling Frame

Sampling Frame: A representation of the elements


(respondents you are trying to reach) of the target
group that consists of a list or set of directions for
identifying the target group.

Sampling Frame Guidelines:


• Each element included only once
• No element excluded
• Sampling frame covers whole target population
• Sampling frame is up-to-date and accurate
Sampling Error
 We cannot expect any given sample to be exactly
representative of the population from which it was drawn

 Nor can we expect two samples, independently drawn


from the same population to be exactly alike

 Therefore each sample drawn form the universe is


expected to be ‘wrong’ to a certain degree

 This is the concept of Sampling Error


3. Select the Sampling Technique

Sampling
Techniques

Non Probability Probability


Sampling Sampling
Techniques Techniques

Sampling relies on the Each element of the population


personal judgement of has an equal probability of being
the researcher: selected for the sample:
Non Statistical: Results cannot Statistical – Results can be
Be projected to population; sampling projected to population
error cannot be measured
Probability Sampling Techniques

Probability
Sampling Techniques

Simple Random Systemic Stratified


Sampling Sampling Sampling

Sample is chosen by Population is divided


Each respondent has
selecting a random into ‘strata’ and chosen
an equal probability
staring point and an randomly from within
of selection
interval from then on each strata
Probability (Random) Sampling

 Population elements (people / objects) have


an equal chance of being selected

 Can estimate how closely the sample result


resembles the population result (i.e. sample
error)

 Allows inferences (conclusions) to be made


about the whole population
Non Probability Sampling Techniques

Non Probability
Sampling
Techniques

Convenience Judgemental Quota


Sampling Sampling Sampling
Quotas are developed
Selection of
Selection of by the researcher and
Sampling
Sampling respondents are
Is left to the
Is left to the selected on this basis
judgement of the
interviewer
researcher
Simple Random Sampling: An Example

• Imagine that our population or universe is


2,000 consumers in the country using
Mobile Phones lines
• Imagine that these users are classified into
6 groups, for example Social Class
• As in real life, consumers of the same
social class concentrate in certain areas
(e.g. rich areas or poor areas)
Simple Random Sampling: Conclusions

• The first point to note is how thin the sample looks


• Secondly, how closely it reflects the universe
structure in spite of its apparent thinness

• Samples work up to a point but there are some


worrying differences

• The sample of 120, as can be seen, is more


comparable to the universe structure

• Therefore, the larger the sample the more accurate


the picture
Sample Size: Things to Consider

• Importance of the decision


• Nature of the research
• Number of variables being investigated
• Nature of the analysis
Sample Sizes Used in Marketing Research
Studies
Type of Study Minimum Size Normal Range
Problem identification 500 1000-2500
Research (e.g. Market share)

Problem Solving Research (e.g. 200 300-500


Pricing)

Product Tests 200 300-500

Test Marketing Studies 200 300-500

TV/Radio/Print Advertising 150 200-300


(per commercial or ad tested)

Focus Groups 6 10-15

You might also like