Page 29 - DMGT506_CONSUMER_BEHAVIOUR

Consumer Behaviour

Notes Pepsodent
Neem
Vicco Vajradanti

Sample Design

A sample design addresses three questions: who is to be surveyed (sampling unit), how many to
survey (sample size) and how should the respondents be chosen (the sampling procedure).
Deciding whom to survey (sampling unit) requires that the researcher must define the target
population (universe) that would be sampled. For example, if the Indian Airlines conduct a
survey, should the sampling unit be business travellers, vacation travellers, or both? Should
travellers under age 30 years be interviewed? Interviewing the correct target market or the
potential target market is basic to the validity of research.
Deciding how many people should be surveyed (sample size) depends on the budget and the
required confidence in research findings. As a rule, large samples give more reliable results
than small samples. If the sampling procedure is credible, sample sizes of less than 1% of a
population can give reliable results.
How should the respondents be chosen (sampling procedure)? If the researcher wants to project
the findings to the total population, then a probability sample should be selected. If the researcher
wants the findings to be “representative” of the population, then a non-probability sample can
be chosen.

Notes Probability and Non-probability Samples
Probability Sample
Simple random sample Every member of the population has an equal chance
of being selected.
Stratified random sample The population is divided into mutually exclusive
groups (such as gender, age group), and random
samples are drawn from each group.
Cluster or area sample The population is divided into mutually exclusive
groups (such as city, village) and the researcher draws
a sample of the groups to be interviewed.
Non-probability Sample
Convenience sample The researcher selects the most accessible population
members to interview and obtain information (such
as shoppers in a departmental store).
Judgement sample The researcher uses her/his judgement to choose
population members who are good prospects for
accurate information (such as doctors).
Quota sample The researcher finds and interviews a predetermined
number of respondents in each of several categories
(such as 50 males and 50 females).

24 LOVELY PROFESSIONAL UNIVERSITY