Page 119 - DMGT509_RURAL MARKETING

Rural Marketing

Notes Organizing a Market Survey for the entire universe can, and usually is, a time consuming,
expensive process. It may take so much time meeting and asking questions from a million
people, that by the time the results appear the product itself may become obsolete. Therefore,
the Survey is done on the basis of a sample of the universe, which is made using well-established
techniques.
Sample Size

The larger number of the people to be interviewed, the better or more reliable the result.
However, in practice, it is better to cover a much smaller sample, may be one percent of the
universe to get reasonably good results, in lesser time and money. It really depends on how the
sample gets selected.

Sample Selection

There can be two ways of selecting a sample.
1. Probability Sample: It ensures that each of the members of the universe have equal
chance or probability of being selected for the survey. It helps in understanding the
levels of errors on account of the sample selected. There are three ways of probability
sampling.
(a) Simple random sample: Every member has an equal opportunity of being in the
survey.
(b) Stratified random survey: The population/universe is divided into mutually
exclusive groups such as, age, income and then random samples are taken from each
group.
(c) Cluster sample: The population is divided as per geographic divisions, like for example
Delhi could be divided into, north, south, east west and central zones and sample
taken from each zone on random basis.
2. Non-probability sample: As the cost and time involved in probability sample surveys is
quite high, MR people use non-probability sample. This can be done in three ways.
(a) Convenience sample: The research person selects the persons most easily met for the
MR.
(b) Judgment sample: Researcher uses his judgment about the sample, which would provide
the most accurate answers.

(c) Quota sample: The researcher meets with a set of persons in each category.
As an example, in sampling methods, if one person is not available then no one else takes his
place except in quota sample method where the researcher has to find out another person in the
same category and meet him.
At times even with proper sampling, errors creep in to the collected data. The reasons of such
errors are given below.
1. A portion of population may have been omitted.
2. People not included in the sample may differ widely from those interviewed.
3. Substitute interviewees may not be right ones if you are not able to meet the person in the
sample.
4. The respondents find the questionnaire difficult.

114 LOVELY PROFESSIONAL UNIVERSITY