# Type of Sampling

When to use it

Probability Strategies

## Simple Random Sampling

When the population members are similar to one another on important variables

Ensures a high degree of representativeness

Time consuming and tedious

Systematic Sampling

When the population members are similar to one another on important variables

Ensures a high degree of representativeness, and no need to use a table of random numbers

Less random than simple random sampling

Stratified Random Sampling

When the population is heterogeneous and contains several different groups, some of which are related to the topic of the study

Ensures a high degree of representativeness of all the strata or layers in the population

Time consuming and tedious

Cluster Sampling

When the population consists of units rather than individuals

Easy and convenient

Possibly, members of units are different from one another, decreasing the techniques effectiveness

Non-Probability Sampling

Convenience Sampling

When the members of the population are convenient to sample

Convenience and inexpensive

Degree of generalizability is questionable

Quota Sampling

When strata are present and stratified sampling is not possible

Insures some degree of representativeness of all the strata in the population

Degree of generalizability is questionable

Notes:

1.                   Reducing sampling error is the major goal of any selection technique.

2.                   A sample should be big enough to answer the research question, but not so big that the process of sampling becomes uneconomical.

3.                   Estimating sample size – in general, you need a larger sample to accurately represent the population when:

a.       The amount of variability within groups is greater, and

b.       The difference between the tw groups gets smaller.

4.                   In general, the larger the sample, the smaller the sampling error and the better job you can do.

5.                   If you are going to use several subgroups in your work (such as males and females who are both 10 years of age, and healthy and unhealthy urban residents), be sure your initial selection of subjects is large enough to account for the eventual breaking down of subject groups.

6.                   If you are mailing out surveys or questionnaire, count on increasing your sample size by 40% to 50% to account for lost mail and uncooperative subjects.

7.                   Remember that big is good, but appropriate is better.  Do not waste your hard-earned money or valuable time generating samples that are larger than you need… law of diminishing returns will set in!