Question 1

Difference between a stratum and a cluster

A stratum and a cluster are units used by different probability sampling methods namely stratified sampling and cluster sampling respectively. The probability sampling methods apply the aspect of randomization to pick a sample implying each unit has an equal chance of being enumerated.

Stratified sampling in research requires that the population be divided into non-overlapping subgroups. A stratum refers to a sub-group in stratified sampling consisting of individual elements that can be examined separately. The cluster, on the other hand, refers to a collection of several individual elements put together to form a research unit known as a cluster. In conducting research the cluster will be viewed as one unit rather than analyzing the individual elements separately.

Question 2

Telephone sampling is a method used by the researcher to acquire a sample to be used in a telephone survey in order to collect data. A random sample is one in which the elements of a population have an equal chance of being picked to be part of the sample. Telephone surveys tend to cost less and save on time. However, obtaining random samples may be prone to biases including;

• Exclusion of households without telephones this means that not all persons have an equal chance of participating in the survey.
• Some households have telephones but are not listed in the directory implying that this group may also be overlooked in the selection of the sample.
• Some of the listed households have more than one telephone number which may increase their chances of being selected.

Question 3

A census is a complete enumeration of the entire population to collect data and make conclusions about the population. Sampling on the other hand is applied by enumerating part of the population and drawing conclusions about the entire population based on the sample findings. A census requires the enumerator to ask all the questions in a questionnaire this can be costly and time-consuming.

Sampling could be used for purposes of quality control to ensure some of the questions are asked to some participants while other general questions may be asked to all respondents especially because some respondents may be illiterate and may provide misleading information. This approach can save time. Sampling can also be used to cut on the cost compared to the traditional method of the census by selecting a large enough sample from the population and using statistical sampling techniques to estimate the number of people in the population.