TYPES OF SAMPLING PROCEDURES

Probability sampling

Non-probability sampling

PROBABILITY BASED SAMPLES

These are procedures in which every member of the sampling frame has an equal chance of being
selected. The chance of every unit in the population to be selected is known. ‘

Selection of specific units for the sample depends entirely on chance.

(a) RANDOM SAMPLING

It is a probability based sample selection procedure.

All individuals in the defined population have an equal chance of being selected.

It is the best single way of obtaining a representative sample.

The sample can be selected using;

Computer Selection

Random Number Tables

Lottery Methods

RANDOM NUMBER TABLES

The numbers move vertically or horizontally.

A random sequence of numbers is produced by selecting from the random lists those participants
who fall in the ranges required eg. when selecting 5 people from 50. Give each person a number 1-
50.

Select those who hold the first 5 numbers which occur as you move through the table.

THE LOTTERY METHOD

When selecting a group of 30 from college students. Give every student a number, draw thirty
numbers from a well shuffled barrel.

LIMITATIONS OF RANDOM SAMPLING

1. It is expensive; the cost of pulling together a large sampling frame is prohibitive.

2. Respondents selected must be interviewed to ensure the integrity of the process. This means up
to three call backs to individuals may be made before classifying a non-response.
STRATIFIED SAMPLING

It is a process of selecting a sample in such a way that identified subgroups in a population are
represented in a sample in the same proportion that they exist in the population.

Steps are similar to those in random sampling except that one is selecting from sub-groups in a
population rather than from the whole population.

SYSTEMATIC SAMPLING

It is done by taking every nth person in a list eg. selecting 20 students from 200.

Make an alphabetic register then select every 10th name. Random select a starting number eg 4.

CLUSTER SAMPLING

This is sampling in groups not individuals.

Groups are randomly selected.

Any intact group of similar characteristics is a cluster.

Steps in cluster sampling are similar to those in random sampling except that the random selection
of groups is involved not individuals.

MULTISTAGE SAMPLING

It is a modification to cluster sampling.

The population would be found in the first stage or primary units. Each of the primary units may be
made up of second stage units and so forth.

A sample is taken of the first stage units and then of the second stage units eg. for national surveys,
the administrative districts for voting are usually rural and urban districts (PSU) then parliamentary
constituencies (SSU) and lastly polling stations (TSU).

NON- PROBABILITY SAMPLING TECHNIQUES

These are sampling techniques in which there is no way to estimate the probability of each element
to be included in the sample. There is no guarantee that each element has some equal chance of
being included.

Non-probability sampling techniques are used when it is not possible to obtain an absolutely
representative sample.
ADVANTAGES

1. Lower cost

2. Faster

3. Smaller sample sizes.

4. Important respondents can be targeted.

DISADVANTAGES

1. Results are purely indicative

2. Sampling error cannot be computed.

3. The degree of representativeness of the sample to the population is not known.

4. Assumptions need to be made about the groupings with the population of interest.

QUOTA SAMPLING

It is a non-probability sampling technique in which a sample is selected based on known proportions

in the populations.

In quota sampling, the population is divided into groups according to selected characteristics eg.
gender, age, eco nomic status or religious groupings etc. One may use census data to find out the
proportion of the adult population who fall under various age or sex classes.

The investigator makes a final selection from the groupings eg. 5% aged, 40% adults and 55% young.

The quota in this sense is a percentage of respondents selected from each given category.

It is one type of purposive sampling in which the investigator selects respondents within the quota
according to their judgement.

The quota sampling technique is very suitable for attitude and opinion surveys.

ADVANTAGES

It has a high speed of data collection.

It is cheap to undertake.

I t allows sampling to take place where a sampling frame may not be available but key characteristics
of the population are known’

Investigators may not need to interview named individuals; they are screened in or out via a small
number of classification questions

The data, when compared to random methods which are perhaps double the cost, has been proved
to be acceptable provided that the research is properly managed.
Cost savings may be used to improve the quality of research through increasing sample sizes or using
a different method in support of the survey.

It is a popular technique thus its popularity shows that it works.

DISADVANTAGES

Whilst known characteristics may be distributed in correct proportions, unknown characteristics that
may be relevant to the survey may not be. Hidden bias may exist that is not discovered.

Researchers may be biased as to the types of respondents they choose to interview or the location
where they choose to carry out the interview. A quota of the young people may be filled at one
youth club but will not be truly representative of the population as a whole.

JUDGEMENTAL OR PURPOSIVE SAMPLING

It is a nonprobability sampling technique in which the interviewer handpicks the respondents that
are to be included in the sample on the basis of expert judgement. This means the investigator
exercises his/ her own discretion in selecting sample observation from the population.

The key assumption underlying this type of sampling is that, with sound judgement or expertise and
an appropriate strategy, one can carefully and consciously select the sample.

The investigator in this case should know the types of people who are typical of the population in
question.

The individuals selected are those judged to have special characteristics or those who are likely to
provide the most useful information for the study.

MERITS/ADVANTAGES

It is low cost sampling technique.

It is convenient to use.

I t saves time.

It is almost as good as probability sampling.

It can give adequate representation of all the characters of the population if it is used on a small
population.

DEMERITS/DISADVANTAGES

It allows for bias or personal prejudice.

The results of the sample cannot be extended /generalised to the population because various
samples drawn from the same population using this technique may give different results due to
absence of scientific application of statistical techniques.

The sample estimates vary widely from the population parameters leaving room for large sampling
errors.

In the absence of scientific application of statistical techniques, it becomes difficult to establish the
significance of the statistical error.

CONVENIENCE/ OPPORTUNITY SAMPLING

It is a non-probability sampling technique where sampling units are selected because they are
conveniently available, accessible and easy to measure eg a class of students, a congregation (people
at a church), people visiting a supermarket and can be stopped for questioning.

ADVANTAGES OF CONVENIENCE SAMPLING

It is cheap to conduct.

It saves time since it studies those who are available.

It does not require a sampling frame so it is easy to conduct.

DISADVANTAGES

The sample may not be representative of the population so it is not possible to generalise results.

It is not possible to identify the sampling error.

SNOWBALL SAMPLING

This is a non-probability sampling technique in which initial respondents are selected at random and
then subsequent respondents are selected by referrals, ie. key individuals are found from initial
interviews. One person’s interview can produce a snowball effect as each new interviewee leads in
turn to several more interviewees. Those who are selected can either report on the feelings of the
community or especially when trying to understand a particular phenomenon under research.

This is very useful where there is a low incidence of the population eg where unusual behaviour is
under consideration.

SELF SELECTING/ VOLUNTARY SAMPLING

This is a technique where volunteers offer to be participants therefore volunteers are self-selecting
eg those who answer advertisements requiring participants.

Field studies may recruit participants in this manner. This is also suitable for studies which really are
sensitive’
ADVANTAGES

Volunteers are likely to cooperate, be open and provide accurate information.

Volunteers are not likely to withdraw before the end of the study.

DISADVANTAGES

Volunteers may not make up a representative sample.

It is not always easy to come up with volunteers’

It is not possible to come up with the sampling error so it is not safe to generalise the results where
such samples have been used.

5 Common Errors in the Research Process

1. Population Specification

This type of error occurs when the researcher selects an inappropriate population or universe from
which to obtain data.

2. Sampling

Sampling error occurs when a probability sampling method is used to select a sample, but the
resulting sample is not representative of the population concern. Unfortunately, some element of
sampling error is unavoidable. This is accounted for in confidence intervals, assuming a probability
sampling method is used.

3. Selection

Selection error is the sampling error for a sample selected by a nonprobability method.

4. Non-responsive

Nonresponse error can exist when an obtained sample differs from the original selected sample.

5. Measurement

Measurement error is generated by the measurement process itself, and represents the difference
between the information generated and the information wanted by the researcher.