You are on page 1of 9

Types of Sampling Techniques

RANDOM SAMPLING

A. Cluster Sampling
1. The population is divided into mutually exclusive groups and the researcher
draws a sample of the groups interviewed.
2. A random sample for clusters is taken and then all units within the cluster are
chosen.
3. Makes sampling quicker and easier as respondents are in one area
4. Respondents are selected to represent the target population as a whole
5. Involves increased risk of sampling error
6. Used where no sampling frame exists
7. Random selection from a chosen area

Cluster Sampling

Cluster sampling was devised in the United States of America to try to overcome the problems
of cost and the lack of a satisfactory sampling frame. Instead of selecting a random sample
scattered over a wide area, it is a few geographical areas that are selected at random and
every single household in each area is interviewed. The areas chosen must be relatively small.
It is not possible to interview every household in cities like London and Singapore. The typical
area selected might be three or four streets in a town, or perhaps an apartment block. The
great advantage of this type of sampling is the saving in time and cost. Many interviews can
take place within a short space of time with a minimum of travelling. Moreover, it does not
require any knowledge of the population before the survey is undertaken; that is, no sampling
frame is necessary.

On the other hand, there is a basic problem with this type of sample. Whilst the population is
heterogeneous, those who live in a small area or the same block of flats will tend to be
homogeneous. They will tend to have the same opinions, the same characteristics or the same
life-style. Naturally one cannot be certain of this, but it seems to be highly probable, and since
the statistician is selecting only a few such areas there is a great danger that the sample will
be biased. One way of trying to offset this tendency to bias is to increase the number of clusters
in the hope that you will then include in your sample every important strata, but every increase
in the number of clusters raises the cost of the survey. There comes a time when you will have
to weigh the risk of selecting a biased sample against the cost of avoiding the danger of bias.
It is, however, a useful type of sample when you have no sampling frame, or when the cost of
taking a random sample is too high.

1
B. Simple Random Sampling
1. Every member of the population has a known and equal chance of being selected.
2. It is used where the target group is large and where a sampling frame is available
3. Accurate if sample is large enough
4. It is possible to calculate the sampling error
5. Requires an homogenous population and is not appropriate if the population
consists of distinct strata

Simple Random Sampling

A simple random sample is often mentioned, but it is actually one of the least used techniques.
In theory, it is easy to understand. However, in practice it is difficult to perform. Technically, a
simple random sample is a set of n objects in a population of N objects where all possible
samples are equally likely to happen. Here’s a basic example of how to get a simple random
sample: put 100 numbered balls into a bowl (this is the population N). Select 10 balls from the
bowl without looking (this is your sample n). Note that it is important not to look as you could
(unknowingly) bias the sample. While the “bowl” method can work fine for smaller populations,
in reality you will be dealing with much larger populations.

Imagine the 12 persons are game pieces. Place the 12 game pieces in a bowl and (without
looking) choose 3. This is simple random sampling.

A simple random sample is chosen in such a way that every set of individuals has an equal
chance to be in the selected sample. It sounds easy, but itis often difficult to employ in surveys
or experiments. In addition, it is very easy for bias to creep into samples obtained with simple
random sampling. Sometimes it is impossible (either financially or time-wise) to get a
realistic sampling frame (the population from which the sample is to be chosen). For example, if
you wanted to study all the adults in the U.S. who had high cholesterol, the list would be
practically impossible to get unless you surveyed every person in the country. Therefore, other
sampling methods would probably be better suited to that particular experiment.

The simplest example of Simple Random Sample would be working with things like dice or cards
— rolling the die or dealing cards from a deck can give you a simple random sample. But in real
life you’re usually dealing with people, not cards, and that can be a challenge .

2
C. Stratified Random Sampling
1. This separates the population into strata.
2. Each strata contributes a pre-determined proportion to the overall sample.
3. Within these proportions respondents are chosen randomly.
4. It is used when the population is defined by categories and a sample frame.
5. It results in a sample which is free from selective bias as it takes account of
significant attributes of population.

Stratified Sampling

When the response we get to our questions is likely to depend in this way on the social group
to which the person being questioned belongs, we will get far better results if we adopt
stratified sampling. If we are investigating the impact of traffic on the 'roads, it is quite possible
that systematic sampling would result in a sample which contained only those who owned
cars. The opinions of those who do not own cars would not be represented, and to that extent
the sample results will be biased. Stratified sampling has been designed to ensure that all
important views are represented in the sample. In this type of sample each social group is
represented in the sample in proportion to the size of that group in the population as a whole.
Let us take a simple example to show how stratified samples are constructed. Suppose we are
investigating opinions on education of those who are still at school or college. We may decide
that these opinions will depend on whether the student is at university, secondary school, or
private school. We know that 20% of students are at university, 10% at private school and 70%
at a secondary school. If we are selecting a sample of 2000 students, our first step is to say that
20% of the sample, or 400 students, must be university students, 200 (10%) from private
schools, and 1400 (70%) from secondary schools. We may now additionally decide that men
and women have markedly different opinions, so we will now divide our three groups (or
strata) into two sub-groups. If, for example, we know that 15% of university students are
women, we would include 60 women among our sample of 400 university students. Each of
these sub-groups can be further subdivided, perhaps in relation to age or to parents’ social
class. It does not matter how far groups are subdivided provided that we are stratifying
according to a characteristic which is relevant to the survey. If, for example, we are investigating
television viewing habits there is little point in stratifying the sample according to political
affiliations. It is probably completely irrelevant. On the other hand, our viewing habits might
well be affected by the size of our family, or whether we live in a rural or urban area and we
would stratify according to these criteria.

3
Once we have decided on the number of people in each stratum of our sample, the persons to
be included must, of course, now be chosen by some random method.

Now, one peculiarity can arise with this type of sampling. If our strata are many, it may be that
an important sub-group of the population is entitled to be represented by only one or two
members. It may be that we feel that we cannot obtain an adequate representation of the
opinions of this group if we question one person only, and, if so, we should include more
representatives of the group, say four or five. While this may seem reasonable, it means, of
course, that members of this group have a far higher probability of being chosen as a part of
the sample than have people in other strata. This type of sampling is known as sampling with a
variable sampling fraction, and in assessing our final results we must make allowance for the
difference in the representation of such strata. In many surveys, when samples are drawn from
a markedly heterogeneous population the sampling fraction is variable rather than uniform.
The principle finds its most important application in the evaluation of stock value by sampling
methods. We can afford to take a very small sample of items costing only a few pence, since, if
our sample is inaccurate it will make little difference to the total value of stock. But if the items
of stock cost a great deal, even a small error may significantly affect our total stock valuation.
Thus, in determining total stock value from a sample valuation of stock, accountants almost
invariably ensure that high-value items have a greater probability of being selected than do
low-value items.

Once again, with stratified sampling, we need a sampling frame which will give us a great deal
of information about the social structure. Fortunately, nowadays governments undertake so
many social surveys and publish so many statistics that there is little difficulty in obtaining the
necessary information. Hence stratified sampling is becoming increasingly popular whenever
the objective is to assess people's attitudes or opinions.

4
D. Systematic Sampling
1. This technique selects every nth person on a list of people in the target population.
2. n = population ./. sample size
3. It is costly and time consuming
4. Easier to extract the sample than that of simple random sampling and it ensures
that the sample is spread across the country
5. Not appropriate if the population consists of different strata
6. Only provides a representative sample of the population if it is arranged in a
random way.

Systematic Sampling

In practice true random sampling is not possible unless there is a good sampling frame and
the population is fairly small. So, for most practical work, investigators resort to methods of
selection which are quasi-random, or not truly random. One of the most popular is to choose
as a member of the sample every nth item on a list, the first sample unit being selected by
some random method. If, for example, we wanted a sample consisting of 2 1/2% of the
population we would firstly select at random the first unit, say it is the 29th invoice in a file,
and then every 40th invoice after that — the 69th, 109th and so on. Can you see why this type
of sampling is not truly random? While the first item is chosen by random methods, every
other item is then preselected. Nevertheless, this method of selection (called systematic
sampling) approximates to random sampling sufficiently well to justify its widespread use. Once
again, we need a reasonably good sampling frame, but, in our opinion, having accepted that
the sample is not truly random, it need not be 100% accurate. However, if we are to use
systematic sampling, we must ensure that there are no ‘cyclical’ patterns in the sampling
frame that match our choice of items. If, for example, we are taking a survey on a housi ng
estate to assess the opinion of the residents about the open plan layout, and every tenth
house is a corner house, then a 10% systematic sample either always, or never, includes a
corner house. Such a pattern as this could substantially affect the results we get.
Now, systematic sampling is useful so long as the population is homogeneous. It is suitable,
for example, if accountants are selecting a sample of invoices to check for errors; or if the
police are taking a sample of motorists passing a certain point to check that they have a valid
driving licence. But you will appreciate that in much statistical work the population is
heterogeneous. This is especially true if we are investigating opinions, because in many cases
a person's opinions are formed by, or result from, the social class to which he belongs. If we
are asking the simple question, 'Do you think that people earning over £10,000 a year pay
too much in taxes' the answer we get will often depend on the income of the person being
questioned. Inevitably the rich will answer `yes', while the poor are more likely to say `no' .
5
NON-RANDOM SAMPLING

Non Probability Sampling


A. Judgment or Purposive Sampling
1. Sampling techniques which involve some human judgment in selecting the
sample.
2. Every member does not have an equal and known chance of being selected.
3. Does not require a sampling frame.
4. Less costly but more effective.
5. Not possible to calculate the sampling error.
6. Only as good as the judgment that went into it.

B. Quota Sampling
1. The population is divided by important variables such as age, income.
2. A quota (limit) is decided for each group.
3. People in each group are interviewed up to the quota determined.
4. Quick and easy but not random.
5. Danger of interviewer bias.
6. The choice of respondents is left to the market researcher.

Quota Sampling

There can be few people who have not been stopped in the street by an interviewer holding a
questionnaire, or who at least know someone who has been so stopped. And the immediate
reaction seems to be, `Why did they choose me?'

The essence of this type of sample is that it is not preselected but, is chosen by the interviewer
on the spot. She is given a certain number (a quota) of questionnaires which it is her job to
have completed during the course of the week, or month. She has a free choice of whom she
asks to answer the questions. This choice, naturally, is subject to some general restrictions. One
can imagine a young male interviewer stopping only girls of about his own age whom he
thought to be pretty. Free choice does not extend as far as this. The interviewer is told that the
completed quota of questionnaires must include a certain number of males and of females; a
certain number from specified age groups such as, under 18, 18 to 40 and 40 to 65. Controls
such as this are easy to implement, but often the quota has to be further subdivided by social
class or occupation. This can make quota sampling very difficult as there is no real definition of
what constitutes, say, the upper middle class, and certainly you can seldom tell a person's social
class by looking at him. Such a requirement necessitates the interviewer being given detailed
definitions and descriptions of what the survey body means by each term it uses. Subject,
however, to controls such as this, the sample is chosen by the interviewer from people who pass
in the street.

6
Now, obviously, such a technique of sampling can be open to a great deal of abuse and bias.
When it was first introduced, many cases were reported of interviewers sitting at home filling
in their own questionnaires without ever interviewing anyone. Of course, this was soon
remedied by the questionnaire asking for names and addresses, and the survey body making a
spot check to ensure that such people had actually been interviewed. Deliberate fraud of this
nature is very rare now. Much more important is the bias to which the quota survey may be
subject. If the interviewer does not start work until, say, ten o'clock, the majority of people she
can stop must of necessity be the housewife or the unemployed; if she starts at, say, nine
o'clock, the vast majority will be clerical or managerial workers; to start even earlier means that
she will probably meet largely manual workers. Thus, the interviewer must be prepared to
spread her work throughout the day. It is little use trying to have all the questionnaires
completed before the morning coffee-break. A further source of bias is that the people you
meet in the street are either going somewhere or doing something; they resent being
interrupted and their answers may be hurried and slapdash, given without serious thought.
These problems necessitate further controls and advice being given to the interviewer. She may
be told at what times to conduct the interviews or even where to conduct them. We must point
out that the success of such a survey depends very much on their following their instructions to
the letter.

Since the quota sample has these weaknesses, why is it becoming increasingly common? Well,
for one thing, it is cheap. It eliminates repeated calls to interview a person who may not be in at
home on the first two or three occasions. In fact, it is estimated that each interview in a quota
sample costs only about half as much as each interview in a random sample. Then, too, we have
in every survey a number of people who do not respond. In a quota sample it does not matter.
They are ignored and we pass straight on to the next person with no loss of time and no cost. In
other types of sample, a substitute has to be found for each person who does not respond. It
might, in fact, be argued that this itself introduces bias into the sample. It may be that the man
who is prepared to fill in questionnaire is some kind of extrovert, and we may be getting the
views of only one type of individual.

Nevertheless, given the existence of the controls and checks we have mentioned, all the
evidence points to the fact that in skilled hands quota sampling gives reasonably satisfactory
results. Statisticians generally accept that it is not a substitute for random sampling statistically,
but it is so quick and so cheap a method of carrying out surveys that it is not likely to be replaced
for a very long time.

7
C. Convenience Sampling
1. Sample those members of the population who can be easily contacted and those
who are willing to volunteer.
2. Quick, easy, and uncomplicated but not possible to extrapolate from the sample
and drew inferences about the population.
3. Prone to bias because of chose of location.

Convenience sampling

Convenience sampling is defined as a method adopted by researchers where they collect


market research data from a conveniently available pool of respondents. It is the most
commonly used sampling technique as it’s incredibly prompt, uncomplicated, and economical.
In many cases, members are readily approachable to be a part of the sample.

Researchers use convenience sampling in situations where additional inputs are not necessary
for the principal research. There are no criteria required to be a part of this sample. Thus, it
becomes incredibly simplified to include elements in this sample. All components of the
population are eligible and dependent on the researcher’s proximity to get involved in the
sample.

The researcher chooses members merely based on proximity and doesn’t consider whether
they represent the entire population or not. Using this technique, they can observe habits,
opinions, and viewpoints in the easiest possible manner.

A good example of convenience sampling is: A new NGO wants to establish itself in 20 cities. It
selects the top 20 cities to serve based on the proximity to where they’re based.

Convenience sampling is applied by brands and organizations to measure their perception of


their image in the market. Data is collected from potential customers to understand specific
issues or manage opinions of a newly launched product. In some cases, it is the only available
option. For example, a university student working on a project and wants to understand the
average consumption of soda on campus on a Friday night will most possibly call his/her
classmates and friends and ask how many cans of soda they consume. Or may go to a party
nearby and conduct an easy survey. There is always a chance that the randomly selected
population may not accurately represent the population of interest, thus increasing the chances
of bias.

A basic example of a convenience sampling method is when companies distribute their


promotional pamphlets and ask questions at a mall or on a crowded street with randomly
selected participants.

8
D. Snowballing Sampling

Snowball sampling is where research participants recruit other participants for a test or study. It
is used where potential participants are hard to find. It is called snowball sampling because (in
theory) once you have the ball rolling, it picks up more “snow” along the way and becomes
larger and larger. The researchers used their own judgment to choose participants. Snowball
sampling consists of two steps:
1. Identify potential subjects in the population. Often, only one or two subjects can be found
initially.
2. Ask those subjects to recruit other people (and then ask those people to recruit.
Participants should be made aware that they do not have to provide any other names.
These steps are repeated until the needed sample size is found. Ethically, the study participants
should not be asked to identify other potential participants. Rather, they should be asked to
encourage others to come forward. When individuals are named, it is sometimes called “cold-
calling”, as you are calling out of the blue. Cold-calling is usually reserved for snowball sampling
where there is no risk of potential embarrassment or other ethical dilemmas. For example, it
would be easier to cold-call participants in a study for families who regularly dine at fast-food
restaurants than it would be to cold-call people who are having extra-marital affairs. Snowball
sampling can be a tricky ethical path to navigate.

Some people may not want to be found. For example, if a study was investigating cheating on
exams, shoplifting, drug use, prostitution, or any other “unacceptable” societal behavior,
potential participants would be wary of coming forward because of possible ramifications.
However, other study participants would likely know other people in the same situation as
themselves and could inform others about the benefits of the study and reassure them of
confidentiality.

Snowballing allows for studies to take place where otherwise it might be impossible to conduct
because of a lack of participants. Snowball sampling may help you discover characteristics
about a population that you were not aware existed. For example, the casual illegal downloader
vs. the for-profit downloader.

It is usually impossible to determine the sampling error or make inferences about populations
based on the obtained sample. Snowball sampling is also known as cold-calling, chain sampling,
chain-referral sampling, and referral sampling.

Benefits of Sampling

1. Saves time
2. Faster results
3. Saves money
4. Can yield statistically valid info about the population as a whole

You might also like