What is a Survey?

Surveys are a method of gathering information from individuals. Surveys have a variety of
purposes, and can be conducted in many ways. Surveys may be conducted to gather information
through a printed questionnaire, over the telephone, by mail, in person, by diskette, or on the
web. This information is collected through use of standardized procedures so that every
participant is asked the same questions in the same way. It involves asking people for
information in some structured format. Depending on what is being analyzed, the participants
being surveyed may be representing themselves, their employer, or some organization to which
they belong.
Features of a Survey

Information is collected from a sample of population

By asking questions using a standardized questionnaire

Produce statistics (quantitative or numerical description about some aspects of the study
population)

Generalizable to the whole population

Types of Survey
Cross sectional
A cross-sectional survey collects data to make inferences about a population of interest
(universe) at one point in time. Cross-sectional surveys have been described as snapshots of the
populations about which they gather data. Cross-sectional surveys may be repeated periodically;
however, in a repeated cross-sectional survey, respondents to the survey at one point in time are
not intentionally sampled again, although a respondent to one administration of the survey could
be randomly selected for a subsequent one. Cross-sectional surveys can thus be contrasted with
panel surveys, for which the individual respondents are followed over time. Panel surveys

usually are conducted to measure change in the population being studied. Cross-sectional
surveys can be conducted using any mode of data collection, including telephone interviews in
which landline telephones are called, telephone interviews in which cell phones are called, faceto-face interviews, mailed questionnaires, other self-administered questionnaires, electronic mail,
Web data collection,
Panel/ Cohort
The panel survey design is that it provides repeated observations on a set of variables for the
same sample units over time. The different types of panel surveys are distinguished by the
different ways of deriving these repeated observations. In a panel survey, repeated observations
are derived by following a sample of persons over time and by collecting data from a sequence
of interviews .. There are many variations under this general description of a panel survey,
including (a) cohort panel surveys, (b) household panel surveys, and (c) rotating panel surveys.
These three types of panel surveys can be distinguished, first, by the sampling units and the
population the survey aims to represent. The focus can be entirely on individuals .
A panel study is defined as a study that collects information on the same individuals at different
points in time. The various data collections are often called waves. A panel study is therefore a
longitudinal study; it differs from other studies that collect information over time, such as time
series

and

cohort

studies,

in

that

it

studies

the

same

persons

longitudinally.

Descriptive Survey
Descriptive research is used to describe characteristics of a population or phenomenon certain
behavior, life style, disease prevalence etc.. The characteristics used to describe the situation or
population are usually some kind of categorical scheme also known as descriptive categories.
Analytical Survey:
Analytical surveys attempt to describe and explain why certain situations exist. In this approach
two or more variables are usually examined to test research hypotheses. The results allow
researchers to examine the interrelationships among variables and to draw explanatory
inferences.

 Steps of survey
1.Establish the goal of the project - What do you want to know
2. Select the sample - Whom will you interview
3. Choose interview methodology - How will you interview
4. Specify variables of interest What information do you want
5. Create questionnaire (i.e. instrument) - What will you ask
6. Pre-test questionnaire Are your questions clear, understandable and follow logical patterns
7. Conduct interviews - Ask the questions
8. Data editing and entry Check for consistency
9. Data analysis Answer research questions

Self-administered Questionnaire
The self-administered questionnaire is a type of questionnaire, either in paper or electronic form,
that a respondent completes on his/her own.

Respondents answer at their convenience.

There is no need to set up interview appointments.

Surveys are delivered wherever the mail or email goes: city or country, home or office.

No interviewer is present to inject bias in the way questions are asked.

The low cost-per-completion makes it an economical method of surveying large samples.

Disadvantage

Response rates can be low

Possible clarity issues
Possible language and literacy issues
If online, possible access issues
Face-to-face interviews

A personal interview survey, also called as a face-to-face survey, is a survey method that is
utilized when a specific target population is involved. The purpose of conducting a personal
interview survey is to explore the responses of the people to gather more and deeper information.
1. High Response Rates
2. Tolerable Longer Interviews
3. Better Observation of Behavior
Disadvantages of Personal Interview Survey
1. High Costs
2. Time-consuming
Specify variables of interest
What information do we want
Variables

Questions

Age

What is your age?

Profession

What is your job?

Marital status

What is your marital status?

Strength and weakness of survey

Strengths of tests (especially standardized tests)

1.
2.
3.
4.
5.

High Representativeness
Standard questionnaire
Availability of reference group data.
Tests are usually already developed.
A wide range of tests is available (most content can be tapped).

Weaknesses of tests (especially standardized tests)

1. Bad design of questionnaire
2. Poor response population
Types of Survey Errors
Coverage errors
Coverage errors occur when the sampling frame excludes some segments of the target
population. Phone books are very convenient, but unlisted households are excluded. Election
polls sample from the frame of registered voters, but the target population is the subset who are
going to vote.
Non-response error
Non-response errors can cause serious bias in survey results. Phone interviewers find it easiest
to reach families with young children, and hardest to reach young singles. Mail surveys tend to
be answered by those who feel strongly about an issue, or by those who feel more civic
responsibility, neither of which is a representative cross-section of the population.
Measurement Errors
Measurement Errors occur when respondents answer `inaccurately' because of question
wording, question ordering, interviewer effect, or other external influences. For example, the
answer to ``Do you approve of affirmative action?'' may be influenced by the gender of the
interviewer. Also, the question ``Do you like living in Davis Hall?'' may be influenced by
preceding it with the question ``Do you have enough parking spots in Davis Hall?''.
Why sampling is important
Sampling is required of any piece of research because of the significant impact that it can have
on the quality of your results/findings. If there are new to sampling, there are a number of key
terms and basic principles that act as a foundation to the subject. This article explains these key
terms and basic principles. Rather than a comprehensive look at sampling, the article presents
the sampling basics . It also provides links to other articles within the Sampling Strategy section
of this website that you may find useful. Some of the key sampling terms you will come across
include population, units, sample size, sampling frame, sampling technique, and sampling bias.

Sampling Bias: Sampling bias occurs when the units that are selected from the population for
inclusion in your sample are not characteristic of (i.e., do not reflect) the population. This can
lead to your sample being unrepresentative of the population you are interested in.
Sampling
Sampling is the process of selecting units from a population of interest so that by studying the
sample fairly generalize research results back to the population from which they were chosen.
The Sampling Process

Execute the Sampling Process

What percentage of Women (20-55) in Dhaka Metropolitan city in Bangladesh were

diagnosed with uterine cancer in 2000?

Study Population?

Sampling Frame?

Study Population?

Women (20-55) in Metropolitan Dhaka in 2000 who do not have a history

of hysterectomy.

Sampling Frame?

List of all women (20-55) in Metropolitan Dhaka in 2000 who do not have a history of
hysterectomy.

 Classification of Sampling Techniques

Simple Random
Sampling

Non-Probability Sample
Non-probability sampling represents a group of sampling techniques that help researchers to
select units from a population that they are interested in studying. Collectively, these units form
the sample that the researcher studies.
A core characteristic of non-probability sampling techniques is that samples are selected based
on the subjective judgment of the researcher, rather than random selection.
Convenience Sampling

Imagine that a researcher wants to understand more about the career goals of students at the
University of Bath. Let?s say that the university has roughly 10,000 students. These 10,000
students are our population (N). Each of the 10,000 students is known as a unit, a case or
an object (these terms are sometimes used interchangeably; we use the word unit). In order to
select a sample (n) of students from this population of 10,000 students, we could choose to use a
convenience sample. Let?s imagine that because we have a small budget and limited time, we
choose a sample size of 100 students.
A convenience sample is simply one where the units that are selected for inclusion in the sample
are the easiest to access. This is in stark contrast to probability sampling techniques, where the
selection of units is made randomly. In our example of the 10,000 university students, we were

only interested in achieving a sample size of 100 students who would take part in our research.
As such, we would continue to invite students to take part in the research until our sample size
was reached. Since the aim of convenience sampling is easy access, we may simply choose to
stand at one of the main entrances to campus of the University of Bath where it would be easy to
invite the many students that pass by to take part in the research.
Purposive sampling

Purposive sampling represents a group of different non-probability sampling techniques. Also

known

as judgmental,

the judgment of

the

selective or subjective sampling,

researcher

when

it

comes

to

purposive
selecting

sampling

relies

the units (e.g.,

on

people,

cases/organizations, events, pieces of data) that are to be studied. Usually, the sample being
investigated is quite small, especially when compared with probability sampling techniques.
The main goal of purposive sampling is to focus on particular characteristics of a population that
are of interest, which will best enable you to answer your research questions.
Quota Sampling

Quota sampling: the researcher decides in advance on certain key characteristics which s/he will
use to stratify the sample
Imagine that a researcher wants to understand more about the career goals of students at the
University of Bath. Let's say that the university has roughly 10,000 students. These 10,000
students are our population (N). In order to understand more about this population, we only need
to examine a sample (n) of students, not the whole population of 10,000 students. Furthermore,
sometimes we are interested in particular strata (groups) within the population. Therefore, the
quota sample involves dividing the population into two or more strata (groups). These strata are
expressed as H.
For example, imagine we were interested in comparing the differences in career goals between
male and female students at the University of Bath. If this was the case, we would want to ensure
that the sample we selected had a proportional number of male and female students relative to
the population. Therefore, the total number of male and female students included in our quota
would only be equal if 5,000 students from the university were male and the other 5,000 students
were female. Since this is unlikely to be the case, the number of units that should be selected for

each stratum (that is, the number of male and female students selected) will vary. We explain
how this is achieved in the next section:

Snowball sampling
Snowball sampling is particularly appropriate when the population you are interested in
is hidden and/or hard-to-reach. These include populations such as drug addicts, homeless
people, individuals with AIDS/HIV, prostitutes, and so forth. To understand more about
snowball

sampling,

how

to

create

snowball

sample,

and

the

advantages

and

disadvantages of this non-probability sampling technique, see the article: Snowball

sampling.

Probability Sampling
Probability sampling is a sampling technique wherein the samples are gathered in a process that
gives all the individuals in the population equal chances of being selected.
In this sampling technique, the researcher must guarantee that every individual has an equal
opportunity for selection and this can be achieved if the researcher utilizes randomization.
Types of Probability Sampling
Simple Random Sampling
Simple random sampling is the easiest form of probability sampling. All the researcher needs to
do is assure that all the members of the population are included in the list and then randomly
select the desired number of subjects.
There are a lot of methods to do this. It can be as mechanical as picking strips of paper with
names written on it from a hat while the researcher is blindfolded or it can be as easy as using a
computer software to do the random selection for you.
Stratified Random Sampling
Stratified random sampling is also known as proportional random sampling. This is a probability
sampling technique wherein the subjects are initially grouped into different classifications such
as age, socioeconomic status or gender.

Then, the researcher randomly selects the final list of subjects from the different strata. It is
important to note that all the strata must have no overlaps.
Researchers usually use stratified random sampling if they want to study a particular subgroup
within the population. It is also preferred over the simple random sampling because it warrants
more precise statistical outcomes.
Systematic Random Sampling
Systematic random sampling can be likened to an arithmetic progression wherein the difference
between any two consecutive numbers is the same. Say for example you are in a clinic and you
have 100 patients.
1.

The first thing you do is pick an integer that is less than the total number of the
population; this will be your first subject e.g. (3).

2.

Select another integer which will be the number of individuals between subjects e.g. (5).

3.

You subjects will be patients 3, 8, 13, 18, 23, and so on.

There is no clear advantage when using this technique.

Cluster Random Sampling
Cluster random sampling is done when simple random sampling is almost impossible because of
the size of the population. Just imagine doing a simple random sampling when the population in
question is the entire population of Asia.
1.

In cluster sampling, the research first identifies boundaries, in case of our example; it can
be countries within Asia.

2.

The researcher randomly selects a number of identified areas. It is important that all areas
(countries) within the population be given equal chances of being selected.

3.

The researcher can either include all the individuals within the selected areas or he can
randomly select subjects from the identified areas.
Mixed/Multi-Stage Random Sampling

This probability sampling technique involves a combination of two or more sampling techniques
enumerated above. In most of the complex researches done in the field or in the lab, it is not
suited to use just a single type of probability sampling.

Most of the researches are done in different stages with each stage applying a different random
sampling technique.

 Different Between Probability sampling and non-probability sampling.

The two main methods used in survey research are probability sampling and non-probability sampling.
The big difference is that in probability sampling all persons have a chance of being selected, and results
are more likely to accurately reflect the entire population. While it would always be nice to have a
probability-based sample, other factors need to be considered (availability, cost, time, what you want to
say about results). Some additional characteristics of the two methods are listed below.
Probability Sampling
complete
sampling
frame:
Contact
information
for
the
entire
population.
Select a random sample from your population: Since all persons (or units) have an equal chance of
being selected for survey, randomly select participants without missing entire portions of audience.
Generalize results from a random sample: With this data collection method and a decent response rate,
extrapolate results to the entire population.
Can be more expensive and time-consuming than convenience or purposive sampling.
Non-probability Sampling
Used when there isnt an exhaustive population list available: Some units are unable to be selected,
therefore you have no way of knowing the size and effect of sampling error (missed persons, unequal
representation, etc.).
Not random: Can be effective when trying to generate ideas and getting feedback, but cannot generalize
results to an entire population with a high level of confidence. Quota samples (males and females, etc.)
Probability sampling
Always
method

for

quantitative

Non-probability sampling
research

Usually for qualitative research method

Sampling units have known probability

of being selected

No such thing

Involves statistical analysis

Does not involve statistics.

Results can be generalized

Results are
generalized

Sample size calculation is involved

Sample size calculation is not involved

not

intended

to

be

There are 5 different types of sample you should be able to define. You should also understand when
to use them, and what their advantages and disadvantages are.

Simple Random Sample

Obtaining a genuine random sample is difficult. We usually use Random Number Tables, and use the
following procedure;
1. Number the population from 0 to n
2. Pick a random place I the number table
3. Work in a random direction
4. Organize numbers into the required number of digits (e.g. if the size of the population is 80,
use 2 digits)
5. Reject any numbers not applicable (in our example, numbers between 80 and 99)
6. Continue until the required number of samples has been collected
7. [ If the sample is "without replacement", discard any repetitions of any number]

Advantages:

The sample will be free from Bias (i.e. it's random!)

Disadvantages: Difficult to obtain

Due to its very randomness, "freak" results can sometimes be obtained that
are not representative of the population. In addition, these freak results may
be difficult to spot. Increasing the sample size is the best way to eradicate
this problem.

Systematic Sample
With this method, items are chosen from the population according to a fixed rule, e.g. every
10th house along a street. This method should yield a more representative sample than the random
sample (especially if the sample size is small). It seeks to eliminate sources of bias, e.g. an inspector
checking sweets on a conveyor belt might unconsciously favor red sweets. However, a systematic
method can also introduce bias, e.g. the period chosen might coincide with the period of faulty
machine, thus yielding an unrepresentative number of faulty sweets.

Advantages:

Can eliminate other sources of bias

Disadvantages:

Can introduce bias where the pattern used for the samples coincides with a
pattern in the population.

Stratified Sampling
The population is broken down into categories, and a random sample is taken of each category. The
proportions of the sample sizes are the same as the proportion of each category to the whole.
Advantages:

Yields more accurate results than simple random sampling

Can show different tendencies within each category (e.g. men and women)

Disadvantages: Nothing major, hence it's used a lot

Quota Sampling
As with stratified samples, the population is broken down into different categories. However, the size
of the sample of each category does not reflect the population as a whole. This can be used where an
unrepresentative sample is desirable (e.g. you might want to interview more children than adults for a
survey on computer games), or where it would be too difficult to undertake a stratified sample.

Advantages:

Simpler to undertake than a stratified sample

Sometimes a deliberately biased sample is desirable

Disadvantages: Not a genuine random sample

Likely to yield a biased result

Cluster Sampling
Used when populations can be broken down into many different categories, or clusters (e.g. church
parishes). Rather than taking a sample from each cluster, a random selection of clusters is chosen to
represent the whole. Within each cluster, a random sample is taken.

Advantages:

Less expensive and time consuming than a fully random sample

Can show "regional" variations

Disadvantages: Not a genuine random sample

Likely to yield a biased result (especially if only a few clusters are
sampled)