Sample and Sampling Technique

Course Code: BBA 2201

Course Instructor
Fatema Khatun
Lecturer (Economics)
Department of Business Administration
Bangladesh Army University of Science and Technology
(BAUST), Saidpur
 Sampling
A finite subset of the population, selected from it with the objective of investigating its
properties is called a sample and the number of units in the sample is known as the sample
size.
Sampling is a tool which enables us to draw conclusions about the characteristics of the
population after studying only those objects or items that are included in the sample.
The main objectives of the sampling theory are :
i. To obtain the optimum results, i.e., the maximum information about the characteristics of
the population with the available sources at our disposal in terms of time, money and
manpower by studying the sample values only.
ii. To obtain the best possible estimates of the population parameters.
 PRINCIPLES OF SAMPLING
i. Large sample size: Logically, it seems that as the sample size increases, the sample is
more likely to reveal the true characteristics of the population and thus provide better
estimates of the parameters. It is known that the reliability of the sample statistic as an
estimate of the population parameter is proportional to the square root of the sample size
n. But due to certain limitations in terms of time, money and manpower, it is not always
possible to take very large samples. Moreover, the effort and cost of drawing large samples
might outlive the utility of the sample study as against the complete enumeration (census).
ii. Random selection: The sample should be selected at random from the population. By
random selection we mean a selection in which each and every unit in the population
has an equal chance of being selected in the sample.
Example: If we are interested in studying the average height of the students in BAUST, then it
is not desirable to resort to 100% enumeration of the students in the university. A fairly
adequate sample of the students from each college may be selected at random and the average
height of the students selected in the samples may be computed. Since the sample is random, it
would be representative of the population and the average so obtained will not differ much
from the true value (i.e., the average computed by the complete enumeration). This difference
is attributed to fluctuations of sampling.
 When to use Census Method ?
i. If the information is required about each and every unit of the population,
there is no way but to resort to 100% enumeration.
ii. In any manufacturing process in industry, 100% enumeration should be taken
recourse to under the following conditions :
a) The occurrence of a defect may cause loss of life or serious casualty to
personnel.
b) A defect may cause serious malfunction of the equipment.
iii. It may also be desirable to carry out complete census if,
a) N, the lot size is small and
b) the incoming lot quality is poor or unknown.
 Advantages of Sample Method
❖Speed, i.e., less time
❖Economy, i.e., Reduced Cost of the Enquiry
❖Administrative Convenience
❖Reliability
❖Greater Scope
❖Infinite or / and Hypothetical Population
❖Destructive Testing
 Limitations of Sampling
▪ If a sample survey is not properly planned (or designed) and executed carefully, the results obtained
will not be reliable and quite often might even be misleading. In this context, it may be worthwhile to
quote the words of Frederick F. Stephen :
“Samples are like medicines. They can be harmful when they are taken carelessly or without knowledge
of their effects…. Every good sample should have a proper lable with instructions about its use”.
Sampling design must be perfect otherwise it might lead to serious complications in the final results.
The omission of a few units in a complete census may be immaterial but non-response or incomplete
response from even one or two units in a small sample might have a significant effect on the final
result.
▪ An efficient sampling scheme requires the services of qualified, skilled and experienced personnel,
better supervision and more sophisticated equipment and statistical techniques for the planning and
execution of the survey and for the collection, processing and analysis of the sample data. In the absence
of these, the results of the survey may not be reliable.
▪ Sometimes the sample survey might require more time, money and labour than a complete census.
This will be so if the sample size is a large proportion of the population size and if complicated weighted
system is used.
▪ Sampling procedure cannot be used if we want to obtain information about each and every unit of
the population. Further, if the population is too heterogeneous, it may be impossible to use a sampling
procedure.
 Sampling and Non-Sampling Errors
Sampling Errors: In a sample survey, since only a small portion of the population is studied, its
results are bound to differ from the census results and thus have a certain amount of error. This
error would always be there, no matter that the sample is drawn at random and that it is highly
representative. This error is attributed to fluctuations of sampling and is called sampling error.
Sampling error is due to the fact that only a subset of the population (i.e., sample) has been used to
estimate the population parameters and draw inferences about the population. Thus, sampling error is
present only in a sample survey and is completely absent in census method.
Sampling errors are primarily due to the following reasons :
1. Faulty selection of the sample: Some of the bias is introduced by the use of defective sampling
technique for the selection of a sample, e.g., purposive or judgment sampling in which the investigator
deliberately selects a representative sample to obtain certain results. This bias can be overcome by
strictly adhering to a simple random sample or by selecting a sample at random subject to restrictions
which while improving the accuracy are of such nature that they do not introduce bias in the results.
2. Substitution: If difficulties arise in enumerating a particular sampling unit included in the random
sample, the investigators usually substitute a convenient member of the population. This obviously
leads to some bias since the characteristics possessed by the substituted unit will usually be different
from those possessed by the unit originally included in the sample.
 Sampling and Non-Sampling Errors
Sampling Errors
3. Faulty demarcation of sampling units: Bias due to defective demarcation of sampling
units is particularly significant in area surveys such as agricultural experiments in the field or
crop cutting surveys, etc. In such surveys, while dealing with border-line cases, it depends
more or less on the discretion of the investigator whether to include them in the sample or not.
4. Error due to bias in the estimation method: Sampling method consists in estimating the
parameters of the population by appropriate statistics computed from the sample. Improper
choice of the estimation techniques might introduce the error.
5. Variability of the population: Sampling error also depends on the variability or
heterogeneity of the population to be sampled.
 Sampling and Non-Sampling Errors
Non-Sampling Errors: Non-sampling errors are not attributed to chance and are a
consequence of certain factors which are within human control. In other words, they are due
to certain causes which can be traced and may arise at any stage of the enquiry, viz., planning
and execution of the survey and collection, processing and analysis of the data.
Non-sampling errors are thus present both in census surveys as well as sample surveys.
Obviously, non-sampling errors will be of large magnitude in a census survey than in a sample
survey because they increase with the increase in the number of units to be examined and
enumerated.
It is very difficult to prepare an exhaustive list of all the sources of non-sampling errors. We
enumerate below some of the important factors responsible for non-sampling errors in any survey
(census or sample).
➢Including vague and faulty definitions of the population or the statistical units to be used,
incomplete list of population-members (i.e., incomplete frame in case of sample survey).
➢Vague and imperfect questionnaire which might result in incomplete or wrong information.
➢Defective methods of interviewing and asking questions.
➢Vagueness about the type of the data to be collected.
 Sampling and Non-Sampling Errors
Non-Sampling Errors:
➢ Exaggerated or wrong answers to the questions which appeal to the pride or prestige or self-interest of the
respondents. For example, a person may over-state his education or income or understate his age or he may give
wrong statements to safeguard his self-interest.
➢ Personal bias of the investigator.
➢ Lack of trained and qualified investigators and lack of supervisory staff.
➢ Failure of respondents’ memory to recall the events or happenings in the past.
➢ Improper coverage: If the objects of the survey are not precisely stated in clear cut terms, this may result in
➢ the inclusion in the survey of certain units which are to be excluded,
➢ or the exclusion of certain units which were to be included in the survey under the objectives
➢ Compiling Errors, i.e., wrong calculations or entries made during the processing and analysis of the data.
Various operations of data processing such as editing and coding of the responses, punching of cards, tabulation
and summarizing the original observations made in the survey are a potential source of error. Compilation errors
are subject to control through verification, consistency checks, etc.
➢ Publication Errors. Publication errors, i.e., the errors committed during presentation and printing of tabulated
results are basically due to two sources. The first refers to the mechanics of publication-the proofing error and the
like. The other, which is of a more serious nature, lies in the failure of the survey organization to point out the
limitations of the statistics.
 Types of Sampling Technique
Simple Random Sampling

Systematic Random Sampling

Random/
Probability
Stratified Sampling

Cluster Sampling
Sampling
Convenience sampling

Judgmental or purposive sampling

Non-random/
Non-probability Snowball sampling

Quota sampling
 Types of Sampling Technique
Probability Sampling
• Under probability sampling, every element in the sampling frame has a known and an
equal chance of getting selected in the sample.
• Probability sampling allows us to make a statistical statement about the accuracy of the
sample results and they are necessary for testing for statistical significance.
• They are also more representative of the population and hence the results are more
generalizable.

Properties of Probability Sampling

• The probability sample needs a sampling frame.
• Each element on the sampling frame has a known probability of selection.
• The selection process is unbiased or random.
 Types of Sampling Technique

Types of Probability Sampling

1. Simple Random Sampling
2. Systematic Random Sampling
3. Stratified Random Sampling
4. Cluster Sampling
 Simple Random Sampling
Simple Random Sampling
• A sample selected so that each item or person in the population has the same chance of
being included. Each possible sample to have an equal probability of being picked.
• Each item in the entire population to have an equal chance of being included in the sample.
Merits of Simple Random Sampling
• Since it is a probability sampling, it eliminates the bias due to the personal judgment or
discretion of the investigator. Accordingly, the sample selected is more representative of
the population than in the case of judgment sampling.
• Because of its random character, it is possible to ascertain the efficiency of the estimates
by considering the standard errors of their sampling distributions. Moreover, large sample
will be more representative of the population and thus provide better results.
• The theory of random sampling is highly developed so that it enables us to obtain the most
reliable and maximum information at the least cost, and results in savings in time,
money and labour.
 Simple Random Sampling
Demerits of Simple Random Sampling
• Simple random sampling requires an up-to-date frame, i.e., a complete and up-to-date list of the
population units to be sampled. In practice, since this is not readily available in many inquiries, it
restricts the use of this sampling design.
• In field surveys if the area of coverage is fairly large, then the units selected in the random sample
are expected to be scattered widely geographically and thus it may be quite time consuming and
costly to collect the requisite information or data.
• If the sample is not sufficiently large, then it may not be representative of the population and thus
may not reflect the true characteristics of the population
• The numbering of the population units and the preparation of the slips is quite time consuming and
uneconomical particularly if the population is large. Accordingly, this method can’t be used
effectively to collect most of the data in social sciences.
• For given degree of accuracy, simple random sampling usually requires larger sample as compared
to stratified random sampling.
• Sometimes, simple random sample gives results which are highly improbabilistic in nature, i.e.,
whose probability is very small. For example, a random selection of 13 cards from a pack of 52 cards
might give all thirteen cards of spades, say. The probability of the happening of such an event in
practice is very small.
 Systematic Random Sampling
• Elements are selected from the population at a uniform interval that is measured in time,
order, or space.
• A random starting point is selected, and then every kth member of the population is
selected.
• Even though systematic sampling may be inappropriate when the elements lie in a
sequential pattern, this method may require less time and sometimes results in lower
costs than the simple random-sample method.
• The basic problems in stratified random sampling are :
i. The stratification of the universe into different starta or sub-groups.
ii. The determination of the sizes of the samples to be drawn from different starta.
• Both these points are equally important. A faulty stratification cannot be compensated
even by taking large samples.
 Merits of Systematic Random Sampling
▪ Systematic sampling is very easy to operate and checking can also be done quickly.
Accordingly, it results in considerable saving in time and labour relative to simple random
sampling or stratified random sampling.
▪ Systematic sampling may be more efficient than simple random sampling provided the
frame is complete and up-to-date and the units are arranged serially in a random order like
the names in a telephone directory where the units are arranged in alphabetical order.
However, even in alphabetical arrangement, certain amount of non-random character may
persist.
 Demerits of Systematic Random Sampling
▪ Systematic sampling works well only if the complete and up-to-date frame is available and if
the units are randomly arranged. However, these requirements are not generally fulfilled.
▪ Systematic sampling gives biased results if there are periodic features in the frame and
the sampling interval (k) is equal to or a multiple of the period.

▪ The relative efficiency of the systematic sampling over stratified random sampling or
simple random sampling without replacement largely depends on the properties of the
population under study. Without a knowledge of the structure of the population, no hard
and fast rules can be laid down and no situations can be pinpointed where the use of
systematic sampling is to be recommended.
 Stratified Random Sampling

➢ A population is divided into subgroups, called strata, and

a sample is randomly selected from each stratum.
➢ Appropriate when population is already divided into groups of different
sizes and wish to acknowledge the fact.
➢ Use when each group has small variation within itself but there is a wide
variation between the groups.
➢ The basic problems in stratified random sampling are :
➢ The stratification of the universe into different starta or sub-groups.
➢ The determination of the sizes of the samples to be drawn from
different starta.
➢ Both these points are equally important. A faulty stratification cannot be
compensated even by taking large samples.
 Merits of Stratified Random Sampling
1. More Representative Sample: A properly constructed and executed stratified random
sampling plan overcomes the drawbacks of purposive sampling and random sampling and
still enjoys the virtues of both these methods by dividing the given universe into a number of
homogeneous subgroups with respect to purposive characteristic and then using the
technique of random sampling in drawing samples from each stratum. A stratified random
sample gives adequate representation to each strata or important section of the population
and eliminates the possibility of any important group of the population being completely
ignored. The stratified random sampling provides a more representative sample of the
population and accordingly results in less variability as compared with other sampling
designs.
2. Greater Precision: As a consequence of the reduction in the variability within each
stratum, stratified random sampling provides more efficient estimates as compared with
simple random sampling.
 Merits of Stratified Random Sampling
3. Administrative Convenience: The division of the population into relatively homogeneous
subgroups brings administrative convenience. Unlike random samples, the stratified samples
are expected to be localized geographically. This ultimately results in reduction in cost and
saving in time in terms of collection of the data, interviewing the respondents and supervision
of the field work.
4. Sometimes it is desired to achieve different degrees of accuracy for different segments of
the population. Stratified random sampling is the only sampling plan which enables us to
obtain the results of known precision for each of the stratum.
5. Quite often, the sampling problems differ quite significantly in different segments of the
population. In such a situation, the problem can be tackled effectively through stratified
sampling by regarding each segment of the population as a different strata and approaching
upon them independently during sampling.
 Demerits of Stratified Random Sampling
• The success of stratified random sampling depends on : (i) Effective stratification of the
universe into homogeneous strata and (ii) Appropriate size of the samples to be drawn
from each of the stratum.
• If stratification is faulty, the results will be biased. The error due to wrong
stratification cannot be compensated even by taking large samples.
• Disproportional stratified sampling requires the assignment of weights to different strata
and if the weights assigned are faulty, the resulting sample will not be representative and
might give biased results.
 Cluster Sampling
• A population is divided into clusters using naturally
occurring geographic or other boundaries. Then,
clusters are randomly selected and a sample is collected
by randomly selecting from each cluster.
• Use when there is considerable variation within each group but the groups
are essentially similar to each other.

In using cluster sampling the following points should be borne in mind :

(i) Clusters should be as small as possible consistent with the cost and
limitations of the survey, and
(ii) The number of sampling units in each cluster should be approximately same.
 Types of Sampling Technique

Non-probability sampling
This sampling method is not a fixed or predefined selection process.
This makes it difficult for all population elements to have equal opportunities to be
included in a sample.

CHAP 1-24
 Types of Sampling Technique

Types of Non-probability sampling

▪ Convenience sampling
▪ Judgmental sampling
▪ Quota sampling
▪ Snowball sampling

CHAP 1-25
 Convenience sampling
❑ A non-probability sampling technique that attempts to obtain a sample of convenient
elements.
❑ The selection of sampling units is left primarily to the interviewer.
❑ It is the least expensive and least time consuming of all sampling techniques.
❑ The sampling units are accessible, easy to measure and cooperative.
❑ It can be used in exploratory research for generating ideas, insights or hypotheses.
❑ It can be used for pretesting questionnaires, or pilot studies.

CHAP 1-26
 Convenience sampling
Despite these advantages, this form of sampling has serious limitations.
❑ Selection bias, including respondent self-selection
❑ Not representative of any definable population
❑ Not theoretically meaningful to generalize to any population from a
convenience sample
❑ Not appropriate for marketing research projects involving population
inferences.
❑ Not recommended for descriptive or causal research

CHAP 1-27
 Judgmental sampling
❑ A form of convenience sampling in which the population elements are
purposely selected based on the judgement of the researcher.
❑ The researcher, exercising judgement or expertise, chooses the elements to
be included in the sample because he or she believes that they are
representative of the population of interest or are otherwise appropriate.
❑ Judgmental sampling is inexpensive, convenient and quick, yet it does
not allow direct generalizations to a specific population, usually because
the population is not defined explicitly.
❑ Subjective and its value depends entirely on the researcher’s judgement,
expertise and creativity.
❑ Can be useful if broad population inferences are not required.
❑ Frequently used in business-to-business marketing research projects,
given that in many projects the target population is relatively small.
CHAP 1-28
 Quota Sampling
▪ Quota sampling may be looked as a special form of stratified sampling.
▪ In this method, the investigator is told in advance the number of the sample units he is to
examine or enumerate from the stratum assigned to him.
▪ In the language of stratified sampling, the quota of the units to be examined by the
investigator from the stratum assigned to him is fixed for each investigator.
▪ The sampling quotas may be fixed according to some specified characteristic such as
income group, sex, occupation, political or religious affiliations, etc.
▪ The choice of the particular units or individuals for investigation is left to the
investigators themselves. They are merely given the quotas with the specific instruction to
inspect (interview) a specified number of units (informants) from each stratum.
▪ Quite often the investigator does not make a random selection of the sample units. He
usually applies his judgment and discretion in the choice of the sample and tries to get the
desired information as quickly as possible.
▪ Moreover, in case of non-response from some of the selected sample units (due to certain
reasons like non-availability of the respondent even after repeated calls by the investigator,
or the inability or refusal of the informant to furnish the requisite information), the
investigator selects some fresh units himself to complete his quota. In doing so, he is likely
to include some purposive units to get the desired information.

CHAP 1-29
 Merits & Demerits of Quota Sampling
Merits
1. Quota sampling is a stratified-cum-purposive or judgment sampling and thus enjoys the benefits of
both. It aims at making the best use of stratification without incurring high costs involved in
following any probabilistic method of sampling. There is considerable saving in time and money as the
sampled units may be so selected that they are close together.
2. If carefully executed by skilled and experienced investigators who are aware of the limitations of
judgment sampling and if proper controls or checks are imposed on the investigators, quota sampling
is likely to give quite reliable results.
Demerits
Since quota sampling is a restricted type of judgment sampling, it suffers from all the limitations of
judgment or purposive sampling, viz.,
❑ It may be biased because of the personal beliefs and prejudices of the investigator in the selection
of the units or/and inspecting them.
❑ It may involve the bias due to the substitution of the sampled units from where there is no
response.
❑ Since it is not based on random sampling, the sampling error cannot be estimated.
In spite of all these shortcomings, the technique of quota sampling is generally adopted in market
surveys, political surveys, or surveys of opinion poll where it is very difficult, rather impossible, to
identify the strata in advance.
 Snowball Sampling
An initial group of respondents is selected, sometimes on a random basis, but more typically
targeted at a few individuals who are known to possess the desired characteristics of the
target population.
After being interviewed, these respondents are asked to identify others who also belong to
the target population of interest. Subsequent respondents are selected based on the
referrals. By obtaining referrals from referrals, this process may be carried out in waves,
thus leading to a snowballing effect.
Even though probability sampling can be used to select the initial respondents, the final
sample is a nonprobability sample.
The main objective of snowball sampling is to estimate characteristics that are rare in the
wider population.
The major advantage of snowball sampling is that it substantially increases the likelihood of
locating the desired characteristic in the population.
It also results in relatively low sampling variance and costs.

CHAP 1-31
 Uses of non-probability sampling
Create a hypothesis
To create an assumption when limited to no prior information is available.
To help with the immediate return of data and builds a base for further research.
Exploratory research
Use widely when conducting qualitative research, pilot studies, or exploratory research.
Budget and time constraints
Use when there are budget and time constraints, and some preliminary data must be
collected.
The survey design is not rigid, it is easier to pick respondents randomly and have them
take the survey or questionnaire.

CHAP 1-32
 SAMPLING DISTRIBUTIONS
Standard deviation
▪Standard deviation (SD) measures the amount of variability, or dispersion, from the
individual data values to the mean.
▪Standard deviation is a descriptive statistic, which means it helps to describe or summarize
the dataset. In simple terms, standard deviation tells, on average, how far each value within
the dataset lies from the mean.
▪A high standard deviation means that the values within a dataset are generally positioned far
away from the mean, while a low standard deviation indicates that the values tend to be
clustered close to the mean.
▪In a nutshell, it measures how much “spread” or variability there is within the dataset.
 SAMPLING DISTRIBUTIONS
Standard Error
➢Standard error (or standard error of the mean) is an inferential statistic that tells, in simple
terms, how accurately the sample data represents the whole population.
➢For example, if you conduct a survey of people living in New York, you’re collecting a
sample of data that represents a segment of the entire population of New York. Different
samples of the same population will give you different results, so it’s important to understand
how applicable your findings are. So, when you take the mean results from your sample data
and compare it with the overall population mean on a distribution, the standard error tells what
the variance is between the two means. In other words, how much would the sample mean
vary if you were to repeat the same study with a different sample of people from the New York
City population?
➢Just like standard deviation, standard error is a measure of variability. However, the difference
is that standard deviation describes variability within a single sample, while standard
error describes variability across multiple samples of a population.
➢A probability distribution of all the possible means of the samples is a distribution of the
sample means. Statisticians call this a sampling distribution of the mean.
Standard Deviation vs. Standard Error
Standard Deviation vs. Standard Error
 Appropriate Use Cases for Standard Deviation
Standard deviation (SD) is primarily used when:
1. Describing variability in a dataset: SD provides a measure of how spread out the data
points are from the mean.
2. Identifying outliers: Data points that fall outside 2-3 standard deviations from the mean
are often considered outliers.
3. Quality control applications: In manufacturing, SD helps monitor process consistency.
4. Comparing datasets: SD allows for meaningful comparisons of variability between
different datasets.

Example: Manufacturing Process Variability

Imagine a pharmaceutical company producing tablets. The ideal weight is 500mg, with an
acceptable range of ±20mg. By calculating the SD of tablet weights, quality control can
determine if the production process is consistent:
•SD = 5mg: 95% of tablets fall within 490-510mg (good consistency)
•SD = 15mg: Only 68% of tablets fall within 485-515mg (poor consistency)
In this case, using standard deviation helps assess product quality and process stability.
 Appropriate Use Cases for Standard Error
Standard error (SE) is typically used when:
1. Estimating population parameters: SE helps quantify the precision of sample statistics
as estimates of population parameters.
2. Constructing confidence intervals: SE is crucial in determining the range within which
the true population parameter likely falls.
3. Hypothesis testing: SE is used in calculating test statistics for various statistical tests.
4. Reporting research findings: SE communicates the reliability of sample means in
representing the population mean.

Example: Political Polling Accuracy

Consider a political poll estimating voter preferences. If a candidate’s support is reported as
52% with a standard error of 3%, we can interpret this as:
The best estimate of the candidate’s true support in the population is 52%.
There’s a 68% chance that the true population support falls within 49-55% (±1 SE).
There’s a 95% chance that the true population support falls within 46-58% (±2 SE).
In this scenario, using standard error provides insights into the precision of the poll’s estimate.
 Central Limit Theorem
If all samples of a particular size are selected from any population, the sampling distribution of
the sample mean is approximately a normal distribution. This approximation improves with
larger samples.
As the sample size increases, the sampling distribution of the mean will approach
normality, regardless of the shape of the population distribution.
This relationship between the shape of the population distribution and the shape of the
sampling distribution of the mean is called the central limit theorem. The central limit theorem
is perhaps the most important theorem in all of statistical inference.
The significance of the central limit theorem is that it permits us to use sample statistics to
make inferences about population parameters without knowing anything about the shape of the
frequency distribution of that population other than what we can get from the sample.
Standardizing the Sample Mean
Standardizing the Sample Mean
Standardizing the Sample Mean
Standardizing the Sample Mean
Standardizing the Sample Mean
 Thank you

CHAP 1-45