Federal University Dutsin-Ma Journal of Educational Research (Fudjer), 2 (1): 47-56.

Research Sampling and Sample Size Determination: A practical Application

Chinelo Blessing ORIBHABOR (Ph.D)

Department of Guidance and Counseling, Faculty of Arts and Education, University of Africa,
Bayelsa State.
E-Mail: chiblessing42004@yahoo.co.uk; chinelo.oribhabor@uat.edu.ng
Phone Number: 08034982069

And

Chioma A. ANYANWU (Ph.D)

Department of Guidance and Counseling, Faculty of Arts and Education, University of Africa,
Bayelsa State.
E-Mail: akudo.anyanwu@uat.edu.ng
Phone Number: 08036648341

Abstract
One of the major issues in planning a research is the decision as to how a sample and the method
to be employed to select the estimated sample in order to meet the objective of the research.
Besides emphasizing the need for a representative sample, the importance of sampling was
examined in this study. Sampling is an essential tool for research in Education. Specific sampling
techniques are used for specific research problems because one technique may not be appropriate
for all problems. Similarly, if the sample size is inappropriate it may lead to erroneous
conclusions. The aim of this paper is to sensitize our researchers on the importance of proper
sampling and sample size determination. The various types of probability and non-probability
sampling techniques were explained in this paper.

Keywords: Research, Survey, Population, Sample Size, Sampling Techniques

Introduction
Research is the orderly investigation of a subject matter for the purpose of adding to knowledge
Krieger (2012). Survey is a method of collecting large scale quantitative data, but it does not use
an experimental design. In this type of research, a number of data-gathering psychometric tools
and procedures are used. These include questionnaires, tests, checklists, rating scales, score
cards, inventories, interviews, etc. The study can be used to ascertain the nature of a
phenomenon from a relatively large number of cases. When you study the entire target
population, the survey is called census. Most of the times, the entire population is too large to be
 Federal University Dutsin-Ma Journal of Educational Research (Fudjer), 2 (1): 47-56.

handled. In this case, you have to use appropriate sampling technique to obtain a truly
representative sample. Krieger (2012) defined population as all members of any well-defined
class of people, events or objects. It means therefore that any entity, group or set which
constitutes a population must have at least one attribute or characteristic which is common to all
of them. The population of a study therefore represents the target of the study as defined by the
aims and objectives of the study. Most of the times, the target population is too large for a
researcher to study. The number of subjects may be too many for the researcher to handle with
limited resources available. The geographical spread or area may be two wide to cover with the
limited time for the research. As a result of cost, time and other constraints, it may be very
difficult to study the entire population. It becomes necessary, reasonable and only feasible to
study a portion of the population which is described as sample.

When the population of the study is too large for a complete census to be taken, it becomes
indispensable to take a sample out of it for an effective research. According to Osuala (2007),
“Sampling is taking any portion of a population or universe as representative of that population
or universe.” To make a research to be effective and plausible, a reasonable portion of the
population should be sampled. This would make for a reliable generalization of the findings. A
sample is a set of individuals or participants selected from a larger population for the purpose of
a survey (Salant & Dillman, 2004). A sample can be defined as a group of relatively smaller
number of people selected from a population for investigation purpose. An optimal sample is
important for minimizing the cost of sampling error, thus indicating the need for selecting an
appropriate sample size. Specifically, (Salkind, 2010) emphasized that an appropriate sample
size is necessary for any research because too small sample size is not a good representative of
the population. Too small sample size may lead to committing Type I error, which is the
probability of wrongly rejecting a particular finding when it in fact to be accepted (Sekaran,
2013). Furthermore, Sekaran (2013) argued that too large sample size is not appropriate because
of possible problem of type II error, which is accepting a particular finding when it is supposed
to be rejected.

Determination of the sample size for parameters

 Federal University Dutsin-Ma Journal of Educational Research (Fudjer), 2 (1): 47-56.

A sample size generally depends on five study design parameters: minimum expected difference
or also known as the effect size, estimated measurement variability, desired statistical power,
significance criterion, and whether a one- or two-tailed statistical analysis is planned (Scott,
2007).
Basis of Minimum Expected Difference or size effects: According to Fox, Hunn & Mathers
(2007), the minimum expected difference is made smaller, the sample size needed to detect
statistical significance increases. The setting of this parameter is subjective and is based on
clinical judgment or experience with the problem being investigated. For example, suppose a
study is designed to compare a standard diagnostic procedure of 85% accuracy with a new
procedure of unknown but potentially higher accuracy. Suppose the investigator believes that it
would be a clinically important improvement if the new procedure were 95% accurate.
Therefore, the investigator would choose a minimum expected difference of 10% (0.10).
Estimated Measurement Variability: This parameter is represented by the expected standard
deviation in the measurements decide within each comparison group (Taherdoost, 2017). As
statistical variability increases, the sample size needed to detect the minimum difference
increases. Ideally, the estimated measurement variability should be determined on the basis of
preliminary data collected from a similar study population. If preliminary data are not available,
this parameter may have to be estimated on the basis of subjective experience, or a range of
values may be assumed. A separate estimate of measurement variability is not required when the
measurement being compared is a proportion (in contrast to a mean), because the standard
deviation is mathematically derived from the proportion (Taherdoost, 2017).
Based on Statistical Power: This parameter is the power that is need from the study. As power
is increased, sample size increases. While high power is always desirable, there is an obvious
trade-off with the number of individuals that can feasibly be investigated, given the usually fixed
amount of time and resources available to conduct a research or investigational study. In
randomized controlled trials, statistical power is customarily set to a number greater than or
equal to 0.80, with many clinical trial experts now advocating a power of 0.90 (Writes, 2012)
Significance Criterion (P-Value): This parameter is the maximum P value for which a
difference is to be considered statistically significant (Taherdoost, 2017). As the significance
criterion is decreased, the sample size needed to detect the minimum difference increases. The
statistical significance criterion is customarily set to 5 percent.
 Federal University Dutsin-Ma Journal of Educational Research (Fudjer), 2 (1): 47-56.

One- or Two-tailed Statistical Analysis: In a some cases, it may be known before the
investigation that any difference between comparison or experimental groups is possible in only
one direction (Bartlett, Kotrlik & Higgins 2001). In such cases, use of a one-tailed statistical
analysis, which would require a smaller sample size for detection of the minimum difference
than would a two-tailed analysis, may be considered. The sample size of a one-tailed study
design with a given statistical significance criterion—for example, α—is equal to the sample size
of a two-tailed design with a significance criterion of 2α, all other parameters being equal
(Bartlett, Kotrlik & Higgins 2001).

Ticehurst (2009) have pointed out the importance of determining an absolute sample size that is
independent of the study population, thus indicating the need for method of determining a
sample size such as statistical power test. Specifically, Cohen (2007) stressed that sample size
should be determined using a suitable power of statistical test. Therefore, in deciding about
sample size for the study being reported, power of a test becomes a viable option. The power of a
statistical test is defined as the probability of rejecting a null hypothesis or rejecting a specific
effect size of a particular sample size at a particular alpha level (Cohen, 2008). The test has the
capacity to detect a difference if it truly exists in the wider population. In addition, even if
sample size to be used in a particular study has been determined through other methods, it is still
appropriate and worthy to use power analysis so that the probability of detecting the effects of
different sample sizes is explicitly known (Ramalu, 2010).
However, Meyer (1979), Fox, Hunn & Mathers (2007) suggest the following range of population
and the corresponding sample size to derive there from:
Population Range Approximate Sample Size
Infinity 384
500, 000 384
100, 000 383
50, 000 381
10, 000 370
5, 000 357
3, 000 341
2, 000 322
1, 000 278

It should be noted that the above range is a suggestion, which works best at the convenience of
the researcher and the applicability of the method to the research. A researcher is therefore not
 Federal University Dutsin-Ma Journal of Educational Research (Fudjer), 2 (1): 47-56.

compelled to follow the suggested range of sample size hook, line and sinker. Another good and
useful suggestion for determining the sample size for a study is the Yaro Yamane Statistical
Formula for finding sample size of a finite population. This method is only applicable when the
numerical strength of the population is known. The formula is:
n = N / [1 + N (e) 2]

Where; n = the sample size

N = the finite population
e = the level of significance or limit of tolerable error
1 = unit or a constant
Thus, if a researcher decides to study a universe which constitutes a total of 1,024 for instance,
with a tolerable limit of error or the level of significance at 0.05, using the above formula, the
sample size would be determined thus: From the above example, the sample size for a study
population of 1,024 is approximately 400, which also is approximately 39% of the population.

Another good method for determining a representative sample size was suggested (Dillman,
2000). Thus, given the population size of 2,400, the sample size was computed using the below
formula:

n = [(N) (p) (1 − p)] / [(N − 1) (B/C) 2 + (p) (1 − p)]

Where n is the computed sample size needed for the desired level of precision; N is the
population size; p is the proportion of population expected to choose; B is acceptable amount of
sampling error, or precision; and finally C is Z statistic associated with the confidence level
which is 1.96 that corresponds to the 95% level. B can be set at 0.1, 0.05, or 0.03, which are ±
10, 5, or 3% of the true population value, respectively. The acceptable amount of sampling error
or precision is set at 0.05 or 5%. Confidence level of 1.96 corresponds to the 95% level. Using
0.05 will lead to a greater sample size than using 0.03; however, it always provides an adequate
sample size for a smaller or greater population (Biemer & Lyberg, 2003).
Where N = 2,400, p = 0.5, B = 0.05, C = 1.96
n= [(2,400) (0.5) (1-0.5)] / [(2400-1) (0.05/1.96)2 + (0.5) (1-0.5)]
 Federal University Dutsin-Ma Journal of Educational Research (Fudjer), 2 (1): 47-56.

n= [(2,400) (0.5) (0.5)]/ [(2399) (0.0255)2 + (0.5) (0.5)]

n= (600)/ (1.56+ 0.25)
n= 600/ 1.81
n= 331.49 ≈ 332

Based on the results of sample size computation, this study needed 290 participants to complete
the survey.

Factors affecting sample size

a. Population — Statistical theory proves that in most types of populations where statistical
sampling applies, the population size is only a minor consideration Noordzig, Tripepi,
Dekker, Zoccali, Tanck, Jager (2010). This is because representativeness is ensured by
the random selection process. Once an adequate sample size is obtained that includes a
good cross-section of items, additional items are not needed regardless of population size.
b. Confidence level/interval — an increase in the desired confidence level or decrease in the
confidence interval will both result in an increase in the necessary sample size (Noordzig,
et al. 2010). For example: An increase of the confidence level from 80% to 90% would
increase the required sample size by approximately 65%, or A decrease in the confidence
interval will increase the required sample size proportionally.
c. Cost benefit ratio — before beginning a research, consideration should be given to the
cost and time needed to conduct the research using statistical sampling techniques
(Sukhatme, 2008). The use of the random method for selecting a sample may impose an
additional time requirement. A research should not be conducted just for the sake of
testing.
When the sample size for a study is determined, the next invaluable stage is the procedure for
selecting the particularly subjects or respondents for the study. The whole process and procedure
involved in selecting the particular subjects or respondents for investigation in a research is
called sampling procedure or sampling technique. The general aim of all sampling methods is to
obtain a sample that is representative of the target population. By this we mean that, as much as
possible, the information derived from the sample survey is the same.
 Federal University Dutsin-Ma Journal of Educational Research (Fudjer), 2 (1): 47-56.

When selecting a sampling method, we need some minimal prior knowledge of the target
population; with this and some reasonable assumptions, we can estimate a sample size required
to achieve a reasonable estimate (with acceptable precision and accuracy) of population
characteristics.
How we actually decide which sampling units will be chosen makes up the sampling method.
Sampling methods can be categorized according to the approach they take to the probability of a
particular unit being included (Scott, 2007). Most sampling methods attempt to select units such
that each has a definable probability of being chosen. Moreover, most of these methods also
attempt to ensure that each unit has the same chance of being included as every other unit in the
sample frame. All methods that adopt this general approach are called probability sampling
methods (Scott, 2007).
Types of Sampling Technique
According to Noordzig et al. (2010), there are two types of sampling techniques:
1) Probability sampling technique;
2) Non-probability sampling technique

Basic probabilistic sampling techniques

A probability technique is a sampling procedure in which the sample is selected based on well-
defined mathematical guidelines whereby every member of the universe has equal chances of
being selected (Noordzig et al. 2010). The types of probabilistic sampling techniques are:

(i) Simple Random Sampling: One of the most outstanding features of this method is that each
member of the population has an equal chance of being selected (Wimmer & Dominick, 2006).
The simple random sampling operates in two ways: replacement of the subject after being
selected and non-replacement of the subject after selection. When it is sampling with
replacement, any member of the population that is selected still has a further chance of being
selected because after the first selection it is returned to the fold. On the other hand, when it is
random sampling without replacement a member of the population that is selected is removed
from the fold and cannot be selected again. Random sampling without replacement is the most
widely used random sampling method.
 Federal University Dutsin-Ma Journal of Educational Research (Fudjer), 2 (1): 47-56.

The most popular approach to simple random sampling is the use of a table of random numbers.
In this method the researcher arbitrarily picks the starting by going up, down, left and right on
the table or random numbers, or even randomly throughout the table (Wimmer & Dominick,
2006). Apart from the use of the table of random numbers, a crude method can also be used in
the simple random sampling. In this case, every member of the population is represented by a
serial number. The serial number is written on a tally and put in a bag and mixed. The researcher
then closes his eyes and picks out one tally at a time till the required number of sample is picked.
The serial number on the selected tally then determines the particular member of the population
to be selected for study.
(ii) Stratified Sampling involves dividing the population into homogeneous non-overlapping
groups (i.e., strata), selecting a sample from each group, and conducting a simple random
sample in each stratum (Cochran, 1993). On the basis of information available from a frame,
units are allocated to strata by placing within the same stratum, those units which are more-or-
less similar with respect to the characteristics being measured. If this can be reasonably achieved,
the strata will become homogenous, i.e., the unit-to-unit variability within a stratum will be
small. In proportional allocation, the sample size in a stratum is made proportional to the number
of units in the stratum (Yamane, 2007; Raj, 2008; Sukhatme, 2008). In equal allocation, the
same number of units is taken from each stratum irrespective of the size of the stratum (Raj,
2008).
(iii) Cluster Sampling: The smallest units into which a population can be divided are called the
elements of the population, and groups of elements of the clusters (Sukhatme, 2008). The
problem with random sampling methods when sampling a population that's distributed across a
wide geographic region lies in covering a lot of ground geographically in order to get to each of
the units sampled (Trochim, 2010). This geographic trotting to collect samples is an expensive
affair. However, without taking samples from across the whole geographic population, it may
become difficult to conclude anything affirmatively about the population. The impasse is to
determine the best size of the cluster for a specified cost of the survey (Raj, 2008; Cochran,
1993). This predicament can be solved if the cost of the survey and the variance of the estimate
can be expressed as functions of the size of the cluster (Raj, 2008). According to (Trochim,
2010), cluster sampling includes:
1. Divide population into clusters
 Federal University Dutsin-Ma Journal of Educational Research (Fudjer), 2 (1): 47-56.

2. Randomly sample clusters

3. Measure all units within sampled clusters

Cluster sampling is ordinarily conducted in order to reduce costs. The variance of the estimate of
the mean in simple random sampling of clusters depends on the sample size, the population
variance and on the correlation of the variate of interest between units within the same cluster. If
the units within a cluster are more similar than units belonging to different clusters, the estimator
is subject to a larger variance; thus the smaller the intra-cluster correlation, the better (Raj, 2008).
(iv)Multi-stage Sampling: Multistage sampling involves combining various probability
techniques in the most efficient and effective manner possible. The process of estimation is
carried out stage by stage, using the most appropriate methods of estimation at each stage.
Sukhatme (2008) mentions that for a given number of elements, greater precision is attained by
distributing the elements over a large number of clusters than by taking a small number of
clusters and sampling a large number of elements from each one of them. Quite often, auxiliary
information is used to improve the precision of an estimate. However, in the absence of auxiliary
information, it may be advantageous to conduct the enquiry in two phases. In the first phase,
auxiliary information is collected on the variate of a fairly large sample. Then a sub-sample is
taken, and information collected on the variate of interest. Then the two samples are used in the
best possible manner to produce an estimate for the variate of interest. The procedure of first
selecting clusters and then choosing a specified number of elements from each selected cluster is
known as subsampling (Raj, 2008). It is also known as two-stage sampling or double sampling
(Cochran, 1993; Neyman, 2008). The clusters, which form the units of sampling at the first stage,
are called first stage units, and the elements or group of elements within clusters, which form the
units of sampling at the second stage, are called sub-units or second-stage units (Sukhatme,
2008). The procedure can be easily generalized to three or more stages and hence known as
multi-stage sampling.
Advantages of probability sampling technique are:
a) This sampling technique reduces the chance of systematic errors.
b) The methods minimize the chance of sampling biases. Sampling bias occurs when the
sampling frame and the population are not consonant (i.e., not the same). This could
 Federal University Dutsin-Ma Journal of Educational Research (Fudjer), 2 (1): 47-56.

happen if we were unable to obtain permission to get access to the list of the population
we are interested in, which is a common occurrence (Trochim, 2010).
c) A better representative sample is produced using probability sampling techniques. For
example, systematic sampling creates samples that are highly representative of the
population, without the need for a random number generator.
d) Inferences drawn from sample are generalisable to the population.

Disadvantages of probability sampling technique are:

a) The techniques need a lot of efforts. In simple random sampling, the selection of sample
becomes impossible if the units or items are widely dispersed.
b) A lot of time is consumed. For example, simple random sampling and stratified random
sampling are tedious and time consuming, especially when creating larger samples.
c) They are expensive. From the point of view of field survey it has been claimed that the
cases selected by random sampling tend to be too widely dispersed geographically and
the cost of collecting data becomes very large (Trochim, 2010).
Non-probabilistic sampling
Non-probability technique, on the other hand, is a sampling procedure that does not follow any
mathematical guideline, and as such the members of the population do not have equal chance of
being selected. This method is more or less based on the principle of availability, not on
prescribed formula or guidelines. Types of non-probabilistic sampling techniques are:
(i) Purposive Sampling: In purposive sampling, sampling is done with a purpose in mind.
Purposive sampling can be very useful for situations where one needs to reach a targeted sample
quickly and where sampling for proportionality is not the primary concern (Anaekwe, 2002).
With a purposive sample, it is likely to get the opinions of ones’ target population, but one is
also likely to overweight subgroups in ones’ population that are more readily accessible. All of
the methods that follow can be considered subcategories of purposive sampling methods.
According to Anaekwe (2002), one might sample for specific groups or types of people as in
modal instance, expert, or quota sampling. One might sample for diversity as in heterogeneity
sampling. Or, one might capitalize on informal social networks to identify specific respondents
who are hard to locate otherwise, as in snowball sampling.
 Federal University Dutsin-Ma Journal of Educational Research (Fudjer), 2 (1): 47-56.

(ii) Expert Sampling: Expert sampling involves the assembling of a sample of persons with
known or demonstrable experience and expertise in some area (Anaekwe, 2002). Often, we
convene such a sample under the auspices of a "panel of experts." There are actually two reasons
one might do expert sampling. First, because it would be the best way to elicit the views of
persons who have specific expertise. In this case, expert sampling is essentially just a specific
sub case of purposive sampling. But the other reason one might use expert sampling is to provide
evidence for the validity of another sampling approach one has chosen.
(iii) Quota Sampling: In quota sampling, the researcher selects people non-randomly according
to some fixed quota. There are two types of quota sampling: proportional and non- proportional.
(a) Proportional quota sampling: Here, the researcher represents the major characteristics of
the population by sampling a proportional amount of each. For instance, if one knows the
population has 40% women and 60% men, and that the researcher wants a total sample size of
100, the researcher will continue sampling until he/she gets those percentages and then stop. So,
if the researcher already got the 40 women for his/ her sample, but not the sixty men, the
researcher will continue to sample men but even if legitimate women respondents come along,
the researcher will not sample them because he/she has already "met his/her quota." The problem
here is that the researcher has to decide the specific characteristics on which he/she will base the
quota. Will it be by gender, age, education race, religion, etc.? (Scott, 2007)
(b) Non-Proportional Quota Sampling: This is a bit less restrictive. In this method, the
researcher specifies the minimum number of sampled units he/she wants in each category. Here,
the researcher is not concerned with having numbers that match the proportions in the
population. Instead, he/she simply want to have enough to assure that he/she will be able to talk
about even small groups in the population. This method is the non-probabilistic analogue of
stratified random sampling in that it is typically used to assure that smaller groups are adequately
represented in your sample (Stuart, 2004).
(iv) Heterogeneity Sampling: We sample for heterogeneity when we want to include all
opinions or views, and we aren't concerned about representing these views proportionately
(Bartlett, Kotrlik & Higgins, 2001). Another term for this is sampling for diversity. In many
brainstorming or nominal group processes (including concept mapping), the researcher would
use some form of heterogeneity sampling because the primary interest is in getting broad
 Federal University Dutsin-Ma Journal of Educational Research (Fudjer), 2 (1): 47-56.

spectrum of ideas, not identifying the "average" or "modal instance" ones. In effect, what the
researcher would like to be sampling is not people, but ideas.
(v) Snowball Sampling: In snowball sampling, the researcher begins by identifying someone
who meets the criteria for inclusion in his/her study (Neyman, 2008). The researcher asks them
to recommend others who they may know who also meet the criteria. Although this method
would hardly lead to representative samples, there are times when it may be the best method
available. Snowball sampling is especially useful when one is trying to reach populations that are
inaccessible or hard to find. For instance, if the researcher is studying the homeless, one is not
likely to be able to find good lists of homeless people within a specific geographical area.
However, if one goes to that area and identify one or two, the researcher may find that they know
very well who the other homeless people in their vicinity are and how one can find them.
(vi) Accidental Sampling: This can be called availability sampling (Scott, 2007). This is
because it makes use of the respondents available at the time. Participation is based on
availability. This is very common with pressmen or journalists. They interview or use anybody
available at the time.
Advantages of non-probability sampling technique:
a) The techniques need less effort. Unlike probability sampling techniques, especially
stratified random sampling, quota sampling is much quicker and easier to carry out
because it does not require a sampling frame and the strict use of random sampling
techniques (i.e. probability sampling techniques) (Stuart, 2004).
b) These techniques need less time to finish up.
c) They are not much costly.
d) In quota sampling, when the respondent refuses to cooperate, he may be replaced by
another person who is ready to furnish information.

Disadvantages of non-probability sampling technique:

a) The sampling techniques are prone to encounter with systematic errors and sampling
biases. The sample size determination becomes very hard since there are no formulas like
in case of random sampling methods. Moreover, the interviewer interviews people who
are easily available and accessible. So, the possibility of collecting valuable data is
affected in Quota sampling.
 Federal University Dutsin-Ma Journal of Educational Research (Fudjer), 2 (1): 47-56.

b) The sample cannot be claimed to be a good representative of the population. Difficulties

in selecting a truly representative sample produces reliable and accurate results only
when they are representative of the whole group. Selection of a truly representative
sample is difficult when the phenomena under study are of a complex nature. Selecting
good samples is difficult (Stuart, 2004).
c) Inferences drawn from sample are not generalisable to the population. There is a potential
for selection bias, which can result in a sample that is unrepresentative of the population.
d) Quota sampling method requires several investigators. Each one cannot be equally
competent. So, the results derived from the study may not be uniform.

References

Anaekwe, M.C. (2002). Basic Research Methods and Statistics in Education and Social
Sciences. Enugu: Podiks Printing and Publishing Company.
Bartlett, j. E., Kotrlik, j. W. & Higgins, c. C. 2001. Organizational Research: Determining
Appropriate Sample Size in Survey Research. Learning and Performance Journal, 19, 43-
50.
Biemer, P.P.L. & Lyberg, K. (2003). Introduction to survey quality. New York: John Wiley.
Cochran, W. G., (1993). Sampling Techniques. Second Edition. John Wiley & Sons, Inc. New
York. Library of Congress Catalog Card Number: 63-7553
Cohen, J. (2007). Statistical Power Analysis for the Behavioral Sciences. New York: Academic
Press.
Cohen, J. (2008). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ:
Lawrence Erlbaum Associates, Inc.
Dillman, D.A. (2000). Mail and internet surveys: the tailored design method. Brisbane: Wiley.
Fox, N.; Hunn, A. & Mathers, N. (2007). Sampling and Sample Size Calculation. The NIHR
RDS for the East Midlands / Yorkshire & the Humber.
Krieger, N. (2012). Who and What Is a “Population”? Historical Debates, Current Controversies,
and Implications for Understanding “Population Health” and Rectifying Health Inequities.
The Milbank Quarterly: A multidisciplinary journal of population Health and Health
Policy, 90 (4): 634–681.
Meyer, J.T. (1979). Fundamental Research Statistics for the Behavioural Sciences. New York:
Holt Rinehart & Winston.
Neyman, J. (2008). Contribution to the Theory of Sampling Human Populations. American
Statistics Association., 33, p101-116.
Noordzig, M.; Tripepi, G.; Dekker, F. W.; Zoccali, C.; Tanck, M. W.; Jager, J. K. (2010).
Sample size calculations: basic principles and common pitfalls. Nephrol Dial Transplant,
25: 1388–1393.
Osuala, E. C. (2007). Introduction to research methodology (3rd ed.). Onitsha: African – First
Publishers Ltd.
Raj, D. (2008). Sampling Theory. Mc-Graw-Hill Book Company. New York. 2008
 Federal University Dutsin-Ma Journal of Educational Research (Fudjer), 2 (1): 47-56.

Ramalu, S. (2010). Relationships between Cultural Intelligence, Personality, Cross-Cultural

Adjustment and Job Performance amongst Expatriates in Malaysia. In: University Putra
Malaysia.
Salant, P. A. & Dillman, D. A. (2004). How to conduct your own Survey. John Wiley & Sons,
Inc. New York.
Salkind, N. J. T. (2010). Exploring research. Prentice Hall Upper Saddle River, NJ.
Scott, M. S. (2007). Determining sample size: How to ensure you get the correct sample size.
Qualtrics, 4 (3): 12-23.
Sekaran, U. (2013). Research methods for business: A skill building approach. 4 ed., New
Jersey: John Wiley and Sons.
Stuart, M. (2004). Issues of Sampling Techniques. Nariobi: Adex publishers.
Sukhatme, P. V. (2008). Sampling Theory of Surveys with Applications. Indian Society of
Agricultural Statistics, New Delhi, India and the Iowa State University Press, Ames, Iowa,
U.S.A.
Taherdoost, H. (2017). Determining Sample Size; How to Calculate Survey Sample Size.
International Journal of Economics and Management Systems, 4 (2): 140-156.
Ticehurst, G. (2009). Business research methods: A managerial approach. NSW Australia:
Addison Wesley Longman.
Trochim, M. W. (2010). http://trochim.human.cornell.edu/kb/sampnon.htm Bill Trochim’s
Center for Social Research Methods, Cornell University
Wimmer, R. D. & Dominick, J. R. (2006). Mass media research: An introduction (8th
ed.).Wadsworth: Holly J. Allen.
Writes, M. (2012). Sample size calculations for trials in health services research. Journal of
Health Series and Policy, 4 (4): 226 – 229.