CHAPTER –FOUR

SAMPLING DESIGN

4.0. Meaning of Research Design

Research design is the decisions regarding what, where, when, how much, by what means
concerning a research project. It is the arrangement of conditions for collection and analysis of data
in a manner that aims to combine relevance to the research purpose with economy in procedure.
In fact, the research design is the conceptual structure within which research is conducted; it constitutes
the blueprint for the collection, measurement and analysis of data. As such the design includes an
outline of what the researcher will do from writing the hypothesis and its operational implications to
the final analysis of data. More explicitly, the design decisions happen to be in respect of:
 What is the study about?
 Why is the study being made?
 Where will the study be carried out?
 What type of data is required?
 Where can the required data are found?
 What periods of time will the study include?
 What will be the sample design?
 What techniques of data collection will be used?
 How will the data be analyzed?
 In what style will the report be prepared?
The Overall research design may split into the following parts.
 Sampling Design which deals with the method of selecting items It be observed for the
given study;
 Observational design which relates to the conditions under which the observations are to
be made;
 Statistical design which concerns with the question of how many items are to be observed
and how the information and data gathered are to be analyzed;
 Operational design which deals with the techniques by which the procedures specified in
the sampling, statistical and observational designs can be carried out.
4.1 SAMPLING DESIGN
Sampling is the process of using a small or parts of a larger population to make conclusions about the
whole population. It is one of the components of research design. Sampling design involves several basic
questions:
 Should a sample be taken?

RESEARCH METHODS IN ACCOUNTING AND FINANCE BY: OLANI BEKELE PAGE 1

  If so, what process should be followed?
 What kind of sample should be taken?
 How large it should be?
 What can be done to control and adjust for non response errors?

Basic sampling and statistical terminology

The most common concepts and terminologies in sampling are the following:
 Population (universe): the totality of the element to be studied.
 Census: is study of the whole population & it involves complete enumeration of the
whole population.
 Element: unit about which information is collected & which provides the basis of
analysis.
 Sample: a small proportion of a population selected for observation and analysis from a
population in accordance with specified procedures.
 Sampling frame: the actual list of sampling units from which the sample is selected.
 Sample size: the number of sampling units selected for observation and analysis.
 Statistics: The summary description of a given variable in the population.
 Variable: a set of mutually exclusive characteristics such as sex, age, income, etc.

4.2. Census Versus survey

The data which is need for a research could be obtained from primary and secondary sources.
There are two methods of study in using primary data for research:
 Census
 Survey
1. Census: is collecting all the necessary data from the whole population under study.
Advantages of census method
 Completeness: data are obtained from each and every unit of population in the study
area.
 Originality: all the target groups are participated in the study.
 Precision (accuracy) & reliability: the results obtained are likely to be more
representative, accurate, & consistent.

RESEARCH METHODS IN ACCOUNTING AND FINANCE BY: OLANI BEKELE PAGE 2

  Less sampling error: - there is no subjective judgment in selection of some study units
from the entire population.
 Wide applicability: - the data collected has a wide application in the country as a whole.
Disadvantage of census method
 Huge resources (HR, financial, time resources) requirements
 Time constraints: requires long period of time
 Impossibility of checking biasness,
 Not appropriate for short term study.
2. Survey: is a research design in which information is gathered from a sample of respondents.
Assumptions or rationales for survey:
 Homogeneity: - survey is based on the assumption that groups have certain essential
common traits. There is no basic difference in the nature of the units of the universe.
 Representativeness: - Because of the law of statistical regularity the groups chosen
through sampling from the whole population have the likely of the characteristics of the
population.
 Relativity: The results of sampling method may not be 100% accurate but they are
sufficiently accurate
 Adequacy: the size of the sample should be adequate enough; otherwise it may not
represent the characteristics of the universe.
Basic survey designs
Survey may be cross sectional and longitudinal designs
 In cross sectional survey, data are collected at one point in time from a sample selected
to describe some larger population at that time.
 Longitudinal studies are survey of respondents at different points in time. In this
method respondents are asked in d/t moments of times.
Errors in survey research
1. Random sampling error: the difference b/n the result of a sample and the result of a
census conducted using identical procedures.
2. Systematic error: error resulting from imperfect aspect of the research design that causes
response error or from a mistake on the execution of the research.
 Systematic error may be respondent & Administrative error.

RESEARCH METHODS IN ACCOUNTING AND FINANCE BY: OLANI BEKELE PAGE 3

 I. Respondent error: are of two types
A. Non response error: The statistical difference b/n a survey that includes only those who
responded and a survey that also includes those who failed to respond. Non response error
occurs because of:
 Not- at- home
 Refusal
 Self selection bias
B. Response bias: error that occurs when respondents tend to answer in a certain direction:
consciously or unconsciously misrepresent the truth. Response bias may:
 Acquiescence bias: When individuals have tendency to agree with all questions or to
indicate a positive connotation
 Extremity bias: results from response style varying from person to person.
 Interviewer bias: Bias in the responses of subjects due to the influence of the
interviewer.
 Auspices Bias: Bias in the responses of subject caused by respondent being influenced
by the organization the study.
II. Administrative error: is an error caused by improper administration of the research task.
 Data processing error
 Sample section error
 Interviewer error
 Interviewer cheating
Advantages of survey study design
 Viability: the results that the researches find out from the study can be obtained in a
short period of time.
 Thoroughly: the area of the study is small.
 Administrative convenience
 Practicability
 Possibility of checking biasness
Disadvantages of survey study design
 Possibility of bias and prejudices
 Difficulty in selecting representative
 Need for specific and specialized knowledge

RESEARCH METHODS IN ACCOUNTING AND FINANCE BY: OLANI BEKELE PAGE 4

4.3 THE NEED FOR SAMPLING

Sample method has some merits over the census method. That are:
1- Sampling saves time and money.
2- It is much convenient as it involves less personal staff.
3- It is useful when population is infinitely large.
4- It can be more accurately supervised and data can be carefully selected.
5- It is useful in case of inspecting the quality of units, which we have to resort to sampling, such as
testing the quality of bulbs, tubes, strength of stencils, testing explosives, etc.

4.4. TYPE OF SAMPLE DESIGN

Sampling method can be classified in to two broad categories:
 Probability (random)
 Non probability ( non random) sampling
1. Probability sampling Method: is a sampling method in which every member of the
population will have a known none zero probability of selection.
Advantages of probability sampling
 It avoids conscious or unconscious biases in the selection of sampling units.
 It permits estimation of sampling error
 It is possibility to evaluate relative efficiency
Disadvantages of probability sampling
o It requires high level of skill & experience for its use
o It requires time for planning & executing
o Costly as compared to non random sampling.
Types of probability sampling
There are various types of probability sampling techniques: the most well known are the
following:
 Sample random sampling
 Stratified sampling
 Systematic sampling
 Cluster (multi-stage) sampling.
1. Simple random (unrestricted) sampling: In this method each unit/individual in the
population has an equal or known opportunity of being selected. It is more satiable in
more homogeneous and comparatively large groups. The sample is chosen by some
method like flipping a coin (head or tail selection), lottery, or table of random number.

RESEARCH METHODS IN ACCOUNTING AND FINANCE BY: OLANI BEKELE PAGE 5

 The simple random sampling takes the form of N cn
N = is the number of population in the sampling frame
n= is the number of population in the sample
Ncn=is the number of combination (subsets of n from N)
F= n/N= is the sampling fraction.
Example: Assume that we take a certain finite population consisting of six elements say (a, b, c,
d, e, f) that is number and suppose that we want to take a sample size n=3 from it. Then there
are 6c3=20 possible distinct samples of the required size and it consisting of the elements, abc,
abd, abe, abf, acd, ace acf, ade, adf, aef, bcd, bce, bcf, bde, bdf, bef, ade, cdf, cef, and def. if we
choose one of these samples in such a way that each has the probability 1/20 of being chosen.
Advantages of simple random sampling
 Free from subjectivity
 Simplicity
 More representative
 Easily assess accuracy
Disadvantage of simple random sampling
 It requires complete lists of the universe
 It is not suitable for respondents (dispersed) over a large geographic area.
 For a given degree of accuracy, this method requires larger sample as compared to
stratified sampling.
 Lack of control of the investigator.
2. Stratified sampling
If the population is heterogeneous with respect to the variable or characteristics under study
then the techniques of stratified sampling is used. It involves dividing the target population in
to smaller homogeneous subgroups & then taking a sample random sample in each sub group.
Assume N is the heterogeneous population divide the population none over lapping
groups ( strata) N, N2, N3 ……….Ni) Using certain classification criteria such as religion,
age…… group, income, geography, race, language and such forth ( thus, N1 + N2 + N3+
……….+Ni=N). tends a sample of F=n/N in each strata.
Example: suppose that we want a sample of size n=30 to be drawn from a population of
size N= 8000 and the population includes Tigrians, Amharas, & Oromos. Assume again

RESEARCH METHODS IN ACCOUNTING AND FINANCE BY: OLANI BEKELE PAGE 6

 N (Oromos) =4000, N2 (Amharas) = 2400 and N3 (Tigrians) = 1600, by adopting
proportional allocation we shall get the sample size as follows:
For strata with N1 (Oromos) = 4000, we have p1=4000/8000
n1=n.p1=30 (4000/8000)=15
for strata with N2(Amharas) =2400
n 2=n.p2=30(2400/8000)=9
for strata with N3 = (Tigrians) =1600
n3=n.p3=30 (1600/8000)=6
Then by using simple random we can select the sample size as 15 from Oromo’s, 9 from
Amharas, & 6 from Tigrians.
Advantages of stratified sampling method
 If a correct stratification has been made, eve a small number of units will from a
representative sample
 It is more accurate and avoids bias to great extent
 It enables to achieve different degree of accuracy for d/t segment of the population
 Replacement of case is easy if the original case is not accessible to study.
 It enables d/t research methods & procedures to be used in d/t strata.
 Greater control of t investigator.
Disadvantages of stratified sampling
 It is very difficult to divide the universe in to homogeneous strata
 It the strata are over-lapping or disproportionate, the selection of samples may not be
representative.
3. Systematic sampling
Under this method a sample is taken from a list prepared on a systematic arrangement
either on the basis of alphabetic order, geographical, numerical or on house number or any
other method. In this method only the first sample unit is selected at random & the remaining
units are automatically selected in a definite sequence at equal spacing from one another.
Steps in systematic sampling
1. arrange population in serial number from 1 to n & determine sample size
2. determine the sample interval by dividing the population by sample size as follows:
N/n=k where k=sample interval
N= population size

RESEARCH METHODS IN ACCOUNTING AND FINANCE BY: OLANI BEKELE PAGE 7

 N= sample size
3. Select any number at random from the first sampling interval
Example: Assume total population = (N) =100
Samples size (n) =20
Then K=N/n=100/20 =5
Now select randomly number from the first interval that is 1 to 5, assume we chose 4
then to select the sample start with 4th unit in the list & take every kth unit (every 5th,
because k=5). That is 4, 9, 14, 19……. & so on to 100 & you wind up with 20 units.
Advantages of systematic sampling
 It is very easy to operate & checking can also be & done quickly.
 It results in representative sample because of its randomness and probability
features.
Disadvantages of systematic sampling
 It works well only if the completed & up-to date frame is available & if the units are
randomly arranged
 Any hidden periodicity in the list will adversely affect the representative of the sample.
4. Cluster (multi-stage) sampling
In this method to select the sample the population is divided into groups (clusters) with
some group randomly selected for study.
The difference b/n stratified and cluster sampling is that, in stratified all elements in a group
have the same characteristics (homogenous), where as in cluster each element in cluster has d/t
or heterogonous characteristic.
Advantages of cluster sampling
 it is more flexible
 it gives significant cost gain
 it is easier & more practical method, which facilitates the field work
Disadvantage of cluster sampling
 less accurate
 It is difficult if units in a cluster are very heterogonous.

B. Non probability Sampling Method

RESEARCH METHODS IN ACCOUNTING AND FINANCE BY: OLANI BEKELE PAGE 8

Non probability sampling is the process of sample selection based on personal judgment and
knowledge. Under this method a desired number of sample units are selected deliberately or
purposely depending up on the object of the inquiry so that only the important items
representing the true characteristics of the population are including in the sample.
Advantages of non- probability sampling
 Requires lower cost
 Requires less time
 Acceptable if level of accuracy of the research results is not utmost importance.
Disadvantages of none—probability sampling
 The research results can’t be projected generalized to the total population.
 The researcher doesn’t know the degree to which the sample is representative of the
population from which is drawn.
 Sampling error can’t be estimated.
Types of non-probability sampling techniques
 Judgment sampling
 Convenience sampling
 Quota sampling
 Purposive sampling
 Snowball sampling
1. Judgment Sampling Technique - the researcher’s judgment is used for selecting items
which s/he considers as representative of the population.
2. Quota Sampling Technique: quotas are setup according to some specified
characteristics/ principles criteria such as age, occupation, political, gender, religion ,
education, race, etc, proportionally or non proportionally.
3. Convenience Sampling Technique: in this a sample is selected according to the
convenience of the investigator. This convenience of the investigator depends up on the
availability of the sources list, accessibility to the unit resources at the command of the
investigator.
The selection is unsystematic careless, accidental, thus this method applied under the
following circumstances
 Lack of clarity about the universe.

RESEARCH METHODS IN ACCOUNTING AND FINANCE BY: OLANI BEKELE PAGE 9

  No clarity about sampling
 Non-availability of source list.
4. Snowball Sampling Technique: This method begins with a few people and then
gradually increases the sample size as new contacts are mentioned by the people you
stated out with.
5. Purposive Sampling Technique: in this method the investigator purposively selects
certain units for study. It is characterized by
 Representative character
 Specific objective
 Freedom from bias
 Complete idea of the nature of the universe & various measurements.
CHARACTERISTICS OF GOOD SAMPLE DESIGN
 It must be result in truly representative sample
 It must be result in small sampling error
 It must be viable in the context of funds available for the research study.
 It must be a systematic bias can be controlled in a better way.
CHARACTERISTICS OF GOOD SAMPLE SIZE
 The large the size of the universes, the bigger should the samples size
 If the resources available are vast a lager sample size could be taken.
 The greater the degree of accuracy desired the larger should be the samples size.
 If universe consists of homogeneous units a small sample may serve the purpose & Vic
verse.
 For intensive & continuous study a small sample may be suitable.

RESEARCH METHODS IN ACCOUNTING AND FINANCE BY: OLANI BEKELE PAGE 10