CHAPTER ONE: DATA COLLECTION AND SAMPLING TECHNIQUES

1.1- INTRODUCTION:
The most important part of a researcher’s work is the collection (gathering) of data which will be
used in his field of studies. Let’s consider the case of a business lady who wants to starts up a
small petit shop in the neighborhood’s the first important aspect he/she will have to considered is
the number of consumers who are there in the neighborhood before she begins to find out what
they consumed the most. All this can be done only through the help of data collection which will
form the statistics of that neighborhood. Before proceeding to the various methods of collecting
data it is important for us to define certain concepts which are important in this chapter.

I. Statistics:
Statistics is the art or science that deals with the collection, analysis and interpretation of
numerical data. Generally, statistics is divided into two parts which are: descriptive statistics and
inductive or inferential statistics.

I-A. Descriptive statistics:

Descriptive statistics deal with the compilation and presentation of data in various forms such as
tables, graphs, diagrams etc. the objective of descriptive statistics is to display the information so
that it can be viewed from which conclusions and generalization can be made on.

I-B. Inferential or inductive statistics:

These are the mathematical tools and techniques that are used to analyze data and use the results
in other to make estimate about the predictions from which conclusion and generalizations can
be made by the population.

II. Data:
Definition 1: data is any raw fact collected from the field for which no work has been done on it.
It is also called unprocessed information. The peculiarity about data is that it is unprocessed,
consequently it may be false or true.

III. Information
It is data that has been processed from which conclusions and generalisation can be made upon.
Information is not raw and has passed through a process of verification and facts finding before it
is being publish. The company tax rate of Cameroon is 33% is an information because the
exactitude of the message can be verified and proven from the general tax code (GTC)

Page 1 of 9
1.2- SOURCES OF DATA:
Generally we have two sources of data: that is the primary and the secondary data. Data is
collected from a population which may be finite or infinite. By population we mean the set of all
items of subjects which is under consideration or which is under studies. When all the total
population is studied it is called a census, but if part of the population is surveyed it’s called
sampling.

I.A- Primary Data: these are data collected from the field for the very first time for which no
work has be done on it. They are collected and used for a specific purpose.

Advantages of primary data:

 The exact wanted information needed is obtained
 The information is obtain in its original form
Disadvantages of primary data

 It is costlier to obtain firsthand information

 It is time consuming
I.B – Secondary data: Secondary data are those data which are obtain from already existing
documents. The data is obtain from one source which has previously been obtain examples of
secondary data include the internet, journals, newspapers etc. the cost of obtaining primary data
is not as expensive as secondary data, the secondary data most at times is not really true.

Secondary data can either be internal or external. When the data is internal it comes from
sources within the organization such as: from the company’s sales journals, research reports in
the company etc. conversely, secondary external data refers to data obtain from already existing
documents out of the organization for example the internet.

1.3- Data Collection methods: there are a wide variety of tools a researcher can used in other
to collect data which may include one of the following methods.
i) Questionnaires:

A questionnaire is a list of prepared questions and answers to be fill by a respondence. The

questionnaires are filled and the respondent simply needs to answer from the right answer from
the choice that follows. A questionnaire may be open ended or closed ended. With the closed-
ended questionnaires, the respondent has to answer amongst alternative choices of answers
provided from the list of option provided by the interviewer, while the open-ended
questionnaires are one that the researcher has to fill the answer of the questions with no
alternative answers provided to the questions. The open-ended questionnaires is important than
Page 2 of 9
the closed ended questionnaires because the researchers have the opportunity to express his view
point in its entirety.

Advantages of Using a Questionnaire

- It covers a wide audience

- It reduces biasness
- It is not too costly and time consuming
- If answered the questions can be kept for future references
- There is the absent of biasness
Disadvantages of Using a Questionnaire

- It may require only short and simple questions

- It is difficult to work with people who are uneducated that cannot read or write
- There is low respondent rate since not all the questionnaires can be collected
- facetious answers from the respondent cannot be taking into consideration
Characteristics of a Good Questionnaires

- the questions should not be ambiguous that is it should not be open to more than one
meaning
- the questions must be easily understood
- The questions should be capable of having a precise answer: “yes or No” “true or false,
good, fairly good, or excellent.
- the questionnaires should not be too long
- the questions should not be personal or offensive
- the questionnaires should cover the exact objective of the research
ii) The Personal Interview

This is a one to one interpersonal conversation between the interviewer and the respondents.
Interview may be a one to one conversation or one to many conversation between the interviewer
and the respondent. During the interview process, questions are asked to the respondents while
feedback are being gotten for the purpose of collecting the data.

Advantages of using the interview in other to collect data

- A high respondent rate is achieved

- Difficult questions not understood by the respondent could be explain
- More information can be collected than using the questionnaire.

Page 3 of 9
 - the interview can be conducted even by the telephone
- the interviewer can identify facetious answers of the respondent
Disadvantages of using the interview as a method of data collection

- Bias may be introduced

- Respondent may be reluctant to give confidential information
- The interviewer may not stick to the selected sample of respondent
- An interviewer without adequate skills may obtain inaccurate information from the
respondent.
- Large sample may not be targeted because of high cost
- It is much time consuming since all the respondent musts be interviewed by the
interviewer.
iii) Direct observation method

This method involves sending observers to records the actual situation of the data while the
respondent is carrying on the work. This is the most common method of data collection method
in the world of commerce and business. One of the main advantages of this method is that the
information is gotten in its originality and without the respondent taking notes that he is being
observed

Advantages of this method

- It is fast and does not consumed much time

- The actual data is collected since the respondent may not know the researcher is
collecting information from him
Disadvantages of this method

- The observers need to be highly skilled and unbiased

- Observers will find it difficult to collect information about people’s attitude and taught
- The present observation does not tell us anything about the past or the future

1.4- Sampling
1.4.1- Definition of a Sampling: sampling is the statistical process of selecting a subset of the
population called a sample. The sampling process consist of several stages: the first stag is to
define the target population. A population in research work can be define as all the items or the
people and the characteristics that the researcher wants to study. The unit of analysis may be

Page 4 of 9
 household, customers, managers or any other entity that the researcher wants to draw inference
from.

The second steps in the sampling process is to choose the sample frame. This is an
assessable section of the target population from where a sample can be drawn.

The final steps in the sampling process is to choose the sample frame using well define
sample technique.

1.4.2- Definition of a sample:

A sample is a small representative subset unit of a population from where data are collected.

1.5- Sampling techniques:

Sampling can either be done using the probability techniques or with the non-probability
techniques

i) Probability sampling
It is a technique in which every unit in the population has a chance of being selected in the
sample. The different types of probability sampling techniques include

A. Random sampling:
With this method of sampling all the members of the population have an equal chance of being
selected in the sample. Example of a random sample will be a local street map, school map, a
school, birth or voters register, telephone directory. There are two method of random sampling:

 Simple random sampling:

Simple random sampling ensures that each member of the population has an opportunity of being
choosing in the sample. To conduct a simple random sampling the following procedures should
be followed:

- Assign each member of the population a code

- Select a starting role in the simple random table
- Select the members of the population who have chosen digits that correspond to the code
that was initially coded with the codes on the sample

 Stratified Random Sampling:

Strata means layers or levels. This method involve the division of a heterogeneous population
into strata’s or layers such as sex, age , income or social class etc. when using the stratified
Page 5 of 9
 random sampling, it is important to know the proportion of each stratum and take it into
consideration when selecting the sample.

B. quasi random Sampling:

Quasi means almost or nearly, this method entails the following method
i. systematic sampling
ii. multi- stage sampling

I- Systematic Sampling:
This method is also known as constant skip or sequence method of sampling. The first items of
the item are selected at random and the second items of the sample is selected at a regular
interval. That are every Nth terms of the population.

II- Multi- stage Sampling:

This method is design for designing and reducing the cost of surveying samples from very large
population or from where the population is spread over a relatively wide geographical area. This
method includes the following.
- Splitting the area into a number of regions
- Randomly selecting a small number in each of the region
- Using any random method to select the required size in the region.

C. Cluster Sampling:
This method is used to sample a population which is distributed over a wide geographical area.
The techniques involve selecting one or more geographical area and then sampling members
from the target population that can be identified.

D. Quota Sampling:
Quotas or a number of items from a different population are set and how items from the
population are selected is left completely in the hands of the interviewer. This method is popular
with market research.

E. Non-probability random sampling:

Here we have convenience sample, expert sample and snowball sample
i. Convenience sample:

Page 6 of 9
Also called opportunity or accidental sample. This is a technique in which a sample is drawn
from that part of the population which is closed at hand, readily available or convenient.

ii. Expert sample:

This is a technique where respondents are choosing in a non-random manner based on their
expertise on the phenomenon being studied. The advantage of this approach is that since expert
turns to be more familiar with the subject matter their opinion are deemed credible.
iii. Snow ball sample:
In snow ball sampling, the researcher starts by identifying a few respondents that match the
criterion of inclusion of the study, and ask them to recommend others they know that match the
selection criteria. This method may be inappropriate because the researcher will find it difficult
to define the sampling frame.

1.6- Summary of chapter

Data Collection Technique Sampling Technique

o Questionnaires: A A. Probability sampling technique
questionnaires is a list of  Simple random sampling: simple random sampling ensures that
prepared questions and each member of the population has an opportunity of being choosing
answers to be fill by a in the sample. To conduct a simple random sampling the following
respondents. The question is procedures should be followed:
in a questionnaire may be  Stratified Random Sampling: strata means layers or levels. This
closed or open ended method involve the division of a heterogeneous population into
strata’s or layers such as sex, age , income or social class etc. when
o Interviews: it is an using the stratified random sampling, it is important to know the
interpersonal conversation proportion of each stratum and take it into consideration when
between the interviewer and selecting the sample.
the respondent. Questions are
asked by the interviewer and  Systematic Sampling:
data is collected This method is also known as constant skip or sequence method of
sampling. The first items of the item is selected at random and the
o Observation: with
this second items of the sample is selected at a regular interval. That is
method researchers are send every Nth terms of the population.
on the field to watch and
collects information on the  Multi- stage Sampling:
respondent while he is on the This method is design for designing and reducing the cost of
Page 7 of 9
 surveying samples from very large population or from where the
field
population is spread over a relatively wide geographical area. This
method include the following.
other methods may include:

o focus groups  Cluster Sampling:

o ethnographies This method is used to sample a population which is distributed over a
o participant observations wide geographical area. The techniques involve selecting one or more
 Quantitative research geographical area and then sampling members from the target
o controlled laboratory population that can be identified.
experiments
o field work  Quota Sampling:

o surveys Quotas or a number of items from a different population are set and
how items from the population are selected is left completely in the
hands of the interviewer. This method is popular with market research.

B. Non-probability sampling techniques

 Simple random sampling:
Simple random sampling ensures that each member of the
population has an opportunity of being choosing in the sample. To
conduct a simple random sampling the following procedures should
be followed:
 Stratified Random Sampling:
Strata means layers or levels. This method involve the division of a
heterogeneous population into strata’s or layers such as sex, age ,
income or social class etc. when using the stratified random sampling,
it is important to know the proportion of each stratum and take it into
consideration when selecting the sample.

 Systematic Sampling:
This method is also known as constant skip or sequence method of
sampling. The first items of the item is selected at random and the
second items of the sample is selected at a regular interval. That is
every Nth terms of the population.
 Multi- stage Sampling:
Page 8 of 9
 This method is design for designing and reducing the cost of
surveying samples from very large population or from where the
population is spread over a relatively wide geographical area. This
method include the following.
 Cluster Sampling:
This method is used to sample a population which is distributed over a
wide geographical area. The techniques involve selecting one or more
geographical area and then sampling members from the target
population that can be identified.
 Quota Sampling:
Quotas or a number of items from a different population are set and
how items from the population are selected is left completely in the
hands of the interviewer. This method is popular with market research.

1.7- Review questions of chapter one

1. Define the term statistics and name the two major branches of statistics
2. Distinguish comprehensively between primary and secondary data
3. Give four reasons why a sample is preferable to the entire population in statistical survey.
4. What for you understand by the term sampling
5. Name five areas to be considered when drawing a questionnaire
6. What do you understand by the term sampling frame? What two methods in statistics you
know that do not require a sample frame
7. Distinguish clearly between the following
i. Sample and population
ii. Sampling and census
iii. Statistics and parameters
iv. Sampling errors and bias
v. Systematic and cluster sampling
8. Statistics will be lost without good sampling technique. Discuss the statement and
describe two methods of sampling that are commonly used

Page 9 of 9