Business Statistics
Chapter 2
Data Collection and Sampling
Mustafe Jiheeye MSc. in Research and Data Analysis Jiheeye9308@[Link] Business Statistics -I Chap 2-21
Introduction
In any statistical study, the first step is to collect a set of related
observations from which conclusions may be drawn. Based on the form of
characteristics of interest, a population could be classified into either
quantitative or qualitative.
COLLECTION OF DATA
Methods of
Sources of data collection of
data
Mustafe Jiheeye MSc. in Research and Data Analysis Jiheeye9308@[Link] Business Statistics -I Chap 2-22
Types of Data
According to the scale of measurement we can classify data as
nominal, ordinal, interval or ratio.
Nominal Data: are the types of data which reflects classification
characteristic (names of groups), but characteristics do not indicate any
mathematical difference.
Data are labels or names used to identify an attribute of the element.
A nonnumeric label or a numeric code may be used.
There are no implied criteria by which the data can be ordered from
smallest to largest
Example:
Gender (Male, Female), Ethnic group (black, white), marital status
(married, single, divorced)
Mustafe Jiheeye MSc. in Research and Data Analysis Jiheeye9308@[Link] Business Statistics -I Chap 2-23
Types of Data
Ordinal Data: Are defined as nominal data but the nominal classifications
can be placed in order.
It applies to data that can be arranged in order. However, differences between
data values either cannot be determined or are meaningless.
Example:
Ranking students in a class as first, second, third, fourth and so on
• Guest speaking may be ranked as superior
Ordinal scales can also use attribute labels such as “bad”, “medium”, and
“good”, or "strongly dissatisfied", "somewhat dissatisfied", "neutral", or
"somewhat satisfied", and "strongly satisfied”.
Example: Students of a university are classified by their class standing using a
nonnumeric label such as Freshman, Sophomore, Junior, or Senior.
Alternatively, a numeric code could be used for the class standing
variable (e.g. 1 denotes Freshman, 2 denotes Sophomore, and so on).
Mustafe Jiheeye MSc. in Research and Data Analysis Jiheeye9308@[Link] Business Statistics -I Chap 2-24
Types of Data
Interval Data: Interval data are defined as ordinal data which have
equality of units. Interval data are the types of information in which an
increase from one level to the next always reflects the same increase in
quality.
Example
Temperature: increasing 1 degree to 2 degree is the same as the
increasing from 40 to 41 degrees.
IQ test.
There is no true zero. For example, IQ tests do not measure people who
have no intelligence. For temperature, 0F does not mean no heat at all.
Mustafe Jiheeye MSc. in Research and Data Analysis Jiheeye9308@[Link] Business Statistics -I Chap 2-25
Types of Data
Ratio Data: Are data that can be arranged in order. In addition, both
differences between data values and ratios of data values are
meaningful. Data at the ratio level have a true zero.
All measurement data like height, weight, volume, and area are ratio
data.
Example:
Weight (one person can lift 200 pounds & another can lift 100 pounds, then
the ratio between them is 2 to 1)
Mustafe Jiheeye MSc. in Research and Data Analysis Jiheeye9308@[Link] Business Statistics -I Chap 2-26
Types of Data
According to the source, we classify data as primary and secondary.
Sources of data
Primary Secondary
data data
Primary Data: Data is said to be primary if they are obtained first hand
for the particular purpose on which is currently working.
Mustafe Jiheeye MSc. in Research and Data Analysis Jiheeye9308@[Link] Business Statistics -I Chap 2-27
Types of Data
Secondary Data: Data which were previously collected for a similar or
different purpose are known as secondary data.
Mustafe Jiheeye MSc. in Research and Data Analysis Jiheeye9308@[Link] Business Statistics -I Chap 2-28
Types of Data
According to the role of time, data are classified as either cross sectional or
time series data
Cross sectional data: Data collected on different elements at the same
point in time or for the same period of time are called cross-section data.
Is a set of observations taken at one point in time
Time series data: Data collected on the same element for the same variable
at different points in time or for different periods of time are called time-
series data.
Is a set of observations collected sequence of times, usually at equal
interval, which may be on weekly, monthly, quarterly, yearly etc.
Mustafe Jiheeye MSc. in Research and Data Analysis Jiheeye9308@[Link] Business Statistics -I Chap 2-29
Methods of Data Collection
The data collection methods vary according to the nature of the
investigation and availability of resources like money, time, skilled human
power.
We explained that there are two sources to data which are primary and
secondary. Accordingly, we have two methods of data collection, primary
and secondary methods.
Mustafe Jiheeye MSc. in Research and Data Analysis Jiheeye9308@[Link] Business Statistics -I Chap 2-30
Methods of Data Collection
Primary Method of Data Collection
Data which are in the hands of the investigators can be allocated by
using different methods. Primary data may be collected using one the
following methods.
I. Observation
This is a method of data collection in which the researcher observes the
behaviour of the individual in the set of observations, be it sample or
population.
This is the best method, as reduces the chance of incorrect data being
recorded, but since it costly, it cannot always be used.
Mustafe Jiheeye MSc. in Research and Data Analysis Jiheeye9308@[Link] Business Statistics -I Chap 2-31
Methods of Data Collection
II. Enumeration method
A well trained enumerators ask the selected group of respondents a set
of questions which should relevant to the study.
It is advisable to employ this method because it gives a face to face
contact with the respondent.
Mustafe Jiheeye MSc. in Research and Data Analysis Jiheeye9308@[Link] Business Statistics -I Chap 2-32
Methods of Data Collection
III. Interview
There are three different types of interviewing; the direct personal interview,
the telephone interview or the indirect oral interview methods
a. The direct personal interview method
This technique is the best suited when the problems are not completely
understood and the researcher presents him/her self before the informant
and question him/her carefully.
Since it is the researcher him self or well trained personnel that has to
make the interview, this method is difficult in developing countries and is
very costly.
Mustafe Jiheeye MSc. in Research and Data Analysis Jiheeye9308@[Link] Business Statistics -I Chap 2-33
Methods of Data Collection
b. Telephone interview
In this method, a set of questions will be prepared and then asked the
respondents using a telephone call. But, the questions should not be too
many so that the respondents may not get bored in a single call
Mustafe Jiheeye MSc. in Research and Data Analysis Jiheeye9308@[Link] Business Statistics -I Chap 2-34
Methods of Data Collection
c. Indirect oral interview
In this method, the researcher contacts third parties called witnesses
capable of supplying the necessary information.
This method is generally adopted in those cases where the information
to be obtained is of a complex nature and the information the
respondents are nor inclined to respond if approached directly
Mustafe Jiheeye MSc. in Research and Data Analysis Jiheeye9308@[Link] Business Statistics -I Chap 2-35
Methods of Data Collection
IV. Mailed Enquiry Method
This is a method of collecting data by sending letters which contain a set
of questions to the respondents and collecting the response afterward.
A list of questions pertaining to the survey (questionnaire) is prepared
and sent to the various respondents by post.
Mustafe Jiheeye MSc. in Research and Data Analysis Jiheeye9308@[Link] Business Statistics -I Chap 2-36
Methods of Data Collection
Secondary Method of Data Collection
Most of the studies researchers finds difficult to collect first hand
information on all related issues and s/he makes use of the data collected by
others.
Users of secondary data cannot have as thorough understanding of the
background as the original researcher, and so may be unaware of such
limitation.
Example:
Monthly imports and exports of a country
Monthly statistics of the production of industries
Annual sales of a company
Mustafe Jiheeye MSc. in Research and Data Analysis Jiheeye9308@[Link] Business Statistics -I Chap 2-37
Sampling Techniques
When secondary data are not available for the problem under the study,
primary data collection method is appropriate by using on the methods
discussed earlier.
The data can be obtained by using either census method or sample
survey method.
Sampling can grouped under two broach categories
1. Random (Probability) Sampling
2. Non-random (non-probability) Sampling
Mustafe Jiheeye MSc. in Research and Data Analysis Jiheeye9308@[Link] Business Statistics -I Chap 2-38
Classification of Sampling Techniques
Sampling
Probability Non-probability
Sampling Sampling
Simple Random Judgement Sampling
Quota Sampling
Stratified Sampling
Convenience Sampling
Systematic Sampling
Cluster Sampling
Mustafe Jiheeye MSc. in Research and Data Analysis Jiheeye9308@[Link] Business Statistics -I Chap 2-39
Sampling Techniques
1. Probability Sampling
In this sampling technique, every single observation of the sampled
population has non zero chance of being included in the sample.
The main purpose of sampling is to create a sample that is representative
of the population it is being drawn from hence it is very difficult to
survey the whole population
In this sub-section, there are five different techniques of taking a
random sample are discussed
Mustafe Jiheeye MSc. in Research and Data Analysis Jiheeye9308@[Link] Business Statistics -I Chap 2-40
Sampling Techniques
a. Simple Random Sampling
Is a random sample in which each and every observation in the
population (N) has the same probability (chance) of being selected for
your sample (n).
Simple random sample can be selected by making use of the Lottery
method or the table of random numbers.
Advantages:
As the sample size increase, it becomes representative of the population
There is no possibility of personal bias affecting the results
Limitations:
Difficult to have up-to-date list of all items of the population to be
sampled
Mustafe Jiheeye MSc. in Research and Data Analysis Jiheeye9308@[Link] Business Statistics -I Chap 2-41
Sampling Techniques
b. Systematic Random Sampling
Systematics random sampling is obtained by selecting one unit randomly
and then choosing additional units at equally spaced intervals until the
desired number of units has been obtained.
Example:
A researcher want to take a systematic random sampling of 50 students from
1000 students. The first step is list all students by giving a serial numbers and
1000
then take random from the first = 20 students.
50
The rest of the sample is obtained by taking every 20th student.
Suppose the first student was 7th, then S1 = 7th, S2 = 27th, … S50 = 987th
Mustafe Jiheeye MSc. in Research and Data Analysis Jiheeye9308@[Link] Business Statistics -I Chap 2-42
Sampling Techniques
c. Stratified Random Sampling
Divide the entire population into distinct subgroups called strata. The strata are
based on a specific characteristic, such as age, income, education level, and so on.
All members of a stratum share the specific characteristic. Draw random samples
from each stratum.
A sample which is obtained by selecting a separate simple random sample from
each stratum is known as a stratified random sample.
Illustration: If a researcher wants to deal with the income inequality situation in
Hargeisa city. The researcher can divide the households in to different groups.
As follows:
o Civil Servant
o Merchant
o Petty Traders & local drink sellers
Mustafe Jiheeye MSc. in Research and Data Analysis Jiheeye9308@[Link] Business Statistics -I Chap 2-43
Sampling Techniques
Procedures in selecting a stratified random sample
Select you base stratification
Decide the number of strata
Decide the sample size within each strata
Advantages:
More representative
Greater accuracy
Greater geographical concentration
Limitations:
Higher cost
Need skilled supervisor
Effective of bias
Mustafe Jiheeye MSc. in Research and Data Analysis Jiheeye9308@[Link] Business Statistics -I Chap 2-44
Sampling Techniques
d. Cluster Random Sampling
Divide the entire population into pre-existing segments or clusters. The clusters
are often geographic. Make a random selection of clusters. Include every member
of each selected cluster in the sample.
The main purpose of cluster sampling is to reduce cost.
Example:
In conducting a survey of schoolchildren in a large city, we could first randomly select
five schools and then we may take a cluster of random sample.
Illustration: Still taking the study of the income disparity condition in Hargeisa . In this
case, the Hargeisa city will be classified by locality (i.e., in to Northern, southern part of
Hargeisa , etc). Once the city is classified in to various clusters, randomly some of the
clusters (i.e., locality in our case) will be chosen and the researcher randomly select
elements from the chosen cluster.
Mustafe Jiheeye MSc. in Research and Data Analysis Jiheeye9308@[Link] Business Statistics -I Chap 2-45
Sampling Techniques
Non-probability Sampling
A sample is said to be non probability sampling if every person does not have
equal chance of being selected. Non-probability sampling can be divided into
3 different methods.
1. Judgement Sampling
A sample is said to be judgement sample when it is selected deliberately by
the researcher by taking some factors into consideration. This technique is
useful only when dealing with small populations.
Mustafe Jiheeye MSc. in Research and Data Analysis Jiheeye9308@[Link] Business Statistics -I Chap 2-46
Sampling Techniques
2. Quota Sampling
In this type of sampling technique, the population is subdivided into
heterogeneous parts and judgement sub samples are taken from each part in
percentage proportion of the parts in the population.
3. Convenience Sampling
A convenience sampling is simply uses results or data that are conveniently
and readily obtained.
Convenience sample is obtained by selecting convenient population units
Mustafe Jiheeye MSc. in Research and Data Analysis Jiheeye9308@[Link] Business Statistics -I Chap 2-47
Instructor: [Link]
Thank you for listening
Mustafe Jiheeye MSc. in Research and Data Analysis Jiheeye9308@[Link] Business Statistics -I Chap 2-48