LECTURES 2

Populations and Samples

Sampling Methods, Parameters and Statistics,

Data Presentation

Populations and Samples, Sampling Methods, Parameters and

11/7/2024 1
Statistics, Data Presentation
POPULATION

Population or universe is the entire collection of measurements about which one

wishes to draw conclusions.

• Population is often very large, such as the number of Zooplankton in the ocean,

• but occasionally populations of interest may be relatively small, such as the

number of persons that have been Vice Chancellors of University of Benin.

Populations and Samples, Sampling Methods, Parameters and

11/7/2024 2
Statistics, Data Presentation
SAMPLE

It is the subset of all the measurements in the

population.

From the characteristics of samples one can draw

conclusions about the characteristics of the

populations from which the sample came.

Populations and Samples, Sampling Methods, Parameters and

11/7/2024 3
Statistics, Data Presentation
 SAMPLING METHODS

It is the process of selecting participants from the population.

 It is important that the sample we select represent the target population by making sure that the
subjects in it are similar to the other members of the target population.

 This is because we want to generalize from the sample to target population.

 The more representative the sample, the more confident the researcher can be that the results can be
generalized to the target population.

 One of the problems that can occur when selecting a sample from a target population is sampling bias.

 Sampling bias refers to situations where the sample does not reflect the characteristics of the target
population.
Populations and Samples, Sampling Methods, Parameters and
11/7/2024 4
Statistics, Data Presentation
Samples can be obtained from the populations in a number of ways; however to reach valid
conclusions about populations by induction from samples, it is important that samples are
obtained in a random fashion.

There are two types of sampling:

 non-probabilistic (non-random) sampling and

 probabilistic (random) sampling

 Only results obtained from probabilistic (random) sampling are generalizable to the
reference population.

 The results obtained from non-probabilistic (non-random) sampling cannot be

generalized, even if sample size is large.

Populations and Samples, Sampling Methods, Parameters and

11/7/2024 5
Statistics, Data Presentation
Non-probabilistic Sampling

Also known as deliberate or purposive sampling, it uses a subjective method of selecting units from a

universe, and is generally easy, quick, and economical.

When samples are selected from a population on the basis of ease of access, it can be called

convenience sampling (accidental or haphazard) – select who ever is available

It is based on straightforward sampling,

Purposive (judgement sampling) - select samples based on experience

Quota sampling techniques – gathering of data from individuals possessing similar characteristics.

 It is useful to perform preliminary studies, focus groups or follow-up studies.

Populations and Samples, Sampling Methods, Parameters and

11/7/2024 6
Statistics, Data Presentation
Probabilistic Sampling

To avoid sampling bias it is important that samples are obtained in a random

fashion.

To obtain a random sample from a population requires that each member of the
population has an equal and independent chance of being selected.
 That is, not only must each measurement in the population have an equal chance of
being chosen as a member of the sample,

 But the selection of any member of the population must in no way influence the
selection of any other member.

Populations and Samples, Sampling Methods, Parameters and

11/7/2024 7
Statistics, Data Presentation
 Why is Random sampling of extreme importance?

 It assures the researcher that the samples represent, to an acceptable extent, the

population from which they were drawn.

 Scientists use it to relate his findings about the sample to the entire population

from which it is drawn. In the absence of random sample, it is impractical to infer

about the population from sample data.

 It ensures that no systematic attempt has been made to bias the sample.

Populations and Samples, Sampling Methods, Parameters and

11/7/2024 8
Statistics, Data Presentation
 RANDOM SAMPLING METHODS

1. Simple Random Sampling (SRS)

2. Interval or Systematic Random Sampling

3. Stratified Random Sampling

4. Cluster Random Sampling

The one chosen will depend on a number of factors (such as time, money etc.).

Populations and Samples, Sampling Methods, Parameters and

11/7/2024 9
Statistics, Data Presentation
 Simple Random Sampling (SRS)

It is most achieved using a set of table of values known as random numbers. The most
important feature of the table is the lack of order or discernible organizational structure.

Another method of achieving random sampling is by writing the names or identification

numbers of all the members of a population which are then placed inside a container from
which the required number is selected.

The use of these two methods requires that we be able to list all the members of the
population. Quite often this is rather impracticable, at times impossible.
Advantages: sample will represent the target population and eliminate sampling bias,

Disadvantage: it is very difficult to achieve (i.e. time, effort and money).

Populations and Samples, Sampling Methods, Parameters and

11/7/2024 10
Statistics, Data Presentation
 Interval or Systematic Random Sampling

This method involves listing all the members of a population on a numerical interval
1, 2, 3, e.t.c.
 If a population is 100 and you want a sample of 25 subjects, divide 100 by 25 =
4. Therefore, select every 4th number, such as 3, 7, 11, 15 etc.

• The basis of randomness in this method is that the first subject in the sample is
selected randomly.

Advantage: Provides a representative sample,

Disadvantage: It is very difficult to achieve (i.e. time, effort and money).

Populations and Samples, Sampling Methods, Parameters and

11/7/2024 11
Statistics, Data Presentation
 Stratified Random Sampling

Where the differences among the members of a population subgroup are seen as factors that
might influence their responses, the preferred mode of action is to divide the population to
their respective homogeneous subgroup and then work out the proportions needed for
the sample to be representative.
 Then a number of individuals or elements are randomly selected from each homogeneous
subgroup to constitute the sample.

Advantage: The sample will be highly representative of the target population and
therefore we can generalize from the results obtained.

Disadvantage: Gathering such a sample would be extremely time consuming and difficult
to do.

Populations and Samples, Sampling Methods, Parameters and

11/7/2024 12
Statistics, Data Presentation
 Cluster Random Sampling
In cluster random sampling a researcher breaks up the population into “workable” segments in order to
reduce research cost in terms of time and money.

 The population or the geographical area is subdivided into clusters or properly defined segments using
simple, systematic or stratified random sampling techniques.

 The aim is to reduce the population or geographical spread to such a size as would permit realistic and
adequate representation of a defined population while greatly reducing the cost of the research.

Note that this is significantly different from the stratified sampling where the homogeneity of the various
subgroups constitutes the basis for segmenting the population.

Advantage: The sample will be highly representative of the target population and therefore we can generalize
from the results obtained.

Disadvantage: Gathering such a sample would be extremely time consuming and difficult to do.

Populations and Samples, Sampling Methods, Parameters and

11/7/2024 13
Statistics, Data Presentation
 SAMPLE SIZE
What constitute the appropriate size of a sample?
It is generally difficult to answer. The answer depends on the size of the population and the purpose of the
study. In addition three criteria usually will need to be specified to determine the appropriate sample size:

1. the level of precision (sometimes called sampling error)

2. the level of confidence or risk, and

3. the degree of variability in the attributes being measured (Miaoulis and Michener, 1976).

 Constraint brought about by time and money sometimes imposes serious limitations on selecting
adequate number of individual objects to be included in a sample.

 But these should not dictate the sample size. — There is no point in carrying out a study that is too
small, only to come up with results that are inconclusive.

Populations and Samples, Sampling Methods, Parameters and

11/7/2024 14
Statistics, Data Presentation
It is imperative that the researcher arrives at the “best” estimates by ensuring that large numbers
of samples are collected from the population being investigated as the larger the sample the closer
the statistics obtained will be to the parameter of interest.

 Several formulas have been devised for estimating the appropriate sample size with reference to
the size of any defined population.

 Yamane (1967:886) provides a simplified formula to calculate sample sizes.

 A 95% confidence level and P = 0.5 are assumed for the Equation.

Where n is the sample size, N is the population size, and e is the level of precision (5% or 0.05).

Populations and Samples, Sampling Methods, Parameters and

11/7/2024 15
Statistics, Data Presentation
 Population Sample
10 10
50 44
100 81
150 110
200 134
250 154
300 172
400 201
500 217
750 254
1,000 278
1,500 306
2,000 322
3,000 341
5,000 357
10,000 370
30,000 379
1,000,000 384

Populations and Samples, Sampling Methods, Parameters and

11/7/2024 16
Statistics, Data Presentation
Simple rule of thumb for determining the sample size
1. The larger the population size, the smaller the percentage of the population required to get a

representative sample

2. For smaller (N<100) population, there is no need in sampling. Survey the entire population.

3. If the population size is around 500, 50% should be sampled

4. If the population size is around 1500, 20% should be sampled

5. Beyond a certain point (N=5000), the population size is almost irrelevant and a sample size of

400 may be adequate.

Populations and Samples, Sampling Methods, Parameters and

11/7/2024 17
Statistics, Data Presentation
 Parameters and Statistics

These are measures that help to describe or characterize a population or

samples.
Example,
Measure of central tendency (Mean, mode, and median ).
Measure of dispersion (Range, standard deviation and standard error).

When a quantity describes or characterizes a population it is called a

parameter.

It is rare to be able to calculate parameters. However by random

sampling of populations, parameters can be estimated very well.
Populations and Samples, Sampling Methods, Parameters and
11/7/2024 18
Statistics, Data Presentation
When a quantity describes or characterizes a sample it is called a statistic.

An estimate of a population parameter is called a statistic.

It is biostatistical convention to represent

 population parameters by Greek letters (e.g Population mean = µ

(mu)) and

 Sample statistics by Latin letters (e.g sample mean = 𝑋ത ).

Although parameters are constant for a population, statistics will vary

from sample to sample for samples taken from the same population.

Populations and Samples, Sampling Methods, Parameters and

11/7/2024 19
Statistics, Data Presentation