You are on page 1of 7

Name : Septi Rahayu

Nim : 1711230034

Class : PBI 6 B

Chapter 5

Population and Sample

A. population
Population is a unit of individuals or subjects in a region and time with certain
qualities that will be observed / examined.
The study population can be divided into "finite" and "infinite" populations.
Population is a population whose exact population is known, while an infinite population
is a population whose population members cannot be known with certainty. In the
research entitled PTS University Student Living Costs, for example, the number of PTS
University students can be known with certainty, hence it is called finite population.
While there is a study entitled "Analysis of Consumer Opinions and Attitudes towards
Cheap Earth Shop Services", it will appear that the study population is all consumers of
the Cheap Earth Store. While the number of consumers of the Cheap Earth Store data is
unknown, the population number cannot be known with certainty. Because this kind of
population is called an infinite population.

B. sample
By examining all students and / or all consumers of the store (a later example), the
researcher must certainly provide time, energy from a large cost. Therefore, researchers
in their research activities are justified in using only samples that can represent the study
population.
Thus the research sample is part of the population that is the subject of research as
a "representative" of the members of the population. As an example of the first research
title, not all students are studied (subject = respondent = data source), but some can be
taken to represent it. Likewise with the second example, which means that not all
consumers become research respondents, but some consumers are able to represent it.
This kind of research is called sampling research.

C. Sampling technique
Sampling technique is a sampling technique. To determine the sample that will be
used in research, there are various sampling techniques that are used. In qualitative and
quantitative research and R&D, Prof. Dr. Sugiyono (2010) states Schematically, sampling
techniques can basically be grouped into two namely Probability Sampling and Non-
probability Sampling. Probability Sampling includes: simple random, proportionate
stratified random, disproportionate stratifed random, and random areas. Nonprobability
sampling includes: systematic sampling, quota sampling, accidental sampling, purposive
sampling, saturated sampling, and snowball sampling.

1. Probability sampling
Probability sampling is a sampling technique that provides equal opportunities for
each element (member) of the population to be selected as a sample member. These
techniques include:
a. Simple random sampling
Simple random sampling is taking members of the sample from a random
population without regard to strata in that population. This method is used when
members of the population are considered homogeneous.

b. Proportionate stratified random sampling


This technique is used if the population has members / elements that are not
homogeneous and proportionally stratified. An organization that has employees from
stratified educational backgrounds, the population of employees is stratified. For
example the number of employees graduating from S1 = 45 people, S2 = 30 people,
SMK = 800 people, high school = 400 people, junior high school = 300 people,
elementary school = 300 people. The number of samples that must be taken includes
the educational strata.

c. Disproportionate stratified random sampling


This technique is used to determine the number of samples, if the population is
stratified but not proportional. For example, employees from certain work units have
3 doctoral graduates, 4 graduate students, 90 undergraduate people, 800 high school
students, 700 junior high school students. Then the 3 S3 graduates and 4 S2 graduates
were all taken as samples. Because these two groups are too small when compared to
the S1, SMU, and SMP groups.

d. Cluster sampling (Area sampling)


The area sampling technique is used to determine the sample if the object to be
studied or the data source is very broad, for example residents of a country, province,
or district. To determine which population will be the source of data, the sampling is
based on a predetermined population area.
For example, in Indonesia there are 30 provinces, and the sample will use 15
provinces, then the taking of 15 provinces is carried out randomly. But keep in mind,
because the provinces in Indonesia are stratified (not the same), the sampling needs to
use stratified random sampling. Provinces in Indonesia have dense population, some
do not, some have many forests, some do not, some have rich minerals and some do
not. This kind of character needs to be considered so that sampling according to the
population strata can be determined.
This area sampling technique is often used through two stages, namely the first
stage determines the regional sample, and the next stage determines the people in the
area by sampling too.

2. Nonprobability sampling
Nonprobability sampling is a sampling technique that does not provide an
opportunity / opportunity for each element or member of the population to be selected
as a sample.
a. Systematic sampling
Systematic sampling is a sampling technique based on the order of population
members that have been given a sequence number. For example members of the
population consisting of 100 people. Of all the members given serial numbers,
namely numbers 1 through number 100. Sampling can be done with only odd
numbers, or multiples of certain numbers, for example multiples of numbers 5, for
that then taken as samples are numbers1, 5, 10 , 15, 20, and so on up to 100.

b. Quota sampling
Quota sampling is a technique of determining a sample of a population that has
certain characteristics to the desired amount (quota). For example, it will conduct
research on community opinion on community service in the matter of Building
Permit. The number of samples determined by 500 people. If the data collection is not
based on 500 people, the research is considered incomplete, because it has not met the
specified quota.
If data collection is done in groups of 5 data collectors, each group member must
be able to contact 100 sample members, or 5 people must be able to find data from
500 sample members.

c. Incidental sampling
Incidental sampling is a technique for determining samples based on coincidence,
ie anyone who accidentally met with the researcher, the results of the data can be used
as a sample, if viewed by people who happen to be found suitable as a source of data.

d. Purposive sampling
Purposive sampling is a sampling technique with certain considerations. For
example, will research on the quality of food, the sample data source is a food expert.
Or research on political conditions in an area, then the sample data source is a
political expert. This sample is more suitable for qualitative research or studies that
do not make generalizations.
e. Saturated sampling
Saturated sampling is a sampling technique when all members of a population are
used as a sample. This is often done when the population is relatively small, less than
30 people, or research that wants to make generalizations with very small errors.
Another term for saturated samples is the census, where all members of the
population are sampled.

f. Snowball sampling
Snowball sampling is a technique for determining a sample that is first small in
number, then enlarges. Like a snowball that rolled a long time to become big. In
determining the sample, one or two people are first selected, but because these two
people do not feel complete the data provided, the researchers look for others who are
considered to know better and can complete the data provided by the two previous
people. And so on, so that the number of samples more and more.
In many qualitative studies using purposive and snowball samples, for example, it
will examine who the riot provocateurs are, then it will be appropriate to use
purposive and snowball.

D. Determine the sample size


To determine the sample of the population used calculations and reference tables
developed by experts. In general, for correlational studies the minimum number of
samples to obtain good results is 30, whereas in experimental studies the minimum
number of samples is 15 from each group and for survey research the minimum sample
size is 100.
Roscoe (1975) quoted by Uma Sekaran (2006) provides a general reference for
determining sample sizes:
1. A sample size of more than 30 and less than 500 is appropriate for most studies
2. If the sample is broken down into subsamples (male / female, junior / senior, etc.), a
minimum sample size of 30 for each category is appropriate
3. In mutivariate studies (including multiple regression analysis), the sample size
should be 10x larger than the number of variables in the study
4. For simple experimental studies with strict experimental control, successful research
is possible with small sample sizes between 10 and 20.

The size or size of this sample is very dependent on the amount of accuracy or
error desired by the researcher. However, in terms of error rates, in social studies the
maximum error rate is 5% (0.05). The greater the error rate, the smaller the number of
samples. However, it is important to note that the greater the number of samples (closer
to the population), the smaller the chance of generalization errors and vice versa, the
smaller the number of samples (away from the population), the greater the chance of
generalization errors.
Some formulas for determining the number of samples include:

1. Slovin formula (in Riduwan, 2005: 65)

n = N / N (d) 2 + 1

n = sample; N = population; d = 95% precision value or sig. = 0.05.

For example, the population is 125, and the desired error rate is 5%, so the
number of samples used is:

N = 125/125 (0.05) 2 + 1 = 95.23, rounded up 95

2. Jacob Cohen's Formula (in Suharsimi Arikunto, 2010: 179)

N=L/F^2+u+1

Information :
N = sample size
F ^ 2 = Effect Size
u = Number of changes related in research
L = Power function of u, obtained from the table

Power (p) = 0.95 and Effect size (f ^ 2) = 0.1


Price of L table with t.s 1% power 0.95 and u = 5 is 19.76
then with the formula they will get the sample size
N = 19.76 / 0.1 + 5 + 1 = 203.6, rounded 203
3. Formulas based on Proportions or Tables of Isaac and Michael
The table for determining the number of samples from Isaac and Michael provides
convenience in determining the number of samples based on an error rate of 1%, 5%
and 10%. With this table, researchers can directly determine the sample size based on
the number of populations and the desired error rate.

E. Example Determining Stamp Size


Research will be conducted to determine the response of community groups to
services provided by certain local governments. The community group consists of 1000
people, which can be grouped by level of education, namely graduates S1 = 50,
Bachelor's Degree = 300, SMK = 500, SMP = 100, SD = 50 (stratified population).
If the population = 1000, the error is 5%, then the sample number is 258, because
the population is stratified, then the sample is also stratified. The stratum is determined
according to the level of education. Thus each sample for the level of education must be
proportional to the population. Based on calculations in the following way the number of
samples for the group S1 = 14, Bachelor (SM) = 83, SMK = 139, SMP = 14, and SD =
28.

S1 = 50/1000 X 258 = 13,90 = 12,9

SM = 300/1000 X 258 = 83,40 = 77,4

SMK = 500/1000 X 258 = 139,0 = 129

SMP = 100/1000 X 258 = 27,8 = 25,8

SD = 50/1000 X 258 = 13,91 = 12,9

Jumlah = 258

So the number of samples is 12.9 + 77.4 +129 + 25.8 + 12.9 + = 258. The number
of fractions can be rounded up, so the number of samples is 13 + 78 + 129 + 26 + 13 + 25
= 259.
In calculations that produce deposits (there are commas) it should be rounded up
so that the number of samples is 259. This is safer than less than 258.
Roscoe in the book Research Methonds For Business (1982: 253) provides suggestions
on sample sizes for research such as the following:
1. The size of the sample that is feasible in research is between 30 to 500.
2. If the sample is divided into categories (for example: male-female, public-private
employee, etc.), then the number of sample members per category is at least 30.
3. If in the study will do the analysis with Multivariate (correlation or multiple
regression for example). Then the number of sample members is at least 10 times the
number of variables studied. For example there are 5 research variables (independent
+ dependent), then the number of sample members = 10 X 5 = 50.
4. For a simple experimental study, which uses the experimental group and the control
group, the number of sample members is between 10 to 20.

F. How to Take a Stamp Member


Probability sampling is a sempling technique that provides equal opportunities for
members of the population to be selected as members of the sample. This method is often
called random sampling, or how to take samples randomly.
Random sampling can be done by random numbers, computer, or by lottery. If the draw
is done by lottery, then each member of the population is given a number first, according
to the number of population members.

REFERENCES
Sugiyono.(2012). metode penelitian pendidikan pendekatan kuantitatif,kualitatif dan
R&D. Bandung : Penerbit Alfabeta

Hendri.(2010). Populasi dan Sampel. Retrieved from


https://teorionline.wordpress.com/2010/01/24/populasi-dan-sampel/comment-
page-4/

Sulistyo, Basuki. (2010). Metode Penelitian, Jakarta : Penaku.

Supardi. (1993). Populasi Dan Sampel Penelitian. UNISIA. no.


17TAHUNXIIITRIWULANV , Retrieved from
http://journal.uii.ac.id/Unisia/article/download/5325/4958

You might also like