CHAPTER 13: SAMPLING

CHAPTER 13: SAMPLING

If the data is not collected from the people, events, or objects that can provide

the correct answer to solve the problem, the survey will be futile. The election process

the right of individuals, objects, or events as a representative for the entire population is called " samples ".

Population, Element, Sample, Sampling Unit, and Subject

Population

Population refers to the group as a whole, incident / event, or something

the researchers are interested to be investigated / researched. For example, if the CEO of a computer company

wants to know what types of advertising strategies employed by computer companies in

Silicon Valley, then all computer companies domiciled there will be

population.

Element

Element is a member of the population. For example, if the 1000 workers in a particular organization

The population of the study, every worker in it's element. Census

a calculation of all elements in the human population.

sample

The sample is the subject of the population. The sample consisted of a number of members chosen from

population. Some, but not all elements will form the sample population. Examples of

1000 workers, 200 workers were sampled from a study.

sampling Unit

The sampling unit is an element or set of elements that are available to be selected in a few

menyampel process stage.

Subject

The subject is a member of the sample, as the element is a member of

population. Suppose that 200 workers is a sample from a population. Each of the 200 workers of

The sample of a subject in a study.

Parameters

OMAR SHAZAKI dilaga (A311 11 259)

The characteristics of the population such as (population), (standard deviation of the population), and 2 (variance

population) called a parameter.

Reasons for Sampling

The reason for using a sample, rather than collecting data throughout pupolasi already

very clear. In a study involving hundreds and even thousands of elements, practically

impossible to collect data, test, or examine each element. This bus is done

but will cost, time, and energy. Research on a sample and not the entire

population elements sometimes also very likely to produce more reliable results.

Representativeness of Samples

Kebuutuhan to select the appropriate sample for the study was not excessive. We

know that rarely sample into the exact replica of the population where the sample was taken. Simply,

the sample does not reflect the exact population

Normality of Distributions

Attributes or characteristics of the population generally normal distribution. Simply, attributes such as height,

most people will be grouped average height. Only a fraction short

and tall. When the nature of the population does not overrepresented or underrepresented in the sample

it will obtain a representative sample.

The Sampling Process

Steps in taking the sample:

1. Limiting the population.

2. Determining the sample frame.

3. Determining the sample design.

4. Determine the appropriate sample size.

5. Carry out the sampling process.

probability Sampling

When elements of the population have a known chance of being selected as a subject in the sample,

then we choose the design of the sampling by means of probabilities. Sampling

how probabilities can be a simple random sampling or complicated.

Unrestricted or Simple Random Sampling

OMAR SHAZAKI dilaga (A311 11 259)

Each element of the population has a known and equal chance of being selected as a subject.

This sampling design has the least bias and give a generalization of the most

large. However, the sampling process can be cumbersome and expensive.

Complex Probability Sampling

The sampling procedure in this way, providing a viable alternative and sometimes

more efficient than the design is not limited. There are five ways to design the sampling method

The most common complex probability sampling is systematic, decision

random sample strata, cluster sampling, the sampling area, and twice

sampling.

Sampling nonprobability

In the sampling design nonprobabilitas way, the probability of elements in the population

being selected as a subject of unknown samples. This means that the findings of the study

of the samples can not be reliably generalized to the population.

Convenience Sampling

Pengoleksian information of members of the population that could be easy to prove.

purposive sampling

Limited to certain types of people who can provide the desired information, either

because they are the only ones who have it, or they confirm some

criteria set by the researchers. There are two types of models, namely judgment sampling and

quota sampling.

Examples of When Certain Sampling Design Would Be Appropriate

Simple Random Sampling: The findings are common in a population.

Stratified Random Sampling: Difference information required in the various strata

in the population.

Systematic Sampling: If the frame sampling large, and a list of elements easily

available in one place.

Cluster Sampling: Various types studied group at a time.

Area Sampling: The research objective is limited to specific regions or localities.

Double Sampling: Wanting more information at a small extra cost.

OMAR SHAZAKI dilaga (A311 11 259)

Convenience Sampling: Want to get information "quickly" to get a "taste"

on a phenomenon or variable interest.

Judgment Sampling: There was input specific information on the important topic of research.

Quota sampling: The involvement of all the groups studied, small groups

the number is not negligible.

Sampling in Cross-Cultural Research

In a cross-cultural studies, we must be sensitive to the issue of selection of samples that fit in

different countries. The nature and type of organization studied, whether the subject is derived from the area

rural or urban, and the type of sampling design used, all should

similar in the different countries to allow a true comparison.

Issues of Precision and Confidence in Determining the Sample Size

Precision

Accuracy refers to how close our estimate population characteristics

in fact. Usually we will estimate the population parameter is within a range,

based on the estimated sample.

confidence

Confidence shows how confident that we estimate will actually apply to

population.

Sample Data, Precision, and Confidence in Estimation

Accuracy and confidence is an important issue in sampling because when we

use sample data to draw conclusions about the population, we hope to

hit a target and determine the level of possible errors. Because the level of point estimates

does not provide a measure of the probability of error, we performed the assessment interval to

ensure a relatively accurate assessment of the population parameter.

Trade Off Between Confidence and Precision

Researchers need to consider at least four aspects when making decisions regarding

The sample size of the study: (1) How much rigor needed in assessing

the characteristics of the population studied. (2) How much confidence that is really needed. (3)

To what extent the variability of the characteristics of the population studied? (4) How do the cost benefit analysis

of increasing the sample size?

OMAR SHAZAKI dilaga (A311 11 259)

Sample Data and Hypothesis Testing

The sample data is not only used to estimate the population parameters, but also test

hypothesis about the population value. The procedure for this test combines information

the same as in the interval estimation, but the purpose behind these two methods is somewhat different.

Determining the Sample Size

The formula in determining the sample size Sx =

In the study, the theoretical framework has several variables of the study, and the question arises

how do we determine the sample size if all factors are included. Krejcie and Morgan

(1970) simplifies the sample size to draw up a table to make sure the model

a good decision.

Importance of Sampling Design and Sample Size

If the design of proper sampling is not used, a large sample size will not be

allow the findings to be generalized to the population. Therefore, the decision-making

samples should consider the sampling design and sample size.

Efficiency in Sampling

The efficiency of the sampling to the level of accuracy achieved when certain ( standard

error), the sample size can be reduced, or for a particular sample size (n), the level of accuracy

can be improved. Selection of the sampling design depends on the purpose of research, extensive,

and the nature of the desired efficiency.

Sampling as Related to Qualitative Studies

Sampling in qualitative study begins by limiting the target population. for engineering

samples, qualitative studies usually use nonprobability sampling where there is no help

statistics. Here used one type purposive sampling which are called theoritical sampling.

managerial Implications

Noting the sampling design and sample size helps managers understand

how the sampling method used by the researchers, also helped to facilitate

managers to know the cost implications of different design and stuff more. case this

help managers understand the risks taken in implementing the changes is based

research.

OMAR SHAZAKI dilaga (A311 11 259)