Chapter SAMPLING

2
AND
SAMPLING DISTRIBUTION

A good survey research paper relies on the precision of the methods, and
procedures of conducting the study. This includes reliability of the selected subjects or
respondents of the study, the validity of information gathered out of the distributed
questionnaires, and the accuracy of measurements used in answering the research
questions and other observations. A study which were conducted in the entire
population assures us of 100% reliability since the responses are obtained from all
members of the population. This means that the data were collected by a complete
enumeration method or the so-called census taking. However, it is impossible for many
types of research to conduct a survey to all members of the population especially if the
population size is infinite or finite but very large. To minimize the time and cost involved
in conducting the survey to a large population, it has been accepted that the information
about the population will be based only from a small portion of the population.
On the other hand, considering only the responses of a small portion of the
population may result into some possible biases due to improper selection of the
samples and errors due to the manner of measuring the desired observations since the
selected sample may not have equally represented the characteristics of the entire
population. Hence, it is very important to consider the methods used in selecting the
sample.

General Objectives:

At the end of the chapter, you should be able to:

1. summarize the advantages and disadvantages of the different sampling

techniques;
 2. identify the biases that may arise in survey sampling and apply the best
sampling in different studies;
3. analyze and generalize on the population based on the available samples or
based on its given parameters and sample statistics; and
4. demonstrate the process of collection of raw data, and the sampling
techniques.

LESSON III. SURVEY SAMPLING

Survey sampling or simply sampling refers to the process of choosing a sample
of elements from a total population of elements. It is a process of selecting a subset of a
population of items for the purpose of making inferences about the whole population.
The two broad categories of sampling are probability and non-probability sampling that
will be discussed in this lesson.

At the end of the lesson, you should be able to:

1. define different terms in understanding the concept of sampling distribution;
2. differentiate population and sample;
3. analyze the quality of survey results using accuracy, precision, and margin
error;
4. determine the factors in attaining the best sampling design; and
5. summarize the advantages of sampling over population

The following terms are defined for us to fully understand the concept of
sampling distribution.

POPULATION VERSUS SAMPLE

We will encounter the terms population and sample on almost every page of this
module. Consequently, understanding the meaning of each of these two terms and
difference between them is crucial.
Suppose a statistician is interested in knowing
 1. the percentage of all voters in a city who will vote for a particular candidate in an
election.
2. The 2019 gross sales of all enterprises in Sultan Kudarat.
3. The prices of all houses in the Philippines.
In these examples, the statistician is interested in all voters, all enterprises, and
all houses. Each of these groups is called the population for the respective example. In
statistics, a population does not necessarily mean a collection of people. It can, in fact,
be a collection of people or of any kind of item such as houses, books, television sets,
or cars. The population of interest is usually called the target population.

Population is the totality of items or things under consideration. It is also the

large set of data. The population that is being studied is also called the target
population. Population size is denoted by “N”. A finite population is µ which consists of
a finite or fixed number of objects such as the following:
1. The net weight of 30,000 canned goods in a production lot.
2. The scores of 500 students who took the departmental examination during a
semester.
3. A number of accidents that occurred during the last three months.

An infinite population has no limit; therefore, its size cannot be determined.

Examples of infinite population are the possible expenses of all possible customers
arriving in a supermarket.

Sample is a subset of a population; hence, the sample must possess the same
characteristics of the population. Sample size is denoted by “n”.

Sampling

Sample
Population Inference
 FIGURE 2.1 Population and Sample
.

The collection of information from the elements of a population or a sample is

called a survey. A survey that includes every element of the target population is called
a census. Often the target population is very large. Hence, in practice, a census is
rarely taken because it is expensive and time-consuming. In many cases, it is even
impossible to identify each element of the target population. Usually, to conduct a
survey, we select a sample and collect the required information from the elements
included in that sample. We then make decisions based on this sample information.
Such a survey conducted on a sample is called a sample survey. As an example, if we
collect information on the 2019 incomes of all families in Sultan Kudarat, it will be
referred to as a census. On the other hand, if we collect information on the 2019
incomes of 50 families from Sultan Kudarat, it will be called a sample survey

KEY TERMS
The following terms are defined as guide whenever the terms are used in the
discussion of the entire module.
 Parameter is a measurable characteristic of a population, such as population
mean denoted by “µ” (mu) and population standard deviation denoted by “σ”
(sigma).
 Statistic is a measurable characteristic of a sample such as sample mean
denoted by “ẋ” (x-bar) or sample standard deviation, denoted by “s”.
 Sampling distribution is a probability distribution of statistics. When we say
sampling distribution of the mean, we are referring to the mean values of every
possible samples that can be obtained from the population.
 Sampling with replacement is used when a population element can be selected
more than one time. After a person or item is selected, it is returned to the frame
 where it has the same probability of being selected again. However, it is
generally considered more desirable to have a sample of different people or
items than to permit repetition of measurements on the same person or item.
 Sampling without replacement is used when a population element can be
selected only once. A person or item once selected, it is not returned to the frame
and therefore cannot be selected again.
 Standards error refers to the standard deviation of the sampling distribution.
Hence, the standard error of the mean is the standard deviation of the sampling
distribution of the mean.
 Variable noted by the letter X and Y, is a characteristic of interest for each person
or thing in a population. It may be qualitative or quantitative.
 Data are the actual values of the variable. They may be numbers, or they may be
in words. Datum is a single value.

Example:
Determine what the key terms refer to the following study. We want to know the
average amount of money first year college students spend at ABC College on school
supplies that do not include books. We randomly survey 100 first year students at the
college. Three of those students spent Php500, Php1,000, Php1,500.

Solution:
 The Population is all first year students attending ABC College this terms.
 The sample could be all students enrolled on one section of a beginning
statistics at ABC College (although this sample may not represent the entire
population).
 The parameter is the average (mean) amount of money spent (excluding books)
by first year college students at ABC College this term.
 The statistic is the average (mean) amount of money spent (excluding books) by
first year college students in the sample.
  The variable could be the amount of money spent (excluding books) by one first
year student. Let X = the amount of money spent (excluding books) by one first
year students at ABC College this term.
 The data are the peso amounts spent by the first year students in the sample.
Examples of the data are Php500, Php1,000, Php1,500.

Quality of Survey Results

The quality of survey results can be described by the researchers using one or
more of the following terms.
1. Accuracy. It refers to the closeness of the parameter of sample statistics to a
population. For example, if the sample mean is 99 and the real population mean
is 100, then the sample mean is accurate with gap of 1 unit.
2. Precision. It refers to the closeness of the estimates and the different samples.
An example of a measure of precision is standard error. Standard error is
inversely related to precision. For instance, when the standard error is small,
then the sample estimates are more precise. Likewise, when the standard error
is large, then the sample estimates are less precise.
3. Margin error. The maximum expected difference between the true population
parameter and a sample of that parameter us expressed by the margin error. The
larger the margin of error, the less confidence one should have that a poll result
would reflect the result of a survey of the entire population.

SAMPLE DESIGNS
A sample design can be described by two factors.
1. Sampling method. It is the process of selecting a part from a given whole. The
primary purpose of which is to make a generalization about the (unknown)
characteristics of a whole. It is also central to the study of statistical inference.
The rules and procedures by which some of the elements of the population are
included in the sample is referred to sampling method. Random sampling,
stratified sampling, and cluster sampling are some common sampling method
that will be discussed in the next lesson.
 2. Estimator. It refers to the process of calculating sampling statistics. Different
estimators can be used in different sampling methods. For instance, the equation
used in the computation of a mean score with a random sample is different from
the equation for computing a mean score with a stratified sample. The formula to
be used in the standard error may vary form one sampling method to another.

Survey objectives and survey resources are two factors where the “best” sampling
design depend. For example, the most economical design that provides a desired level
of precision may be selected by a researcher. In cases wherein budget is limited, the
researcher must then choose the design that provides the greatest precision without
going over budget.

Advantage of Sampling over Population

1. Economy or reduced cost relative to doing a complete enumeration of the
population.
2. Timeliness - with fewer observation to gather and process, the time needed to
finish the study is shorter.
3. Provides greater scope and coverage – for a given period, more items can be
covered with fewer observation.
4. May generate more accurate results.