Sampling

Design

By:

Dr. Eunice B. Custodio

BulSU-Graduate School

Philippines
Sampling Design

Sampling is the part of statistical practice

concerned with the selection of individual observations
intended to yield some knowledge about a population of
concern, especially for the purposes of statistical
inference.
The sampling process consists of seven simple stages:
1. Defining the population of concern -1,400 employees – 5-10years in
service
2. Specifying a sampling frame, a set of items or events possible to
measure
3. Specifying a sampling method for selecting items or events from the
frame
4. Determining the sample size -
5. Implementing the sampling plan
6. Sampling and data collecting
7. Reviewing the sampling process
Sampling Methods
It is incumbent on the researcher to clearly define the target population. There
are no strict rules to follow, and the researcher must rely on logic and judgment. The
population is defined in keeping with the objectives of the study.
Sometimes, the entire population will be sufficiently small, and the researcher
can include the entire population in the study. This type of research is called a census
study because data is gathered on every member of the population.
Usually, the population is too large for the researcher to attempt to survey all
of its members. A small, but carefully chosen sample can be used to represent the
population. The sample reflects the characteristics of the population from which it is
drawn.
Sampling methods are classified as either probability or nonprobability.
Probability Samples
In probability samples, each member of the population
has a known non-zero probability of being selected. Probability
methods include random sampling, systematic sampling,
stratified sampling and cluster sampling.
1. Random sampling is the purest form of probability sampling.
Each member of the population has an equal and known chance
of being selected. When there are very large populations, it is
often difficult or impossible to identify every member of the
population, so the pool of available subjects becomes biased.
 Also, Random Sampling (simple random sampling) is
the most straightforward design. From a list of the population,
select n subjects at random. If we wished to survey the Bulacan
State University student population, we would generate a list of
all registered students. Then from this list we would use a random
number generator on a computer or a table of random numbers to
select say n=75 students randomly from the master list.
2. Stratified Sampling. Sometimes the population in question can be broken into a
few large groups, called strata. Then within a group or stratum, a simple random
sample can be selected within each group. The collection of all these samples will
then resemble the population as a whole.
For example, consider the student population again. We could break all
students into freshmen, sophomore, junior, senior status. And so, there would be
4 strata or groups.
STRATA >>>>> RANDOM >>MARGIN OF ERROR>>SLOVINS
Grade 7 =100 = 300>>random> systematic random 300-list
Grade 8 = 90 =
Grade 9=45
Grade 10 = 70
2,300
 Then from within each group we could select a simple
random sample. The four simple random samples would then
form the basis for our study of the Guandong University student
population. The idea behind stratified random sampling is to use
natural groupings where subjects are alike within groups. This
similarity will help reduce variation in the sample results and will
give us more accurate inferences on a question of interest.
3. Cluster Sampling. This kind of sampling will seem similar to
stratified sampling at first, but there are important distinctions.
Cluster sampling is useful if the population can be broken into
many small groups or clusters. Then from this list of clusters, a
simple random sample of clusters is chosen. For a cluster that is
chosen, each subject in the cluster is surveyed. A population can
be broken into households. There are many of these that make up
the population, and it would be easy to get such a list from official
records. From this list of households, we could select a simple
random sample of households, and then for each selected
household, we would survey each member residing there.
4. Systematic Sampling. This design is carried out by selecting
every kth subject from the master list. It is an approximation to
simple random sampling, but contains no randomness. Once it is
decided that every 12th subject will be included in the study, the
sample composition is determined completely, with no remaining
uncertainty.
Nonprobability Sampling
In nonprobability sampling, members are selected from
the population in some nonrandom manner. These include
convenience sampling, judgment sampling, quota sampling,
and snowball sampling.
1. Convenience Sampling. This is a potentially dangerous form of
sampling. As an example, suppose we wish to study the BSU student
population again. Now suppose we stand outside the student center
at 4:10 pm on a Monday afternoon and ask students the survey
questions. We will have studied the students available or convenient
to our sampling system. There is tremendous potential for bias in
this design. Why? Well, what kind of students will be hanging
around the student center at 4:10pm? They would tend to be hungry
students, tend to live on campus, and tend to not be in science
majors. These features could be related to the responses to our
survey, thus causing biased results. This type of survey is to be
avoided when possible.
2. Judgment sampling is a common nonprobability method. This
is also called Purposeful sampling. The researcher selects the
sample based on judgment. This is usually and extension of
convenience sampling. For example, a researcher may decide
to draw the entire sample from one "representative" city, even
though the population includes all cities. When using this
method, the researcher must be confident that the chosen
sample is truly representative of the entire population.
3. Quota sampling is the nonprobability equivalent of stratified
sampling. Like stratified sampling, the researcher first identifies
the stratums and their proportions as they are represented in the
population. Then convenience or judgment sampling is used to
select the required number of subjects from each stratum. This
differs from stratified sampling, where the stratums are filled by
random sampling.
4. Snowball sampling is a special nonprobability method used when
the desired sample characteristic is rare. It may be extremely
difficult or cost prohibitive to locate respondents in these
situations. Snowball sampling relies on referrals from initial
subjects to generate additional subjects. While this technique can
dramatically lower search costs, it comes at the expense of
introducing bias because the technique itself reduces the
likelihood that the sample will represent a good cross section from
the population.
 The advantage of probability sampling is that sampling
error can be calculated. Sampling error is the degree to which a
sample might differ from the population. When inferring to the
population, results are reported plus or minus the sampling error.

In nonprobability sampling, the degree to which the

sample differs from the population remains unknown.
Limitations of Sampling
Sampling depends on subjects’ motivation, honesty,
memory, and ability to respond. Subjects may not be aware of their
reasons for any given action. They may have forgotten their
reasons. They may not be motivated to give accurate answers; in
fact, they may be motivated to give answers that present
themselves in a favorable light.
 Structured surveys, particularly those with closed-ended
questions, may have low validity when researching affective
variables.
Although the chosen survey individuals are often a random
sample, errors due to non-response may exist. That is, people who
choose to respond on the survey may be different from those who
do not respond, thus biasing the estimates.
 Survey question answer-choices could lead to vague data
sets because at times they are relative only to a personal abstract
notion concerning "strength of choice".
For instance the choice "moderately agree" may mean
different things to different subjects, and to anyone interpreting the
data for correlation. Even yes or no answers are problematic
because subjects may for instance put "no" if the choice "only
once" is not available.
Data Processing
If statistical treatment techniques are used, this fact should
be mentioned and correspondingly reflected in the section on data
analysis. If one has used a technique that is, strictly speaking not
appropriate to the data this fact should be noted and explanation
made.
Methods and Procedures of
Collecting Data
Secondary data may be obtained through library search. The guidelines below
will help you in data collection.
– Search for Bibliography
– Extracting Information
– Select information to record, information which is useful and which is not.
– Determine how to record what we extract from the published materials.
– Set up and operate the orderly recording system.
– Evaluating Secondary Data.
Data Gathering Procedure
A description of the adoptation, construction and
administration of instruments should be included. Instruments
include tests, questionnaires, interview, guides and/or schedules, etc.
Apparatuses, devices and laboratory equipment, a drawing or
photograph may be useful
Categorization of Data
Primary. Data coming from the original sources of
materials and are collected especially for the task at hand.
Secondary. Data or information obtained from the works of
others who have collected it for other purpose. Organization files
and library holdings are the most frequently used secondary
sources, but statistical compendia, movie film, printed literature,
audio tapes and computer files are also widely used sources.
Primary data may be obtained through
the following field procedures:
1. INTERVIEW
– Personal interview is a two way purposed conversation initiated by an interviewer to
obtain information that is relevant to some research purpose.
– Telephone Interview are even more useful as the major or sole data collection method.
This popularity is possible because of the widespread acceptance of the telephone as a
necessary family communication device.
– Interviewing /by Mail/e-mail. The so-called mail survey is an interview using a self-
administered open-ended questionnaire The value of using mail is that we can contact
respondents who might otherwise be inaccessible.
– Observation. Most appropriate overt behavior of subjects under
investigation in various natural or controlled settings.
– Non-Behavioral Observation. Record Analysis involves historical
or current, public or private records: they may be written, printed,
sound recorded, photographic or video recorded.
– Physical Condition Analysis. Typified by store audits to
determine availability or merchandise studies of plant safety
compliance, analysis of inventory conditions, and the analysis of
the financial statements of organizations.
– Physical Process Analysis. Includes time studies of manufacturing
processes, traffic flows in a distribution system, paperwork flow in an
office and the study of financial flows in our banking system.
– Behavioral Observation. Non-Verbal Analysis. Includes body
movement, expressions and even exchanged glances.
– Linguistic Analysis. One simple type, familiar to students, is the tally
of “ahs” that a professor emits during a class; study of sales
presentation content; study of what, how and how much information is
conveyed in a training situation; Extra-Linguistic Analysis.
2. TABULATION OF DATA
Data Collection Methods. It is the collection of
relevant/appropriate and sufficient/adequate data at a given time.
The data collected are needed to answer the problem of the study.
The data which will be collected is intended for tabulation.
There must always be data that should be collected,
organized, analyzed, and interpreted in any research undertaking.
No research activity can be succeeded without necessary data so
the data collection is important phase of the research process.
 According to Frederick Lamson Whitney, data are the
things we think with. They are the raw materials of reflection,
until by comparison, combination, and evaluation they are stepped
up to higher levels of generalization, where again they serve as
basic materials for higher and higher thinking.
Data can be classified in different ways. It may be
objective or subjective. They may be quantitative or qualitative.
They may be attributes or variables. Data may also be classified
according to the methods or types of research that the researcher
may use. It can also classify as primary or secondary or as raw or
derived.
basic statistical tools for descriptive-
survey research designs
Descriptive statistics such as the use of frequency
distribution, mean, percentages and statistical description can
be used in describing the respondents.
The data must be examined to developed scores for each
variable. Responses to the individual questions must be tabulated,
mean scores and standard deviations must also be developed.
Numerical data must be encapsulated using frequency
distribution, mean and statistical description.
 To measure the significant relationship among the
different variables, Multiple Regression Analysis can be used.
to test whether there exist a significant relationship between
the variables as well as its significant relationship, the Analysis of
Variance (ANOVA) and F-test can employ.
The data could be processed by a computer system using
Microsoft Excel Application and Statistical Package for Social
Sciences (SPSS).