The Who, How, And Why of Research

8/4/2019 The Who, How, And Why of Research

1/31

THE WHO, HOW, AND

WHY OF RESEARCH


2/31

Objectives

To Understand Sampling Procedures and

Strategies

To Understand Research Strategies

To Understand Research Designs

To Identify Reasons For ConductingResearch


3/31

RESEARCH PROCESS


4/31

POPULATION, SAMPLE, & PARTICIPANTS

Population= the universe of people or objects

to which the study could be generalized.

Sample= the subset of people or objects that

will participate or be included in the study.

Participants= other term of the sample in

social science research.

Sampling= method for including people/objectin a study, or the process of selecting a small

number of elements from a larger defined

target group of elements such that the

information gathered from the small group will

allow judgments to be made about the larger

groups

Sampling Frame= a complete list of all the members of the population from which asample will be drawn. For instance, if you were doing a phone survey and selecting

names from the telephone book, the book would be your sampling frame , from which

you will draw your sample. Governmental census is always conducted using

geographical regions as the sampling frame.


5/31

Steps in a Sampling Process

Execute the Sampling Process

Determine the Sample Size

Select Sampling Technique(s)

Develop the Sampling Frame

Define the Population of Interest


6/31

Step 1:

The target population is the collection of elements or objects thatpossess the information sought by the researcher and about whichinferences are to be made. Thus, defining the target population is todetermine from whom or what the data to get in order to answerthe research question(s).

The target population should be defined in terms of elements,sampling units, extent, and time.

o An element is the object about which or from which theinformation is desired, e.g., the respondent.

o A sampling unit is an element, or a unit containing the element,that is available for selection at some stage of the samplingprocess.

o Extent refers to the geographical boundaries.

o Time is the time period under consideration.


7/31

Illustration (Defining Population)

A researcher intends to study which is the most favored taxi in

Jakarta and why. To get the answer, he formulates the following

research questions:

Among the cabs in Jakarta, which do people in Jakarta like to

ride in most?

Do the passengers prioritize hiring the taxi due to lower fare?

Is their inclination to prioritize the taxi based on the of the

driver hospitality ?

Do they like riding the taxi due to the cars cleanliness?

After studying some relevant literature, the researcher

decides that the population of this study will be the residences

of Jakarta who often use taxi as their transportation mode.


8/31

Step 2: Develop the Sampling Frame

A sampling frame is a complete list of all the members of the populationfrom which a sample will be drawn.

A sample is only relevant for the sample frame, not appropriate for makinginferences beyond the defined sample frame

Differentiate target population and sample populations

Target populations are all the individuals relevant to a study (i.e.,who/what we want to include)

The sample population is who/what we actually sample

Goal is for these to be equal

The target and sample populations can differ when some of the targetpopulation cant be sampled

After considering all relevant factors, the researcher decides to conduct thesurvey through telephone. Thus, he uses Jakarta phone book as his samplingframe. In other words, the people whose names are listed in the phone bookare the target population, and those who often ride a taxi are the sample.


9/31

Sampling technique selection refers to the

procedures we use to select members from the

sample frames for the sample. There are many ways

to do this, and they are discussed in the comingslides


10/31

Classification of Sampling Techniques

Sampling Techniques

Non-probability

Sampling Techniques

Probability

Sampling Techniques

Convenience

Sampling

Judgmental

Sampling

Quota

Sampling

Snowball

Sampling

Systematic

Sampling

Stratified

Sampling

Cluster

Sampling

Other Sampling

Techniques

Simple Random

Sampling


11/31

Convenience Sampling

Convenience sampling attempts to obtain a sample

because they are available (or convenient) to study. Often,

they are selected because they happen to be in the right

place at the right time. Some examples of this sampling are:

use of students,

use of members of social organizations

mall intercept interviews without qualifying the

respondents

people on the street interviews


12/31

Judgmental (Purposive) Sampling

Judgmental sampling is a form of convenience sampling in

which the population elements are selected based on the

judgment of the researcher that the people he chose are typical

or average, or possess a unique characteristic.

The researcher never knows if the sample is representative

of the population, and this method is largely limited to

exploratory research.

Examples:

Primary school teachers with more than ten years experience

expert witnesses used in court


13/31

Quota Sampling

Quota sampling may be viewed as two-stagerestricted judgmental sampling. The first stageconsists of developing control categories, or quotas,of population elements. In the second stage, sample

elements are selected based not randomly but onconvenience or judgment.

For example, in a university there are 12.000students, of which 60% are male and 40% are

female. If a researcher would like to interview 120students of both sexes, he should do meet 72 malestudents and 48 female students.


14/31

Snowball Sampling

Also called network, chain,

or reputational, snowball

sampling begins with a fewpeople or cases and then

gradually increases the sample

size as new contacts are

mentioned by the people youstarted out with.


15/31

Simple Random Sampling (SRS)

Simple random sampling involves picking a certainnumber of participants out of the total number ofpossible participants in the sampling frame.

This technique is used when the needed population

is relatively small, self-contained, or clearly defined.For instance, a city of say 40,000 inhabitant. Theresearcher simply obtains a list of all residents. Hethen can use a sequence of numbers from a randomnumbers table (or draws of a hat, flips of a coin), to

select, say 10%, 20%, or some portion of names onthat list. He should, however, make sure he does notdraw from any letter of the alphabet more heavilythan others.


16/31

Systematic Sampling

In using this method, a researcher moves through the

sampling frame list and selects one out of every fixed

number of entries. For example, if a phone survey needs400 respondents out of 10,000 people from a phone book

or purchased calling list, then in systematic sampling the

researcher will pick every 25th person (4% of 10,000 is400).

This is a form of random sampling, but because selection

is based on where one is in the list, not everyone has anequal chance of inclusion. For example, if we randomly

decide to start at the 10th person on the list, then the 11ththrough 34th people will not be included, even though

selection began with a random process.


17/31

Stratified Sampling

Stratified random sampling involves selecting researchparticipants based on their membership in a particular subgroupor stratum. This technique allows the sample to look more likethe population in terms of mirroring the different subgroups.

Unlike simple random sampling, stratified random samplinginvolves selecting research participants based on theirmembership in a stratum. Dividing the sampling frame into strata(plural of stratum) allows the researcher to sample peopleproportionately based on the size of each stratum.

For example, according to the latest census, 100.000 peoplelive in city X and 45.000 (45%) graduated from high school.In a stratified sample of 1,000 city residents, researcherswould make sure that 450 of the 1,000 sample residents weresenior high school graduates.


18/31

The target population is first divided intomutually exclusive and collectivelyexhaustive subpopulations, or clusters(usually counties, census tracts, or otherboundaries).

Then a random sample of clusters is

selected, based on a probability samplingtechnique such as SRS.

For each selected cluster, either all theelements are included in the sample (one-stage) or a sample of elements is drawnprobabilistically (two-stage).

Elements within a cluster should be asheterogeneous as possible, but clustersthemselves should be as homogeneous aspossible. Ideally, each cluster should be asmall-scale representation of the population.


19/31

Types of Cluster Sampling

Cluster Sampling

One-StageSampling MultistageSamplingTwo-StageSampling

Simple Cluster

Sampling

Probability

Proportionateto Size Sampling


20/31

The correct sample size for a research depends on thepurpose of the study and the nature of the populationunder scrutiny. In relation to this ones knowledge to theoverall demographics; age, sex, class, etc. of the populationis necessary. The greater the diversity and differences that

exist in the population, the larger your sample size shouldbe.

The sample size of 30 is often held to be the minimumnumber of cases if researchers plan to use some form ofstatistical analysis on their data.

However, in reality, the larger the sample the better, forgreater size of sample will not only give greater reliabilitybut also enables more sophisticated statistics to be used.


21/31

Two Common Ways of Sample Size Determination

A. Subjective methods (less sophisticated methods)

The rule of thumb approach: e.g.. 5% of population

Conventional approach: e.g. average of sample

sizes of similar other studies;

Cost basis approach: The number that can bestudied with the available funds;

B. Statistical formulae (more sophisticated methods)

Confidence interval approach.


22/31

Execute the Sampling Process

After considering his objective and budget/time limitation, theresearcher intending to survey the most favored taxi in Jakartaand the reasons why it is so finally decides to use the clustersampling method.

He first divides the residents of Jakarta listed in thephonebook into five cluster (East, West, North, South, andMiddle Jakarta). In each region, he randomly phones 50.000persons, expecting that a half of them (25.000) ride cabs quiteoften. Those who reply that they use taxi quite often will then

be interviewed. Using this scenario, the researcher expect toget 250.000 respondents (around 5% of the 5 mil. residents ofJakarta) for the study.


23/31

SELECTING RESEARCH STRATEGY

EXPERIMENTAL RESEARCH

in laboratory settings

Using independent variable which affects the dependent variable (outcome)

Disadvantage: the experiment may be unlike the real world

DESCRIPTIVE & CORRELATIONAL RESEARCH

describes the attitudes and behaviors observed during the investigation

Commonly uses naturalistic observation which collects data where people areordinarily found

SURVEY RESEARCH

could use and analyze primary secondary data.

Conducted using tlphone surveys, mail surveys, email surveys, face-to-face interview


24/31


25/31

Can involve collecting quantitative information Can describe categories of qualitative information such as

patterns of interaction when using technology in the classroom.

Does not fit neatly into either category

Involves gathering data that describe events and then organizes,

tabulates, depicts, and describes the data. Uses description as a tool to organize data into patterns that

emerge during analysis.

Often uses visual aids such as graphs and charts to aid the reader

Quite effective within the educational setting since educational

research experiences cannot be as realistically controlled aslaboratory experiments. Descriptive research allows for the human

element of research.


26/31

Survey is a method of obtaining information about a

population from representative sample.

Survey research studies large and small populations

(or universes) by selecting and studying sampleschosen from the populations to discover the relative

incidence, distribution and interrelations of

sociological and psychological variables.

Survey is differentiated to cross-sectional surveys

and longitudinal survey.


27/31

Types of Designs

(frequency of data collection)

ONE - SHOT DESIGN

one group of participants is studied only one time

Used in surveys, experiments, or field studies.

LONGITUDINAL DESIGN

studies the same people over multiple data-collection periods.

CROSS - SECTIONAL DESIGN

studies several different groups of people of different ages to comparewhether age differences exist in the behavior or attitude being studied

REPEATED INDEPENDENT SAMPLES DESIGN

investigate a different sample of people over repeated trials to trackchanges in behaviors or attitudes


28/31

to verify existing theories,refining, modification orrestating the theories orinterpreting new theories

lays down the foundationfor the applied science thatfollow

Basic Research

Associated with specificproblem & have immediatepractice application.

It aims at finding out a

solution for an immediateproblem faced by a society

Applied Research


29/31

The purpose of research should be clearly defined & commonconcepts are used.

The procedural design of research should be carefully planed toestimate their effect upon finding or to yield results that one set inobjective.

Data should be adequate & authentic for analysis & the method ofanalysis used should be appropriate.

The reliability & validity of the data should the checked carefully.

Conduction should be confined, justified by the data of researchand be limited to those for which it can provide with an adequatebasis.

Greater confidence in research is warranted if the researcher hasexperience and a good reputation in research.


30/31

To generate new knowledge-explore the reality.

To describe the phenomenon under investigation.

To frame verifiable generalization and

understanding that explain how the variableinvolve in the situation behave.

To generate more knowledge & understanding ofthe phenomenon that occur & to build theoriesbased on research result.

To refine the existing method & to develop new &more effective method design & procedure.


31/31

References

Creswell, John W. 2008. Educational Research: Planning, Conducting,and Evaluating Quantitative and Qualitative Research. New Jersey:Pearson

VanderStoep, Scott W. & Johnston, Deirdre D. 2008. Research Methodsfor Everyday Life: Blending Qualitative and Quantitative Approaches.San Francisco: Jossey-Bass

The Who, How, And Why of Research

Documents