Chapter 3:
Methodology
What are the sections in Chapter 3?
Research Design/Methodology
Research Locale
Subject/respondents of the Study
Sampling procedure/sample size
Instruments Used; scoring procedures
Conduct of the study
Ethical considerations
Treatment of Data
Research Design
Research design is the plan or method which the
researcher used in answering the problems set in the
study and in the conduct of the research.
This is categorized as to the procedure upon which
the researcher collects and analyzes the data.
The two broad categories for research designs are
the QUALITATIVE research and the QUANTITATIVE
research.
Both have their own merits and flaws, but they could
be complimentary to each other.
Research Design
QUANTITATIVE QUALITATIVE
MORE RESPONDENTS LESS PARTICIPANTS
SHORT CONTACT TIME LONGER CONTACT TIME
SURVEY QUESTIONAIRE INTERVIEW
MORE NUMBERS/TABLE AND MORE WORDS/PARAGRAPHS/
GRAPHS NARRATIVES
CLOSED-ENDED QUESTION OPEN ENDED QUESTION
Research Methodologies
Descriptive Research
Correlational
Causal-comparative
Historical
Experimental Research
Research Locale
The research locale indicates the place or
location where the study is conducted or
undertaken.
Its description should have relevance to the
topic/content of the study.
The locale could indicate the population, the
boundaries and the variables.
In many instances, the researcher provides a
map for its geographical illustration.
When this is done, the map should
reflect what the researcher highlights in
the study.
This map should be accurately plotted
out with correct scaling and legends,
otherwise if the locale is adequately
describe, the map could be omitted.
Subject of the study
The subject of the study could be “people” or
“non-people”.
For people it can either be taken from
population.
If the total population is very large then the
sampling is used.
The researcher should describe the subject of
the study as to the size of the population or
the sampling
Respondents
The respondents are those who
participate in answering the
instruments used in the study. The
respondents could also be the subjects
of the study, hence sometimes the
subheading for this section is the
subject-respondents.
Sampling Procedure
In conducting a research work we are often
faced with the dilemma on what constitute a
reliable and acceptable sampling. When the
population to be studied is very small, we do
not need to get a sample, as we can take the
whole population. This is because the
population size is manageable. There are
instances when we take the whole population
even if it is very large, and this happens when
we conduct a census.
Population
Gay (1976) defines a population as a
group to which the researcher would
like the result of a study to be
applicable to the general body. He
used the word “generalizable”.
Kerlinger (1973) defines population as
all members of any well-defined class of
people, events or objects.
Sample
Ferguson (1976) defines a sample as
any sub aggregate drawn from a
population. It is the small group that you
observe from a given population where
your generalization is made.
Sampling
Sampling (Ary, Jacob and Razavieh) is the
process which involves taking a part of the
population by making observations on this
representative group and generalizing the
findings to the bigger population.
It refers to strategies which enable one to
pick a subgroup from a larger group and then
use this subgroup as a basis for making
judgment about the larger group (Vockell,
1983)
The steps used in sampling
identification of a population,
determination of a required sample
size
selection of a sample.
Sample size
Slovin’s Formula
n = N/(1+Ne2)
where n = sample size, N = population size, and e =
desired margin of error, which is the percent
allowance for non precision because of the use of the
sample instead of the population.
For example, in a population of 9000 and a margin of
error set at 2%
n = 9000/[1 + 9000 (.02)2]
n = 1,957
Sampling Strategies
Probability
simple random sampling
lottery
systematic sampling
stratified sampling, cluster sampling.
Non-probability sampling
Purposive or deliberate sampling. This is
sampling taken on purpose.
Quota sampling. One identifies a set of
important characteristics of a population
and then select the desired sample in a non
random way until the quota is reach.
Convenience sampling. This is based on
the convenience of the researcher, like
through telephone, or survey, etc.
Instruments
Questionnaires requires the respondents to
write or answer questions
Checklists -respondents are given a list of
items where they simply check based from
the given question
Interviews are done by the researcher orally
to get information about some
issues/concerns.
Focus Group Discussion/Interview is a
carefully planned and moderated informal
discussion where one person’s ideas bounce
off another’s creating a chain of relation of
informative dialogue
Observations are concerned with the
recording of events and behaviors of those
being studied in their natural setting
Semantic Differential – rating scale
designed to measure the connotative
meaning of things, events and concepts
Q-sort is ranking of variables typically
printed on cards according to some
instructions
Likert Scale – scale showing agrrement
or disagreement of attitudes or opinions
Treatment of data
Could be qualitative or quantitative
Statistical treatment makes use of
statistical tools and techniques to
discover relationships between
variables, develop generalizations that
may be used to explain phenomenon,
and predict future occurrences.
Descriptive statistics are those that
involve mere counting and tabulation
like frequency counts, percentages,
averages, spreads.
Inferential statistics requires statistical
tests of significance where one can
make inferences. It can be parametric
or non-parametric
Parametric tests are the z-test, t-test and F-
test.
Non-parametric tests do not depend always
on specific type of distribution like the normal
curve. They are also called distribution-free
statistics, applied to both nominal and ordinal
data.
The chi-square is a commonly used non-
parametric test
Mean and Standard Deviation
Mean gives the average of the data
Standard deviation - Considered as the
most useful index of variability or
dispersion
Test for comparison
– Test whether the coefficient of
correlation is significant at a particular
level
– T-test for difference between means (t-
test for independent means, t-test for
dependent means
Analysis of Variance
– To determine if there are significant differences
among the means of more than two groups;
ANOVA; The F-value is the ratio of two
variances or mean squares
– MANOVA – multivariate analysis of variance -
general form of analysis of variance which
incorporates two or more independent variables
in the same analysis
Different correlational
techniques
Pearson-product moment correlation –
measure of relationship between two
variables that are usually of the interval
type of data
Spearman Rank-order correlation
coefficient- measure of correlation
between two sets of ordered pairs
Kendall’s tau – measure o correlation
between ranks
Kendall’s coefficient of concordance – useful
to determine the relationship among three or
more sets of ranks
Point-biserial coefficient special type of
Pearson product moment correlation
coefficient widely use d in test construction,
test validation and analysis
Biserial correlation coefficient – used In test
construction, test validation and analysis like
the point biserial coefficient
Phi coefficient –sometimes called fourfold
coefficient used when each of the variables
are dichotomous
Tetrachoric correlation coefficient – measure
of correlation between data that can be
reduced to two dichotomies
Partial correlation – whenever 2 or more variables are
correlated, there may be possibility that yet other
variables may explain any relationship that are found
Multiple regression – technique that enables
researches to determine a correlation between a
criterion variable (dependent) and the best
combination of two or more predictor variables
(independent)
Coefficient of multiple correlation - indicates the
strength of the correlation between the combination
of the predictor variables and the criterion variable
Coefficient of determination – symbolized by r2, the
coefficient of determination is the square of the
correlation between one predictor variable and a
criterion variable
Discriminant function analysis – technique used in
similar fashion as the multiple regression analysis
Factor analysis – technique that allows to determine if
many variables can be described by a few factors
Path analysis – used to test the possibility of a causal
connection among three or more variables
Some non-parametric
techniques
Techniques for independent samples: The
median test – applied to see if two groups
come from population having the same
median; The Mann-Whitney U test – non
parametric equivalent to the t-test when one is
dealing with ordinal data; The Kruskal-Wallis
One way Analysis of Variance - non
parametric equivalent to the One-way ANOVA
when one is dealing with ordinal data
Techniques for dependent samples:
The sign test non-parametric equivalent test
for t-test for dependent or correlated samples.
The Wilcoxon’s Matched pairs Signed ranks
test – is quite similar to the sign test; more
powerful and efficient since it combines the
sign of the difference with the relative size of
the difference;
The Friedman Two way Analysis of
variance – both sign test and Wilcoxon’s
matched pairs signed-ranks tests are
used for two correlated samples.
In case there are three or more
correlated or matched groups involved
the Friedman Two-way ANOVA is
appropriate
Tips for writing Chapter 3
First, examine the problems you set in the study
Check the type of research methodology based from
the problems
Conduct a needs assessment
Know how to describe the locale in relation to the
study
Determine the subject-respondents, sample size
from the given population in your study
Decide on what instrument to use. Surf the
net to check if there are available instruments
Prepare the instruments, have these checked,
tried-out, validated. Get the reliability when
needed.
Consider the scoring procedure, the criteria
used, and the rating scale
Get the approval of the thesis committee prior
to the launching of instrument
Get appropriate and adequate sample
size to launch the instrument
Retrieve the accomplished
questionnaire; survey; etc.
If interview/FGD(FOCUSS GROUP
DISCUSSION) is conducted transcribe
the discourses
Tabulate the data/ put the transcriptions
in matrices
Determine the treatment of data to be
utilized for analysis
Analyze the results based from the
scoring criteria/descriptive ratings/
qualifying statements
Remember
Always follow correct technical format
Be systematic and succinct in the
discussion on the procedures
Be clear and consistent in your
presentations
Maps and figures have to be clear and
accurate
The conduct of the study have to be
discussed in detailed to guide the
reader
Avoid being repetitive
Check sentences and edit your work.
You can ask another person to read
your write-up if this could be
understood.
Research ethics have to be strictly
followed
Informed consent form should be
accomplished by the respondents of the
study, especially for minors.
Follow correct protocol. Be courteous in
the conduct of the study.
THANK
YOU!
