Вы находитесь на странице: 1из 37

METHODS OF

RESEARCH

by
Dr. Ronald M. Henson
Communication Research Consultant
Methodology Defined
 Method – set of
procedures/steps followed in
research
 Logy – science

Therefore:
Methodology is defined as “the
science of procedures”
What Is Best Method?

 Based on objectives
 Helps gather the data
 Verifies the hypothesis
 Answers the problem/sub-
problems
Begin With Your Design
 A program that guides in
collecting, analyzing and
interpreting observations,
domain of the study
- Descriptive
- Experimental
- Exploratory
Descriptive Survey Design
 One-shot survey – determine state
of knowledge, attitude and practice
of people before or after
intervention
 Before-after survey – see change in
2 surveys: baseline (before
intervention) and terminal (after
intervention)
 Static group comparison – see
certain effects/impact of
intervention after application
Methods in Survey
 Questionnaire – set of closed-
ended questions
 Personal interview – face-to-
face gathering of data using a
schedule
 Telephone survey – semi-
personal method of collecting
data
Survey Procedures
 Determine what you need to know
and objectives
 Construct questions for the survey
 Pre-test the questions
 Identify respondents
 Select a group and pose questions to
the sample
 Collect and organize the information
gathered
 Report in a meaningful way
Experimental Design
 Posttest-only group – determine the
effect of communication intervention
to change knowledge, attitude and
practice, with experimental and
control areas
 Pretest-posttest control group – no
initial information on respondents,
baseline and terminal surveys
(before and after intervention is
applied surveys) using same
questionnaire
Variables of Research
LEVEL OF COMMITMENT OF
GOVERNMENT EMPLOYEES
AFFECTING JOB PERFORMANCE
 Independent – predictor variable

LEVEL OF COMMITMENT
 Dependent – resultant variable

JOB PERFORMANCE
 Intervening – control variable

AGE, INCOME, EDUCATION,


POSITION, SEX OF EMPLOYEES
Variables to Indicators
 LEVEL/TYPES OF
COMMITMENT (independent
variable)
- affective commitment
(emotional attachment)
- continuance commitment
(economic and social costs)
- normative (obligation)
Variables to Indicators

 SOCIO-DEMOGRAPHICS
(intervening variable)
- sex (male or female)
- age (young, mid-age, old)
- position (top, middle, bottom)
- income (low, moderate, high)
- education (elementary, high
school, college, post-college)
Variables to Indicators
 JOB PERFORMANCE
(resultant/dependent)
- outstanding
- very satisfactory
- satisfactory
- fair
- poor
Sampling
 A fraction of a population
N = population n = sample
 A desired sample size (no.)

 Based on objectives of study

 Evaluation of identified

respondents
 Economic sense
Sampling Procedures
 Identify general target
respondents of study.
 Indicate geographical location.
 Collect information about the
target respondents.
 Decide on specific area where
study will be conducted.
 Choose the best sampling
scheme.
Sampling Scheme
 Incidental/accidental sampling –
people you meet first until you get
the desired sample size
 Purposive sampling – having
established criteria in selecting
sample by location, sex or education
 Simple random – selection of
respondents with similar
characteristics, and equal chance of
getting represented, has replacement
Sampling Scheme
 Systematic sampling with
random start – with a list and
nth number, then chosen
randomly
 Stratified sampling – group
divided into layers (strata), done
systematically then chosen
randomly
 Cluster sampling – done in
groups or various areas with
more than one population
How To Determine
Sample Size
Sloven Formula
- where N (population) is over
1 + N (population) x .01, or .
05, or .10 squared
Where: .01, .05, .10 is margin of
error
Sampling Data
135 housewives involved in underground
economy and within age bracket 20-25
from Tondo, Manila
 Sampling frame (general category) –

housewives
 Sampling unit (specific) – who read

tabloids and are in age bracket 20-25


 Sampling size (number) – 135

 Sampling locale (place) – Tondo,

Manila
 Sampling scheme (technique) – simple

random
Instrumentation
The world is full of
measurements, always have
indicators to gauge…how tall,
how cold, how intelligent, how
effective?
 Survey-questionnaire – the basic
tool to measure information
gathered from respondents
Instrument Design
 Writing and formatting the
survey-questionnaire
- Easy to read
- Give clear instructions
- Encircle or check box
- Consider research objectives in
design, every objective
corresponds to survey questions
Pointers in Designing Instrument

 Make sure your language is


appropriate
 Avoid long questions
 Limit question to single idea,
don’t compound your ideas
 Establish your frame of
reference
Pointers in Designing Instrument
 Do not ask embarrassing
questions
 Avoid biased or leading
questions
 Use closed-ended questions
but make open-ended if
needed
 Get into filter questions when
necessary
Pointers in Designing Instrument
 Avoid ambiguous wording
 When asking sensitive
questions, use introductory or
transition statements
 Begin with interesting or easy-
to answer questions
 Use matrix questions when they
are always repeated
Preparing the Questionnaire
(for self-administered)
 Indicate title of the research
 Give instructions to respondents on
how to answer the questionnaire
 Greet the respondents
 Assure confidentiality of data
 Provide space for questionnaire no.
 Organize questions according to
grouping (socio-demographics,
awareness, attitude, etc.)
Questionnaire Contents
 Contents of the questionnaire
will depend on research
objectives
- Introduction
- Socio-demographics
- Awareness/exposure questions
- Knowledge questions
- Attitude/perceptions
- Practice/behavior questions
Pre-testing Questionnaire
 Means of verifying the
applicability of questionnaire to
field setting
 Testing with 20 sample-
respondents
 Find out weaknesses and
strengths of instrument with
respondents
Why Pre-Test?
 Uneasiness and confusion
 Unclear issue in each question
 Difficult words
 Questions which cause
embarrassment and resistance
 Appropriateness of question
sequence
 Mechanical difficulties
How To Determine Acceptability
of Questionnaire
 When significant portion left
unanswered
 Wrong answers, right questions
 Answers show too little variance, like
when answers are all 3s on a 5-point
scale
 Wrong respondent, right
questionnaire
 Physically incomplete, missing pages
How Valid Is Your Instrument
 Measurement can be reliable
but not valid, but when valid it
can be reliable
Analogy:
- A washing machine is reliable if

it washes clothes
- But it cannot be valid if it

washes 10k when the machine


can only wash 5 k
Measurement Validity
 Content validity – reflection of
attributes or content of concepts
 Criterion-related validity –
based on criterion already
considered valid
 Construct validity – inferred
from theory, assess based on
theoretical framework
Using Likert Scale
 Developed by Rensis Likert, a 5-
point measurement on attitude
and perception, levels of
agreement to statements
- Strongly Agree (SA) 1
- Agree (A) 2
- Neutral (N) 3
- Disagree (D) 4
- Strongly Disagree (SD) 5
Some Examples of Likert
There is proliferation of sex
stories in newspapers. (negative
-)
__SA __A __N __D __SD

•Newspapers carry values in their


stories. (positive +)
__SA __A __N __D __SD
Using Semantic Differential
Developed by Osgood and
Tannenbaum, this 7-point scale
measures the meanings people
create in response to specific
stimulus
Statement: Newspapers perform
informational function.
1 2 3 4 5 6 7
High __ __ __ __ __ __ __ Low
Main Points in Measurement
 Concepts are only mental images
used to bring observations
 Concepts do not exist in real world,
measure them using empiricism
 Conceptualization is the process of
making vague concepts into
measurements
 Precision refers to exactness of
measures
Main Points in Measurement
 Reliability refers to measurement
procedure which can be replicated in
a given phenomenon
 Validity refers to measurement that
relates to commonly accepted
meanings of a particular concept
 Reliable measures diminish the
richness of meaning in general
concepts
Data-Gathering Procedures
 Allows smooth implementation
of research
- Coordination and linkage
- Timetable
- Ocular inspection
- Field work
- Respondents/informants
Making Sense of Data
 Gathering data
 Arranging and tabulating
 Summarizing data
 Analyzing data
 Interpreting data
 Reporting findings

Вам также может понравиться