1

Data collection
tools
Types of data gathering methods
available
Questionnaire
Interview
Observation
Focus group discussion
Rating scales
Inventories
Content analysis
Physical measurement instruments
Questionnaire
A collection of statements, questions, or ratings through
which one tries to elicit understanding, values, feelings,
attitude, preferences etc of research subjects
Items may be structured or unstructured, closed or open-
ended
Structured items reduce the tendency of respondents
giving fake answers
Open-ended allow for collection of rich descriptive data
Close-ended
questionnaire
Respondent is constrained or limited to choosing one or
several options listed e.g.
Do you like mathematics? YES NO
I like mathematics AGREE DISAGREE
I like mathematics
Strongly agree Agree Disagree Strongly disagree
Open-ended
questionnaire
If well constructed give rich information under unlimiting or
unconstrained situation
Summarizing and classifying responses from items is difficult, likely to
involve subjective judgment, and thus be a source of invalidity
Advantages of
questionnaires
Economy of time and fund
Uniformity of response for structured questionnaire – hence high
reliability
Anonimity is encourage- frankness in responses is enhanced
A

DISADVANTAGES OF
QUESTIONNAIRES
Disadvantages of
questionnaires
Assumes that all respondents are literate therefore may lead to
problems with sampling
Response rate tends to depend on length of questionnaire, complexity
of questions, relative importance of study, reputation of the sponsoring
institution, quality and design of the questionnaire
Disadvantages of
questionnaires
Compared to the interview, the motivation of respondent cannot be
aroused especially for mailed questionnaires
Reliability of
questionnaires
The extent to which a questionnaire measures
consistently what it is measuring depends on the clarity
of items and cooperation and honesty of respondents
Reliability can be estimated through test-retest, parallel
form, parallel form-delayed, split half method and Kuder
Richardson/Cronbach measures of reliability
Validity of
questionnaires
How well a questionnaire measures what it was designed
to measure
Touches on how well the behavior was defined, how
items were worded and presented
Validity can be ascertained by correlating results from
questionnaire with those collected using another method
(e.g. observation), or by finding the correlation between
results of different behaviors that do not relate
(convergent and discriminant validity)
Constructing
questionnaires
Questions can consist of items of different scale types
Likert scale
Rating scales
Semantic differential scale
Tips for constructing questionnaires
Be brief polite and clear in the introduction
Give only factual information
Use simple, clear and direct instructions
Be sure statements represent what you want to measure
Use simple, not compound and complex statements
Avoid words like ‘all’, ‘none’, ‘never’ ‘only’
Tips for constructing questionnaires
Word sentences to avoid prestige bias
Avoid embarrassing and threatening statements
Try to obtain good rapport with respondents
Phrase statements using the first person so as not to
sound accusing
Some questions may be strategically repeated to ensure
reliability of the respondent
Tips for constructing
questionnaires: what to avoid
factual statements
irrelevant items
statements susceptible to multiple interpretations
Long incomprehensible sentences
double negatives
statements that provoke guessing
emotionally loaded statements
Interviews
A face to face interpersonal role situation during which the interviewer
asks questions and the interviewee responds
Interviews can be structured or unstructured
Unstructured interviews are more open as regards sequence, words
and questions
Interviewing skills
required
Questioning skills
Listening skills
Note taking skills
Moderation skills
Inter personal skills
b

ADVANTAGES OF
INTERVIEWS
Advantages of
interviews
The interviewer is very flexible and readily adaptable to
a variety of situations
It is the best option in cases where only oral
communication is possible e.g. when responses are to
be gotten from children, the illiterate
Interviewer is able to record not only what is said but
also facial expressions and gestures (non verbal cues)
Interviewer can develop and maintain some level of
motivation and rapport with respondent
b

DISDVANTAGES OF
INTERVIEWS
Disadvantages of
interviews
Time consuming
Responses are diverse and may constitute special
reliability and validity problems
Negative interviewer effects due to personal abilities,
error in asking questions and probing, dressing…
Need to train interviewers on interviewing skills
Interviews may end up with opinions rather than facts
Reliability and validity
of interviews
Validity of the interview depends much on the personal
characteristics and skills of the interview

Reliability can be estimated by correlating results of a

single interviewer interviewing the same respondents
twice (intra-interviewer) or have two interviewers
interview the same person (inter-interviewer)
Reliability and validity
of interviews
The most effective way of improving reliability and validity of interviews
includes constructing many questions that reflect the behavior under
study and training interviewer well on interviewing skills and recording
responses
 Observation
Definition: systematic selection and recording of behaviors of people a given
environment.
A well defined behavior could be observed in a natural or contrived situation
In contrived situation the researcher arranges conditions that appear to be
natural but which have been artificially prepared
Covert versus Overt
observation
Covert observation is when the observed persons are NOT informed
about being observed. The danger with this is that it is ethically
contestable. This kind of observation is done in open spaces – stadium,
airport, café etc. where there is no need to make an agreement.
Overt observation is done with the participant aware that they are being
watched
 What to observe
According to Merriam (1998):
(a) The physical setting
The researcher should know the physical environment and
the behavior in which the setting is designed for. Other
important things such as; objects, resources, technological
devices etc. The researcher has to identify preliminary
concepts and what data will be gathered as indicators of
those concepts.
 What to observe
(b) The participants
The researcher should know who are going to be involved when carrying out
the research project, as well as the roles they will be carrying out. It is also
advisable that the researcher may identify “key informants” as they will help
the researcher to familiarize himself /herself with the environment.
 What to observe
(c) Activities and interactions
The researcher has to indicate when he/she is going to begin and how long the
activity will last. He/she has to state the activities that will be carried out and
show if there is interrelation between people and activities.
 What to observe
(d) Conversation
It is important that the researcher knows what he/she is
going to discuss and with whom will s/he talk to.
The silences and non verbal behavior that add meaning to
the exchange should be noted.
Any behavioral differences between researcher and the
informant or even the informant and his/her group, these
differences should be noted. When reporting, quote
directly as much as possible.
 What to observe
(e) Subtle factors
The researcher should take into consideration the informal and unplanned
activities, symbolic and connotation meaning of words, non verbal
communication such as dress and physical space, physical clues and things
that were to happen and the reason why they may not be happening.
 What to observe
(f) Your own behavior
An observer should know his/ her role. He/she has to understand what
he/she says or do because these may affect the scene being observed. The
observers comments should also be included as part of field notes.
 Researcher’s role
Complete observer - The researcher is hidden or be in a completely public
setting such as a library, a bus stop etc. The researcher does not become
part of the social process
Participant observer - The researcher and the participant are equal
partners, the researcher works with the participant as s/he observes and
records
 Reporting of the
findings
The researcher has to jot down notes which will be used
when constructing field notes
In addition to notes researchers may use photographs,
video tapes, and audio tapes as means of accurately
capturing a setting.
Other sources : official records, letters, newspapers
accounts, diaries, and reports as well as published data
used in a review of literature
Advantage & disadvantages of
observation
Observation allows researcher to gather data from
respondent’s true self BUT
Major problem is that the researcher is the observer;
biases and personal opinions may influence
interpretation of findings
If the subject is conscious that he is being observed, he
may try to change his behavior
Enhancing validity and
reliability
Observer should know exactly what they are to observe
Observe for extended periods of time so that respondent relaxes and
exhibits their true behavior
AVOID COMPLEX PRESENTATIONS IN RESEARCH TOOLS
POSSIBLE CHOICES IN INFERENTIAL STATISTICAL
ANALYSIS
Type Of test Parametric Method Non-parametric Equivalent
Test of differences between t-test for independent sample
-Wald Wolfowitz runs test
Groups (Two sample- comparing -Mann-Whitnet U test
mean value for same variable-Kolmogrov-Smirnov two
of interest) sample test
ANOVA/MANOVA -Kruskal Wallis Analysis rank
(Multiple) test
-Median test
Test of difference between t-test for dependent samples -Signed test
variables (cf two variables in same -Wilcoxon’s matched pairs
sample)
Repeat ANOVA -Friedman’s two way analysis
(cf for more than two -Cochran Q
variables in same sample)
Test of relationship between Correlation coefficient -Spearman rho
variables -Kendall tau
-Coefficient Gamma
-Chi-Square
-Phi Coefficient
-Fishers Exact test
1

DISCUSSION
QUESTIONS?
CONCERNS?
FEARS?
THREATS?