Methods for Collecting Data

h n ic
ly tec ept
e P o at D
T h S t
ath &
M
Outline
Levels of Measurement Summarized
Methods of Collecting Data
Experimentation
Survey & Questionnaire
Observation
Comparing Experiment, Survey &
Observation
Documentation: Primary & Secondary Data
Steps in Constructing a questionnaire
Closed-ended & Open-Ended questions
Summary & Next Lecture
 Levels of Measurement -
Summary
Measurement is the process of determining the
value/label of a variable or a value/label assumed
by a variable for an element.
Four levels of measurement namely: Nominal,
Ordinal, Interval & Ratio
Knowledge of the level of measurement assists
in deciding on how to interpret data values and
selecting tools for data analysis
Data elements classified as either nominal or
ordinal measurement are usually referred to as
categorical data.
Data elements classified as either interval or ratio
measurement are usually referred to as
 Method for Collecting Data
Common methods of data collection are:
Surveys/Questionnaires
Observations
Use of Documented Data
Experiment
Registration method*
Computer simulation*
 Experimentation
Definition: An experiment is a method of
collecting data where there is actual human
interference with the conditions by controlling the
environment and using a randomization mechanism
to determine the treatment to be administered to
an experimental unit.

Experiments are ideal in studying the effect of the

explanatory (independent) variable on the response
(dependent)variable.
 Definition of terms
The explanatory variable/s are the variables
in the study whose values are believed to have
an effect on the value of the response variable.
A response variable is a variable that exists as
a consequence of the explanatory variable.

The treatments or factor levels are the

values or categories of the explanatory variable
that are being considered in the study.
The extraneous variable/s are those variables
that may have an effect on the response
variable but their effects are not of interest in
the study.
 An example
 Suppose that we wish to study the effect of jogging on lung capacity.
We could take a group of homogeneous subjects (subjects whose
physical characteristics are very similar) and randomly assign them to
several groups. One group jogs 1 mile each day and another group
jogs 5 miles each day. To further control the experiment, we have a
control group that does not jog. Ideally we would require that all
subjects be treated the same (diet, sleep, emotional stress, etc.)
except for the distance that they jog.

 In this study, the distance jogged is the “explanatory variable” or

the factor under study, while lung capacity is the “response
variable”. There are three categories in the explanatory variable (no.
of miles jogged in a day): 0, 1, 5. We call the categories of the
explanatory variable the “treatments” or factor levels.

 There are also other variables that we believed may have an effect on
the response variable but their effects are not of interest to us. We
call them “extraneous variables”. In this study, the identified
extraneous variables are diet, amount of sleep, emotional stress
etc.
 Survey Method
The survey method is a method of collecting
data by asking people questions using a
questionnaire.
The people from whom the questions are asked
in a survey are called respondents.

NB: The questionnaire may contain varied

types of questions on the respondent’s actual
behavior, attitudes, experiences, socio-
demographics, etc. Researcher hopes that the
respondent will give truthful answers.
 Types of Questionnaires
The data elements are classified into distinct,
mutually exclusive, and exhaustive categories.

Self-administered Questionnaire
The respondent himself reads the questions and
writes his/her responses on the self-administered
questionnaire.

Interview Schedule
An interviewer asks the questions and writes the
responses of the respondent on an interview
schedule.
 How to Disseminating the Self-
Administered Questionnaire?

Distribution by mail/post or hand or as inserts

in reading materials such as newspapers

On-line surveys where the respondent reads

the questions and sends his/her responses
electronically via the internet.
 Methods of Communicating the
Interview Schedule
Personal one-on-one interviews
Telephone interviews
Focus group discussions (FGD) where a
moderator follows an interview guide to direct
a freewheeling discussion among a small group
of people.
FGDs are usually used to:
[Link] the research hypothesis
[Link] a list of possible responses to a
question
[Link] an explanation for the results of a study
 Adv of Self-Administered
Questionnaire
Questionnaire can be mailed or handed to the
respondents or uploaded on the internet or e-
mailed for thereby allowing multiple/simultaneous
administration
It is appropriate in obtaining objective information
that requires no further explanation since
information is limited to respondent’s written
answers to the prearranged questions.
Respondents can be freer to express their views
and have less pressure to answer immediately due
the absence of an interviewer.
Can promote high response rate when mailed with
self-stamped and addressed envelopes to avoid
expense on the part of the respondents.
 Adv of Interview Schedule
The use of the interview schedule is flexible and
adaptable. It can be used to handle different types
of persons, such as those who are illiterate or too
young to read and write.
It allows for probing and clarification of responses
to achieve specific and more accurate responses.
Interviewer may note nonverbal as well as verbal
behaviour in face-to-face interviews.
May result in a much higher response rate than the
use of self-administered questionnaires, especially
for topics that concern personal qualities or
negative feelings.
Callbacks and persistence of interviewers can
convince reluctant respondents to participate in the
survey.
 Disadv of Interview
Schedule
Poorly-trained and dishonest interviewers may
potentially increase errors in measurement
through biased or fabricated responses.

It is generally more expensive and more time-

consuming.
 Factors to Consider…
Considerations in choosing the method of
communicating the questions in a survey:
[Link] of data needed
[Link]
[Link]
[Link] rate
[Link] coverage
[Link] of good interviewers
 Observation Method

The observation method is a method of

collecting data on the phenomenon of interest
by recording the observations made about the
phenomenon as it actually happens.
 An Example

Researchers can study the adjustment of

college freshmen (Year0) to campus life by
observing their behaviour in various settings.
Researchers can observe animals in their
natural habitat
 Notes on Observations
Method
Observation of subject/s is usually done in its
natural setting.
Applicable when subjects are unable to
verbalize their feelings because they cannot
speak. For example, studies of reactions of
babies to different stimuli.
The data collected using the observation
method is usually the subjective perceptions
and interpretations of the researcher on the
event under study.
Usually, the researchers give a qualitative
narrative of what they have observed about the
phenomenon of interest.
 Notes Cont…
However, it is still possible to collect objective
quantitative data using the observation method
through:
[Link] recording
[Link] count recording
[Link] recording
[Link] recording and
[Link] sampling.
 Methods of Collecting Quantitative Data
using the Observation Method
Duration Recording
 The researcher records the duration of time a
particular kind of behaviour lasts. A stopwatch keeps
track of the duration of the behaviour. For example,
we may record how long a student uses the cell
phone.

Frequency-count Recording
 The researcher records the number of times a
particular behaviour occurs in a given time span.
Usually, the researcher has a list of behaviour types
and then he counts every time a particular behaviour
occurs. This procedure is preferred when the duration
of the behaviour is from one to five seconds. For
example, researcher counts how many times the
teacher says the word “OK” in a 2 period class.
 Methods of Collecting Quantitative Data
using the Observation Method Cont…
Interval Recording
The researcher partitions time into fixed time
intervals and counts the number of time intervals
where the behaviour occurred. For example, the 2
period class is divided into 5-minute intervals and
the researcher counts how many 5-minute intervals
when the teacher says the word “OK”.

Latency Recording
The researcher measures the length of time
between the stimulus and the first occurrence of
the behaviour of interest. For example, a
researcher records how long before a patient goes
to sleep after taking a sleeping pill.
 Methods of Collecting Quantitative Data
using the Observation Method Cont…
Time Sampling
The researcher records the phenomenon of
interest or presence/absence of behavior under
study at every specified time schedule. For
example, the nurse measures the temperature of a
patient every 15 minutes.
Comparison of the Experiment, Survey &
Observation Methods
Data Collection Method
Aspect Experiment Survey Observation

Ability to establish Super Poor Poor

cause-and-effect
Assessing the Sometimes Generally Oftentimes
reliability of difficult possible difficult
generalizations
about a well-defined
population
Realism of data Least realistic Realistic Most
realistic
 Use of Documented Data
The researcher can obtain documented data
from previous studies of individuals or private,
government, nongovernment agencies.

The researcher may find documented data in

published or written reports, unpublished
documents, periodicals, and others.
 Classification of Documented Data by
Source
Primary Data
Primary data is data documented by the data
collectors themselves.

Secondary Data
Secondary data is data documented by an
individual/agency, other than the data
collectors.
 Adv of Primary Data
The primary source frequently provides vital
information that are crucial in assessing the
applicability and accuracy of the collected data.
This includes the definitions of terms and
statistical units used in the survey, methodology
including a copy of the questionnaire and a
discussion of the sampling design or
experimental design.
The primary data are more comprehensive.
Secondary data have already been filtered by
secondary source to address their purpose.
The secondary data may contain mistakes due
to errors in transcription made when the figures
where copied from the primary source.
 Basic Steps in Constructing a
Questionnaire
Step 1. Decide on the type of questionnaire,
whether self-administered or an interview schedule.
Step 2. Write the questions. Decide on the type of
question/s, wording of the question and the
response categories for those questions that
include a list of response options from which the
respondents will choose their answers.
Step 3. Put together all the questions formed in
Step 2 and decide on their ordering.
Step 4. Write an introductory statement or cover
letter.
Step 5. Write the special instructions to
interviewers or respondents that will appear on the
questionnaire.
Questionnaire
Cont…
Step 6. Create and design the form. Consider a
layout that makes it easy to understand the
questions and easy to record the answers. The
design of the form should also facilitate easy data
capturing/entry into the computer.
Step 7. If it is a nationwide study, translate the
questionnaire to other major languages. Let
someone else translate the translated
questionnaire back into the original language.
Compare it with the original questionnaire and
check that there had been no changes in the
meanings of all the questions.
Step 8. Pretest the questionnaire to a few
respondents to detect and correct errors in the
questionnaire.
 Types of Questions
Two general types
A closed-ended question is a type of question
that includes a list of response categories from
which the respondent will select his/her answer.
e.g. How do you spend your pocket money?
( ) books ( ) supplies ( ) food ( ) music cds ( )clothes
( ) others, please specify ____________
An open-ended question is a type of question
that does not include response categories.
e.g. How do you spend your pocket money?

____________________________________________________
____________________________________________
 Advantages
Closed-ended Questions Open-ended Questions
Easy to code and analyze Respondent can freely
High response rate since answer
are simple and quick to Can reveal new ideas
answer and views that the
Response categories researcher might not
make questions easy to have considered
understand Good for complex issues
Can repeat the study and Good for questions
easily make comparisons whose possible
i.e. replication is easy responses are unknown
Respondents need not be
Get detailed answers
articulate
Shows how respondents
think
 Variations of Closed-Ended
Questions
Two-way question; e.g. [ ] Yes or [ ] No options
Multiple-choice questions
Checklist questions
e.g. What type/s of music do you like?
[ ] Classic [ ] Traditional [ ] Rock [ ] Jazz [ ] RnB
[ ] Hip hop [ ] Other, please specify _________ [ ]
None
Ranking questions
e.g. Which media do you find more helpful in
disseminating information? Write 1 for the most
helpful … 7 for the least helpful
[ ] Radio [ ] TV [ ] Newspaper [ ] Magazine [ ]
Bill boards [ ] Mailed leaflets [ ] internet
 Variations of Closed-Ended
Questions
Rating scale question
e.g. Indicate your satisfaction with the
cafeteria services on campus by encircling the
appropriate number on the scale below.
1 2 3 4
5
Very Dissatisfied Very
Satisfied
 Summary

Next Lecture: Sampling & Sampling Methods