Data Collection and Analysis,
Interpretation and Discussion
1
Summary
Chapter one
Basic concepts (definition, type, sources of knowledge,
characteristics, purpose …of research )
Chapter two
Research Process (general, for quantitative survey, quantitative
experimental, for qualitative , design science / engineering research)
Special focus on first two steps ; problem selection and literature
review
Chapter three
Details on the three research designs/approaches ; quantitative,
qualitative, and design science
The purpose and focus of each approach and also comparing them
Data collection and sampling issues suitable for each approach
Specific methods under each approach
2
Every research study requires
information.
The question however is, from where
to get the information (data) and how
to get them?
Have you ever collected data so far?
What is your experience?
3
Data Collection – in Quantitative
and Qualitative, Design research
4
Sources of Data
Data can be acquired from different sources. But
different sources can be generally be categorized in two
major groups as
Primary and Secondary data sources.
It refers to those that were written or come into being
by the people directly involved in the research
Are statistical material which the investigator originates
for the purpose of the inquiry at hand
Are data collected afresh and for the first time and this
happen to be original in character.
The difference between secondary and primary data is
one of degree, data, which are primary in the hands of
one person, may be secondary in the hands of another.
5
Examples of Primary Sources
Archives and manuscript material.
photographs, audio recordings, video recordings, films.
journals, letters and diaries.
speeches.
published books, newspapers and magazine clippings
published at the time.
government publications.
oral histories.
6
Secondary sources of data
This data is that which has been collected by someone else
and which has already been passed through the statistical
processes. Thus studies made by others for another
purpose represent secondary data.
7
Advantages of Secondary data:
It is found more quickly and cheaply
Helps much on past events and distant places, where the
researcher cannot afford to collect the data right now.
8
Disadvantages of Secondary data:
The information does not meet ones specific needs, since
it is collected by others for their own purpose, definitions
would differ, units of measurements would differ, and
different time periods may be involved.
It is difficult to asses the accuracy of the information
because one knows little about the research design or
the conditions under which the research took place.
It is often out of date: a study made five years ago may
not be relevant today.
These may be free from personal bias and prejudices.
It may not be adequate.
9
Data Collection Methods
Primary data may be collected through a variety of
methods. The investigator can use one of the following
two methods: Observation or Survey
The personal observation Method (non reactive
research): Much of what one knows comes from
observation.
While such observation may be the basis for knowledge,
the collection process is haphazard.
Observation includes the full range of monitoring
behavioral (related to persons, body movement, motor
expressions, exchange of glances) and non behavioral
activities (record analysis, physical condition analysis,
physical process analysis) and conditions.
10
Observational methods
Observational methods may be gathered by employing
structured or unstructured methods that are either
disguised or undisguised. Observations may be made in a
contrived or a natural setting and may be secured by a
human or mechanical observer.
Direct and indirect observation: Direct observation
refers to observing behaviour as it actually occurs.
Indirect observation refers to observing some record of
past behaviour.
11
Data Collection – Quantitative and
Qualitative, design research
Data collection is after
developing instruments
Pilot and Main (actual)
In case of survey, qualitative studies
In case of experimental , analytical or predictive:
determine the data sources
In case of design science :
Setting the objectives of the design artifacts and later after
the construction of the artifact.
Requires sufficient preparation - has to be reported
Construct /variable identification and definition
Determining type of information
12
Cont…
It is recommended that you create a matrix that aligns your
research questions with the information you assess is needed
and the methods that you will use to collect that information.
Four areas of information are typically needed for most qualitative
studies as an example : contextual, perceptual, demographic, and
theoretical
Even further constructing a matrix that lists the research
questions along the horizontal axis and the interview questions
(as one method of data collection) down the vertical axis can
further indicate the extent to which your interview questions have
achieved the necessary coverage of your research questions.
13
Cont…
14
Cont…
15
Cont…
Developing Data collection Instruments
Two general approaches (in case of people respondent)
Adopting from similar previous researches
Interview questions , questionnaire items, observation
checklists
Mention how and what part is adopted
Crafting yourself based / inline with specific objectives
No orphan question or objectives should be observed
And pilot study/member checking to validate
Describe procedures of data sources accessing and acquisition (in
case of non-human data sources)
16
Cont…
Data collection is followed by analysis
In strict sense in case of quantitative but not in qualitative
In qualitative you collect- analyze- collect…
Until “saturation “
In design research (before the suggestion of the design and
after construction for evaluation)
Then analysis could start
by arranging; presentation
Categories, Tables, graphs etc…
by description of the nature of the data (like in NLP and DM)
The case of data understanding in machine learning
By structuring requirements/set of criteria
17 Refer: pdf document named Data [Link]
Make sure that your data
collection and analysis is
logical!!!
18
