6/17/2022
1 2
• Introduction
• Data collection is a component of research
implementation carried out following the proposal
development stage
CHAPTER 5 • It is a systematic method of gathering data from different
sources in order to find out answers to the research
questions (objectives)
Methods of Data Collection • Research data can be collected from a variety of sources
using different methods
• “Garbage in Garbage Out”-False data provides wrong
output and wrong conclusion
3 4
Types of data: • Using primary and secondary data sources have their own
1) Primary data and advantages and limitations
2) Secondary data sources Advantages of primary data
Primary sources: • Primary data is usually up to date since collected in real-
• Provide raw data and first-hand evidence time
• It allows the researcher a direct access to the study subject
• The researcher has full control over the data collection
Secondary sources:
process
• Provide a second-hand information generated by others.
• Primary data is more reliable than secondary sources
• It refers to the data collected in the past by someone else
• Secondary sources are usually once primary data but Disadvantages of primary data
become secondary later when used by others. • More expensive compared to secondary data.
• Examples include journal articles and books • Time-consuming
1
6/17/2022
5 6
Advantages of Secondary data Methods of Data collection
Easily accessible compared to primary data
• Method of data collection depends on the research
Are usually freely available approach and nature of data to be collected
Less time-consuming Based on the nature research approach
It makes possible to carry out historical studies
I. Qualitative research approach
It helps to develop new insights into existing primary data
II. Quantitative research approach
Disadvantages of Secondary data
May not be reliable, the researcher may need to verify the
data
Some data might be exaggerated due to personal bias
It may become outdated
7 8
Qualitative research approach:
Common Methods of Data Collection
• An approach focusing on non-numerical data and related
1. Review of literature
ways of analysis
2. Questionnaires
• Emphasizes on qualitative results.
3. Focus group discussions (FGDs)
• Example, it explores attitudes, behavior and experiences
4. Field observations and ground surveying
of respondents
5. Experiments
Quantitative research approach:
• It involves numerical or statistical data.
• Finally the data is analyzed quantitatively
2
6/17/2022
9 10
1. Literature review Structured:
• Used to collect secondary data sources • Contains closed-ended questions only
• This involves collection of data from published or • Has standard answers (e.g. yes/no, boxes to tick in)
unpublished documents
• Used to generate statistics in quantitative research.
• Examples: textbooks, government reports, published
articles etc. • Eg. how many people use a service?
• 2. Questionnaires • The answer might be 70% or 500
• Used to collect primary data sources Semi-structured:
• The three common types of questionnaires are:
• Contains both open-ended and close-ended questions.
• Structured
• Semi-structured • Is more commonly used
• Unstructured
11 12
Unstructured: • Therefore, once you choose a questionnaire as a tool of
Contains open-ended questions exclusively data collection, you need to decide whether to construct a
closed-ended, open-ended or a combination of questions
• Such questionnaires do not contain boxes to tick, but instead
a blank section for the respondent to write an answer • In open questions, respondents use their own words to
• Mainly used in qualitative research,
answer a question,
• Some researchers may quantify the answers during the • Whereas in closed questions prewritten response
analysis stage categories are provided
• As there are no standard answers to such questions, data
analysis is more complex.
• eg. what people think about a service?
3
6/17/2022
13 14
• In addition, the researcher has to decide whether the Structure of questions
questionnaire is to be:
• Questionnaires:
Self-administered or Interviewer administered
• Should be short and simple
Self-administered:
• Should not be double-barreled(two question in one)-
• Self-administered questionnaires can be sent through post,
or distributed via the internet
instead split it into two
• The respondent fills it in on his/her own, away from the • Avoid nonsense and technical terms whenever
researcher possible
Interviewer administered: • Avoid uncertainty (words with double meanings)
• Respondents are asked by the researcher (or enumerator)
and interactive discussion occurs for open-ended questions
15 16
Advantage of FGD
3. Focus group discussion/FGD
• Enable to obtain a wide range of responses from
• Focus groups refer to discussion groups or group participants
interviews • Allows interactive conversation among participants
• A group of people are asked to come together to • Helps respondents to remember issues they might be
discuss on a certain issue forgotten
• The discussion is led by a moderator or facilitator who • The group effect is a useful resource in data analysis
introduces the topic, asks specific questions, controls
deviations from the topic
• Make sure that no one should dominates the discussion
to ensure each of the participants makes a contribution.
4
6/17/2022
17 18
Disadvantage 4. Field observation and ground surveying
• Some people may be uncomfortable in a group setting
• Field data collection using tools such as GPS, theodolite
• Every one may not contribute to the group discussion and total station
and may be dominated by some of the group members
Example:
• Others participants may influence individual views • Capturing locations of ground reference samples for land
• Difficult to extract individual views during data analysis.
use/land cover classification using Remote Sensing
images
• Taking geodetic reference points with DGPS
• Tracking road networks etc
19 20
5. Experiments Sample size determination
• Experimental research method where the relationship
between two variables is going to be examined. • From a given population the sample size is
• One of the variables can be manipulated, and the other is determined by:
measured.
• These two variables are classified as dependent and 𝑁
𝑛=
independent variables. 1+𝑁 𝑒 2
• In experimental research, data are mostly collected based on Where
the cause and effect of the two variables being studied. N – population size
• This type of research is common among medical
n – sample size
researchers, and it uses quantitative research approach.
e – margin of error
