Methods of Data Collection

Types of Data Collection

There are two primary approaches to gathering information about a situation,

person, problem, or issue.

When conducting research, you may need to collect the required information;
however, sometimes the required information is already available and

you just need to extract it. Therefore, methods of data collection can be
classified as: - Primary,

where the researcher collects the data, and Secondary, where someone has collected
the data

Primary can be further divided into three categories -

Observation, Interviews, and Questionnaires. Then, secondary

can also be categorized as Documents, Media, and Records.

Observation can be further divided into two - Participant and Non-participant.

Interviewing can also be divided into two: structured and unstructured.

None of the methods of data collection provides 100 percent accurate and reliable
information. The quality of the data

collected depends on several other factors, which we will identify as we discuss

each method.

Your skill as a researcher lies in your ability to address the factors that may
affect the quality of your data.

One of the main differences between experienced and inexperienced researchers lies
in their understanding

of these factors. Therefore, it is important for a beginner to be aware of them.

Observation is a way to collect primary data. Observation is a purposeful,

systematic, and selective

way of watching and listening to an interaction or phenomenon as it occurs. When

you are more

interested in the behavior of individuals, observation is the best way to collect

the required information.

There are two types of observations: 1. Participant observation, 2. Non-participant

observation. Participant observation is when you,

as a researcher, participate in the activities of the group being observed in the

same way as its members,

with or without their knowledge of being observed. For example, you want to study
the

reactions of the general public to disabled individuals. You can study their
reactions by acting as a disabled person yourself.

This type of data collection can also be of good quality. Non-participant

observation, on the other hand, is when you, as a researcher, do not

get involved in the activities of the group but remain a passive observer; meaning

watch and listen to the activities taking place and draw conclusions from them.

For example, you want to study the tasks performed by an engineer in a power plant.
As an observer, you could follow engineers in the power plant and take notes on
what they do in their daily work.

After several observations, conclusions can be drawn about the tasks engineers
perform in a power plant. The

use of observation as a method of data collection may suffer from a number of

problems.

As a beginner, you should be aware of these potential issues. # 1 -

When individuals or groups become aware that they are being observed, they may
change their behavior.

Depending on the situation, this change can be positive or negative -

it can increase or decrease, and this can affect your observation output.

There is always the possibility of observer bias. If an observer is not impartial,

they can easily introduce bias (in observation)

and there is no easy way to verify the observations and the conclusions drawn from
them.

Interpretations from observations can vary from observer to observer.

There are many ways to capture observations.

The selection of a recording method depends on the purpose of the observation.

The way an observation is recorded also determines whether it is a quantitative or

qualitative study.

Narrative / Descriptive recording - In this form of recording, the researcher

records a description of the interaction in their own words.
Such type of recording clearly falls into the domain of qualitative research.
Typically, a researcher makes brief notes about observation

and shortly after completing the observation, makes detailed notes in narrative
form.

The biggest advantage of narrative recording is that it provides a deeper insight

into the interaction. However, a disadvantage

is that an observer may be biased in their observation, and thus the

interpretations and conclusions drawn from the observation may also be distorted.

Using scales - some observers prefer to develop a scale to assess different aspects
of the interaction or phenomenon.
The recording is done on a scale developed by the observer / researcher. The
biggest advantage of using scales in

observational recording is that you don't need to spend time making detailed notes

and can therefore focus on the observation itself.

On the other hand, the problems with using a scale are that it does not provide
specific and in-depth information about the interaction.

Recording on electronic devices/media - Observation

can also be recorded on a videotape or other electronic devices and then analyzed.

The advantage of capturing an interaction in this way is that the observer can
review it multiple times

before an interaction is interpreted or conclusions are drawn from it.

he or she can also invite other professionals to view the interaction and get their
judgment.

one of the disadvantages is that some people may feel uncomfortable or behave
differently in front of a camera.

Therefore, the interaction may not be a true representation of the situation.

Interviewing is asking questions to respondents and recording their answers.

There are mainly two types - structured and unstructured

In a Structured interview, the researcher asks a predetermined set of questions,

using the same wording and sequence of questions

as indicated. An interviewer uses a written list of questions,

open or closed, prepared for use by an interviewer in a face-to-face

interaction (this can be face-to-face, by telephone, or via other electronic

media).

One of the main advantages of the structured interview is that it provides uniform
information,

which ensures the comparability of data.

Unstructured interview

Unstructured interviews are particularly very flexible.

The strength of unstructured interviews is the almost complete freedom they offer
in terms of content and structure. You are free

to order them in any way you like. You also have complete freedom in terms

of the wording you use and the way you explain questions to your respondents.

You can ask questions at any time and address issues

and you can change them at the other (moment).

Unstructured interviews are dominant in both quantitative and qualitative research.

The difference lies in how information obtained through them in response to your
questions is likely to be used. In quantitative research, the answers

are categorized into measurement scales and quantified. In qualitative research,

the answers are used as descriptors

integrated with your arguments and sequence of logic.

Questionnaire is one of the most well-known methods of primary data collection.

A questionnaire is a written list of questions, the answers to which are recorded

by respondents. In a questionnaire, respondents

read the questions, interpret what is expected, and record the answers.

Questionnaires can be administered in 3 ways. 1) Mailing questionnaire is a

commonly used approach for collecting information.

in Mailed questionnaire collection, you send the questionnaire by mail to potential

respondents. It is obvious that this approach assumes that you have access

to their addresses. Usually, it is a good idea to send a pre-addressed, self-

addressed envelope

with the questionnaire, as this may increase the response rate. A mailed
questionnaire should be accompanied by a cover letter.

One of the biggest issues with this method is low response rate.

Group administration - One of the best ways to administer a questionnaire is to get

a captive audience, such as students in a classroom,
people attending an event, participants in a program, or people in one location.

This ensures a very high response rate, as few will refuse to participate

in your study. Since you have personal contact with the study population, you can
also explain the purpose, relevance, and importance of the research

and can clarify any questions from respondents.

Administration in a public place - Sometimes you can administer a questionnaire in

a public place, such as a shopping mall, a health center, a hospital,
school, or pub. Of course, this depends on the type of research population you are
looking for and where it is likely to be found.

Usually, the purpose of the study is explained to potential respondents, and their
participation in the study is requested.

Apart from taking a little more time, this method offers all the advantages of
administering a questionnaire jointly.

We have seen that for both interviews and questionnaires, you need to ask a number
of questions. We will learn this in the next slide. There are two types

of questions you can formulate for your questionnaires or interviews. 1) Open-ended

-

In an open question, the possible answers are not given. In the case of a
questionnaire, the respondent writes the

answers in their own words, but in the case of an interview, the researcher records
the answers in a summary. For example,

if you ask Mention your age. If you handed out the questionnaire,

the respondent will write down 28 years. If you are interviewing, you will get the
input of the respondents of 28 years, and then you do that

record on the interview questionnaire. In a closed question, the possible answers

are set out in the questionnaire

and the respondent or the researcher ticks the category that matches the
respondents. When deciding whether to use

open or closed questions to obtain information about a variable,

visualize how you plan to use the generated information.

This is important because the way you frame your questions determines the unit of
measurement that can be used to classify the answers.

The unit of measurement dictates what statistical procedures are

can be applied to the data and how the information can be analyzed and presented.

Age of early adult, adult, middle-aged, old is not an appropriate measuring

instrument unless armed with measuring.

For example, Early adult 18-25 years, adult 26-34 years

Middle ages 35-50, 50 or older

Now we will learn to design effective questions.

There are five simple rules you should follow when designing questions.

Always use simple and everyday language - your respondents may not be highly
educated
and even if they are, they may still not know some of the 'simple' technical jargon
you are accustomed to.

Especially in a questionnaire, be extra careful to use words that your respondents

will understand, as you will not have the opportunity to explain it

questions to them. In this example question, many respondents, even some who are
well-educated, will not understand 'antipathy'

and therefore, they may answer or not answer the question without understanding.
Antipathy - dislike / avoidance / aversion

Avoid ambiguous questions - an ambiguous question is one that contains multiple

meanings.
and that can be interpreted differently by different respondents.
This results in different answers, making it difficult, if not impossible,

to draw valid conclusions from the information.

In the example, some people are not disabled

they will answer "No." Some with a disability but able to move will also answer
"No."

Do not ask double-barreled questions - A double-barreled question is a question

within a question.
The biggest problem with this type of question is that one does not know which

specific question a respondent has answered. In this example

The question consists of two parts: how often do you travel by bus and how much
money is spent on bus travel? In this type of question.

some respondents may answer the first part, while others may answer the second
part, and some may answer both parts.

Do not ask questions based on assumptions - In such questions, the researcher

assumes that respondents
fit into a certain category and seeks information based on that assumption.

This question is asked without checking whether respondents drive a car. In

situations like this, it is important to check

first if a respondent fits into the category you are inquiring about.

The sequence of questions in a questionnaire or in an interview schedule is

important as it affects the quality of information,
and the interest and even willingness of a respondent to participate in a study.

So far, we have discussed the primary source of data collection where the required
data were

collected by you for the specific purpose you have in mind.

There are occasions when your data have already been collected by someone else

and you just need to extract the required information for the purpose of your
study.

Both qualitative and quantitative research studies

use secondary sources as a method of data collection.

In qualitative research, you typically extract descriptive (historical and current)

and narrative information, and

in quantitative research, the extracted information is categorical or numerical.

Secondary data are mainly collected from previous research

studies of previously conducted similar studies

or through media - you can search for media publications, government documents
newspapers to find information or you can go for historical records (library,
journals, diaries, blogs). This will be the source

your secondary data. When suing data from secondary sources, you must be very
careful

as there may be certain issues with the availability, reliability, format, and
quality of the data.

The extent of this problem varies from source to source. While using such data,

some things to watch out for are - 1. Validity and reliability

The validity of information can noticeably vary from source to source. For example,

information obtained from a census is likely to be more valid and reliable than
that obtained from most personal diaries. 2. Availability of data -

It is common for researchers to assume that the required data will be available,
but you should not make this assumption.

Therefore, it is important to ensure that the required data are available before
proceeding with your study.

Format - Before deciding to use data from secondary sources, it is equally

important to check if the data are available in the required format.
For example, you may need to analyze data from a specific age range of 30 years or
older,

but in your source, data are available only for samples younger than 30 years. When
you have built your research instrument, whether it is an interview schedule or a
questionnaire, it is important that you test it

before using it for actual data collection. Pre-testing a research instrument

involves a critical examination of the understanding of each question

and its meaning as understood by a respondent.

A pre-test should be conducted under actual field conditions on a group of people

similar to your study population. The goal is not

to collect data but to identify any problems potential respondents may have in
understanding or interpreting a question.

Your goal is to determine if there are issues with understanding how a question is
asked, if its meaning is communicated,

or if different respondents interpret a question differently and determine

if their interpretation differs from what you were trying to convey.

If there are issues, you should review the text to make it clearer and unambiguous.