Copyright © 2020 Statswork. All rights reserved 1

Collecting Your Own Data-Primary Research

Dr. Nancy Agens, Head,

Technical Operations, Statswork

info@statswork.com

I. INTRODUCTION

In every research studies, the

scientist wants to identify the truth or they

wish to develop meaningful insights into

the problem. In order to make any valid

conclusion about the research problem or

hypothesis, we need valid evidence.

Moreover, the evidence can be represented

by collecting data (numeric or non-

numeric in nature). In any statistical

analysis, data collection is the

fundamental step and there are two forms

of data collection; Primary and

Secondary method of data collection. I

will describe you the meaning of primary

data, examples of primary data collection

and different ways of collecting the same.

II. PRIMARY DATA

The researcher collected the

Primary data through surveys, experiment,

interview, etc, and it is considered to be

the best research methodology as the data

are collected from the original source.

However, before collecting the primary

data, the main task of the researcher is to

identify the sample population and the

mode of data collection.

III. EXAMPLES OF PRIMARY DATA

COLLECTION

Suppose if you are interested in

identifying the sales of a product in the

market and make a profitable

business plan, then you may conduct a

sample survey about the product to

selected customers and identify the

pattern of sales and fix a target for

future years (Niraula, 2019).

In many psychological experiments,

they often collect primary data through

lab experiments, questionnaire, etc, to

understand the relationship between

the variables under study.

Suppose a research student want to

find out which cool drink they like

most. Then he/she collect the data by

conducting experimental research. That

is, he/she take a sample of say 30

people and ask them to choose which

drink taste good from the 5 different

cool drinks presented before them and

recorded the findings for the study.

IV. PRIMARY DATA COLLECTION

METHODS

Primary data can be collected in

different ways and are explained as

follows:

Data collection through Interview –

In this method, there are two groups:

one is the researcher asking questions

to the respondents or the interviewer

and the other group is the people

answering to the questions or

interviewee. An interview can be done

either telephonic conversation or direct

(face-to-face) interview. The benefits

of the interview method are that the

information is clear and deep, and

missing data can be easily identified

and we can control the size of the

sample. The limitations of this method

are it involves longer time to complete,

manual mistakes while recording or

coding the data and cost incurred is

high (Pochiraju & Seshadri, 2019).

Questionnaires and Surveys – In this

case, the researcher frame a set of

questions to be asked to the

Copyright © 2020 Statswork. All rights reserved 2

respondents and collect the data by

circulating them. Mostly, a pilot study

should be conducted when we make

use of this data collection method.

These data can be collected either

offline mode (direct survey like a

census) or on-line mode (google

forum, email, etc). The benefits of this

data collection are the respondent gets

enough time to answer the questions,

and the cost of conducting this method

is comparatively low. The limitation of

this is once the question is framed, we

cannot alter in the middle and it

requires more time.

Data collection through observation

- This type of data collection is often

used in behavioural science. The

researcher uses a tool to collect the

data. For example, in a teaching

requirement process, the person is

asked to present a demo to evaluate the

capability of the teacher. The benefits

of this method are that it is not related

to the past or future outcome and is

objective in nature. The limitation is

limited information and selected target

samples.

Data through Experiments – This

type of data collection is often carried

out in the medical field. Suppose a

researcher wants to identify the

performance of a new drug and he/she

conduct an experiment by taking a

sample subject. Here the subjects are

pre determined because only the

patients with certain disease are tested

by giving the appropriate drug. The

benefits of this method are there is no

non-response in the data and it is also

objective in nature. The limitation is

that human error is possible and

expensive.

The main advantages of collecting

primary data over the secondary data is

that (1) it provides freedom to collect data

according to your research objective (2)

the recorded data is more accurate than

secondary data (3) you don’t need to share

the data to others unless if you wish to (4)

you can have proper control in the sample

size and finally (5) the information is up to

date (Gray, 2009).

Statistical support service Once

the collected the data then proceed the next

phase as statistical analysis. The statistical

analysis is used to describe the data

structure, classify the data, correlate or

identify the relationship, validating the

data, predicting the future outcomes in the

study. Data analysis service offers the

serveral tests that available in the literature

for analysing primary data. For example, if

the sample size is small, then one may use

the t-distribution test and if the sample is

large enough then one may use z-test, etc.

In addition, there are few situations that

the data may be non-numeric in nature,

then one can use the classification and

clustering technique for analysing the

primary data (De Vaus, 1986).

To sum up, primary research

involves the proper collection of data and

appropriate research objective. The first

step of any primary research is to identify

the data type and choose the appropriate

data collection method and proper

statistical tool to make reliable results. The

primary data collection method is often

accurate and gives real time conclusions.

Primary research can be either qualitative

or quantitative or a combination of both

types in nature. On the basis of the data

type you can execute either qualitative or

quantitative analysis to make

conclusions.

REFERENCES

[1] Gray, D.E. (2009). Doing Research in the Real

World. 2nd Ed. [Online]. United kingdom: SAGE

Publications. Available from:

https://books.google.co.in/books?id=l4dEQTQl1N

YC&dq=doing+research+in+the+real+world&sou

rce=gbs_navlinks_s.

[2] Niraula (2019). A review of research process, data

collection and analysis. Insights in Biology and

Medicine. [Online]. 3 (1). pp. 001–006. Available

from: https://www.heighpubs.org/hjbm/ibm-

aid1014.php.

[3] Pochiraju, B. & Seshadri, S. (2019). Essentials of

Business Analytics. International Series in

Operations Research & Management Science. B.

Pochiraju & S. Seshadri (eds.). [Online]. Cham:

Copyright © 2020 Statswork. All rights reserved 3

Springer International Publishing. Available from:

http://link.springer.com/10.1007/978-3-319-

68837-4.

[4] De Vaus, D.A. (1986). Surveys in social research.

[Online]. Allen & Unwin. Available from:

https://espace.library.uq.edu.au/view/UQ:225795.