Challenges of Data Collection for Research in a Developing Country
Aiyub
1,2
, Sri Novitayani
1
, Marthoenis
1
1
Department of Psychiatry and Mental Health Nursing, Syiah Kuala University, Banda Aceh, Indonesia,
2
Doctoral Program of Mathematic and Applied Sciences, Syiah Kuala University, Banda Aceh, Indonesia
Keywords: Challenges, Data Collection, Survey.
Abstract: Data collection is a substantial step in conducting a research. Researchers usually face enormous challenges
in this step. While they have to maintain the validity of inferences, they also must anticipate such
unexpected factors that might intrude the results. One of the most important things in maintaining validity of
research results is to keep validity of data through a good and correct data collection process. The
challenges become greater when the data collection is carried out in developing countries where the levels
of education, socio-economic status of respondents as well as the enumerators’ capacity are significantly
varied. The present study aims to investigate the challenges of research data collection process through
interview by using questionnaire in a developing country of Indonesia. The study was conducted by
qualitative method with Focus Group Discussion (FGD) as data collection method. The respondents were
enumerators involved in a community based health survey in districts of Indonesia. Data were analyzed
using the qualitative content analysis method to identify themes, categories, and sub-categories. The study
had been identified two themes, namely: internal challenges (questionnaire, research status, sample election,
enumerator) and external challenges (research permit, respondent and research location) as the challenges of
data collection process. The themes were identified in the study will be guidance for researchers in planning
and conducting data collecting process through interviews by using questionnaire for enhancing validity of
research results in Indonesia and other developing countries.
1 INTRODUCTION
Data collection is an important step in conducting
a study. Researchers wish to obtain valid and
accurate data for rational finding. Survey is
designed to obtain information about the
prevalence, distribution and interrelation of
phenomenon (Polit, & Beck, 2012). Therefore,
researchers must design a well-structured plan to
anticipate every challenge faced on the field where
researchers take place in order to produce valid
and accurate data.
With regard to the characteristics of
respondents, the challenges of data collection
between developed and developing country are
different. In general, Hing et al., (2011) in a
literature review mentioned that poor design and
low response were two problems of health survey
using questionnaire in both developed and
developing countries. There are very few studies in
developing country were conducted to explore the
challenges of data collection from heath survey
using questionnaire, including in Indonesia. Whilst
enumerators who were involved in a community
based health survey in districts of Indonesia
reported some challenges faced in the field.
Therefore, this study aims to reveal challenges of
data collection as expressed by survey enumerators
in Indonesian settings.
The current study focuses to explore the
challenges were faced by enumerators in data
collection process. The data were collected using
interview and a Focus Group Discussion (FGD)
approaches. A deep interview was used to explore
more participants’ experiences about the
challenges of data collection. This article is
expected to be a guideline for researchers in
planning data collection procedure of health survey
in developing country, especially survey data
collection using enumerators.
164
Aiyub, ., Novitayani, S. and Marthoenis, .
Challenges of Data Collection for Research in a Developing Country.
DOI: 10.5220/0008396000002442
In Proceedings of the Aceh International Nursing Conference (AINC 2018), pages 164-169
ISBN: 978-989-758-413-8
Copyright
c
2022 by SCITEPRESS Science and Technology Publications, Lda. All rights reserved
2 METHODS
2.1 Study Design
This study was designed as qualitative method
with explorative approach. The aim of study was to
investigate the challenges of data collection
process in a community based health survey.
2.2 Participants
A total of nine enumerators involved in a
community based survey about mental disorders
and diabetes in the districts of Banda Aceh and
Aceh Besar, Indonesia.
2.3 Data Collection
Data were collected through focus group
discussion (FGDs). The FGD was conducted in
April 2018 and undertaken in a two hours meeting
with two sessions and 15 minutes break between
the sessions.
2.4 Data Analysis
Data were analyzed using qualitative content
analysis. This method was used to search for
manifest and latent meanings of transcript of
interview to formulate condensed meaning units,
codes, sub-categories, categories and themes
(Graneheim & Lundman. 2004).
The analysis was done in the following steps:
1) created transcript of interview; 2) the transcript
of interview read repeatedly to formulate
condensed meaning unit; 3) according to
condensed meaning unit, then researchers
formulated codes, sub-categories, categories, and
themes that were suitable to study purposes. Data
analysis are summarized in Table 1.
3 FINDINGS
Based on data analysis as shown in Table 1 above
(Subsection 2.4 Data Analysis), the internal
challenges consist of four categories, i.e.
questionnaire, research status, sample selection,
and enumerator. Whilst the external challenges
were divided into three categories, i.e. research
permit, respondent and research location.
Furthermore, each category was classified in
some sub-categories. For example, questionnaire
was distributed in two sub-categories (i.e. answer
design and question design), research status was
distributed in two sub-categories (i.e. professional
research and undergraduate research), sample
selection was distributed in two sub-categories (i.e.
rigid and flexible), and enumerator was distributed
in three sub-categories (i.e. enumerator’s fees,
enumerator status and enumerator capacity),
research permit was distributed in two sub-
categories (i.e. content of research permit and
coordination of research permit), respondent was
distributed in two sub-categories (i.e. education
level and experience and perception and culture),
and research location was distributed in two sub-
categories (i.e. rural. remote area and urban area).
Table 1: The Data Analysis Using Qualitative Content Analysis Method.
Themes
Categories
Sub-Categories
Codes
Condensed Meaning Unit
Internal Challenges
Questionnaire
Answer design
Likert scale
Likert scale raise confusion of respondents
Likert scale raise inconsistency of respondents’
answer.
Answer for
financing questions
Provide answer choices.
Question design
The numbers of
questions
Too many questions make respondents bored &
annoyed
Repeated questions make respondents bored &
annoyed
Question context
Mental health question decrease response of
respondents
Mental health question make respondents
uncomfortable.
Economic question make respondents hope for
humanitarian aid.
Challenges of Data Collection for Research in a Developing Country
165
Themes
Categories
Sub-Categories
Codes
Condensed Meaning Unit
Research
Status
Professional
research
Increase
participation
Professional research more acceptable.
Undergraduate
research
Lack of
participation
Undergraduate research less acceptable
Sample
Selection
Rigid
Time consuming
When respondents were not met, enumerators must
back for interview again.
Flexible
Saving time
If respondents can be replaced when they were not
met, enumerators not need to back for interview.
Enumerators
Enumerator’s
fees
Payment model of
enumerator’s fees.
It is not fair to pay enumerator’s fees per
questionnaire.
For fair payment, enumerators must have contract
with basic salary, transport fees, allowance and
insurance.
Enumerators
status
Education status
Students were considered less professional
Graduates were considered more professional
Citizen status
The presence of foreigner increase respondents’
acceptance.
The presence of foreigner also can raise suspicion.
The presence of foreigner triggered respondents
dramatizes data for humanitarian aid purpose.
Enumerators
capacity
Adaptability
Adaptability can increase acceptance.
Technique of adaptation was different between
urban and rural community.
Communication
skills
Clear explanation of research purpose, procedures
and benefits can enhance respondent’s acceptance.
Questionnaire
Understanding
Questionnaire understanding will enhance the
smoothness of data collection process.
External Challenges
Research Permit
Content of
research permit
Containing all
activities carried out
Complete information of research permit can
enhance the smoothness of data collection process.
Coordination of
research permit
All government
levels
Coordinate and communicate the research permit
with all government levels can enhance the
smoothness of data collection process.
All formal security
levels.
Coordinate and communicate the research permit
with all formal security levels can enhance the
smoothness of data collection process.
Coordination of
research permit
must be done before
data collection
process.
Coordination and communication of research permit
early can anticipate various problems in data
collection process.
Respondents
Education level
and experience
Education level will
influence
respondent’s
acceptance.
The higher level of respondent’s education, it will
be more critical.
Respondents with lower levels of education will be
more influenced by negative rumors from social
media.
The higher level of respondent’s education, the
respondent’s acceptance will be better.
Experience will
influence
respondent’s
acceptance.
Positive experience will enhance respondent’s
acceptance.
AINC 2018 - Aceh International Nursing Conference
166
Themes
Categories
Sub-Categories
Codes
Condensed Meaning Unit
Perception and
culture
Perception and
culture will
influence
respondent’s
acceptance.
Gender differences will influence respondent’s
acceptance.
Respondent’s participation was influenced by
negative rumors.
Stigma enhances reluctance of respondents to
answer mental health questions.
Respect for guest as a culture in Aceh society has
been increased respondent’s acceptance.
Local habits influence respondent’s acceptance.
Research
Location
Rural/ remote
area
Distance
Difficult to reach and time consuming.
Accessibility
Difficult and risk to find location.
Urban area
Distance
Respondents live in densely areas so that it is easy
to meet.
Accessibility
Near and easy to reach.
4 DISCUSSION
The present study reveals a wide variety of
challenges during a community based health
survey in Aceh Province, Indonesia. Overall, two
main themes presented as the challenges in data
collection, the internal and external challenges.
Internal challenges include the design of
questionnaire and its contents, the research status
whether it’s for academic of professional purpose,
sampling methods, sample selection and the
enumerators themselves. External factors include
the difficulties in obtaining research permit from
each health of villages, respondent selection, and
research location.
Hing et al (2011) mentioned that the main
source of error in survey include non-measurement
errors such as survey design, sample frame and
sampling, non-response and item non-response,
and measurement errors such as survey instrument
and data collection processes. Their statements
somehow also expressed by the informants of the
present study.
Questionnaire is one of challenging of data
collection process. Design of questions and
answers can influence accuracy and validity of
data. Even thought Hing, et al., (2011) mentioned
that face to face interview is particularly useful for
long questionnaire with complex questions. But in
this study participants reported that too many and
repeated questions made respondents bored and
annoyed. Besides they have not so much time to sit
and answer questionnaire.
Using Likert scale in a survey is quite popular
for studying phenomena from opinion and attitudes
in social and medical research (Janhunen, 2012;
Gob, McCollin, & Ramalhoto, 2007). But using
Likert scale in this study was reported as a
challenging because it can affect inconsistencies of
respondent’s answers. Participants said that some
respondents changed their answer when
enumerators validated their answer again. Some
respondents, especially those with lower education
level have difficulty in understanding the Likert
scale, therefore their response on the questions
might lead to bias. Therefore, enumerator should
ensure that the answers provided by subject are
valid. Here, enumerator's knowledge about Likert
scale must be good prepared so that they can guide
respondents to choose right answers according to
their thinking about.
According to participants’ statement, many of
respondents in this study were difficult to answer
questions of income and expenditure. Respondents
with non-permanent worker have not fixed income
and expenditure. Also, since the majority of
Acehnese population work as farmer, where their
incomes are not regular, but largely depend on
harvesting time. Thereof for these types of
questions, the researcher must provide some
answer choices in minimum to maximum range.
This manner of income should be considered in the
future study.
Moreover, research status is also an internal
challenge in data collection process. Respondents
in the field asked repeatedly about research status
(professional or undergraduate research). In
commonly, society pays little attention for
undergraduate research. Therefor enumerators
must be able to convince respondents about direct
and indirect benefits of the research for
respondents.
Sampling method is very important in data
accuracy and validity. Sample size and
Challenges of Data Collection for Research in a Developing Country
167
representative are two things that must be
considered to preserve data integrity and enhance
the validity of study findings (Polit, & Beck,
2012). However, the challenge is when
enumerators cannot meet the samples which have
been selected. Participants reported that they spent
much time for coming back for interview. This
created more difficulties when interviews were in
remote areas. Distance and accessibility are two
challenges faced by enumerators. Whilst
replacement of samples will disrupt data
representation, therefore it is important for
researchers to allocate enough time and budget in
research planning.
Results of this study show that enumerators are
important element in collecting data of survey.
Skills, experience and capability of enumerator
greatly influence the outcome of an interview
(Sahu, 2013; Haunberger, 2010). Therefore in this
study, enumerators were selected by considering of
their education qualification and experience. It is
reasonable because participants reported that many
respondents in the field asked about enumerator’s
education status. As discussed above, respondents
pay little attention for undergraduate research.
Before collecting data, enumerators must be
trained because in the field, capacity of
enumerators will determine validity and reliability
of data. Clear understanding of questionnaire will
enhance the smoothness of data collection process
and respondents were pleased during the interview
process.
Haunberger (2010) mentions that interviewer
attitude and behaviour will influence respondent’s
response. Therefore, participants mentioned that
enumerators must understand local culture for
more adaptability. It is different way to adapt
between urban, rural and remote communities.
Therefore, the presence of a foreigner was
responded differently. Some of respondents were
excited, some others were suspicious, and the
others thought would get humanitarian aid.
Communication skills of enumerators are needed
to explain research purpose clearly. Participants
mentioned that clear explanation about research
purpose and activities will increase participation of
respondents.
Beside of internal challenges, we found also
external challenges. The first external challenges
are research permit. Enumerators faced problems
in the field according to research permit. Some
community leaders (Geusyik) asked enumerators to
have recommendations of all civil administration
levels (district, sub-district and village), and some
others asked to have recommendation of all
security department levels (district and sub-district
police station). Meanwhile, the enumerators just
had recommendation of provincial government.
According to regulation, bureaucracy of
research permit in Indonesia is regulated by at least
four levels of formal regulation, namely:
Constitution of Indonesia, Number 18, 2002,
Regulation of Ministry of Home Affairs, number 7,
2014 and number 64, 2011. In addition, foreigners
or foreign institutions are also regulated by
government regulation, Number 41, 2006, and
Regulation of the Minister for Research and
Technology, Number 08/M/IX/2007. All the
regulations oblige researchers to have
recommendation from Central Government,
Province, and Regency/City depend on research
location or researchers background (Indonesia
Ministry of Research & Technology, 2012;
Agustine, et al., 2018).
In application of the regulations, different
departments have different procedures which often
raise negative implications for researchers and
research institution (Agustine, et al., 2018).
Furthermore, lack of information about research
procedures both from institution providing
research recommendation or others information
resources make researchers be confuse. Meanwhile
in the field, researchers must follow any unwritten
rules of community leaders. Therefore it was very
important to get information and complete all
research recommendation before collecting data to
minimize problems in the field. It was also
recommended by participants in interview.
The second external challenge was education
level of respondents in data collection. Participants
reported that respondents with high levels of
education were more cooperative in interview,
albeit rather more critical. Furthermore,
respondents’ experiences, culture and rumors will
influence the interview process. Instance, negative
stigma makes respondents reluctant to answer
questions of mental health. But cultures of
Acehnese people in respecting guest make
enumerators more acceptable. Even though,
enumerators must spend more time to follow the
tradition. Because of this, it is very good to select
enumerators from local community who have been
familiar with local culture, besides they have
knowledge in relate to research theme.
The third external challenges are research
location. Distance and accessibility were reported
as a challenge in data collection process. For
remote area, participants faced some difficulties
AINC 2018 - Aceh International Nursing Conference
168
like: difficult to reach location, time consuming
and high risk for enumerators. Therefore,
participants suggested to extend data collection
time and to create a fair scheme of payment for
enumerator’s fees. It was not fair to pay
enumerators fees per questionnaire, because
enumerators will face different difficulties
according to research location. For fair payment,
some participants proposed to make a contract with
basic salary, transport fees, allowance and
insurance.
5 CONCLUSIONS
There were two challenges that were faced by
enumerators in conducting survey in developing
countries, namely internal challenges
(questionnaire design, research status, sample
selection, and enumerators) and external
challenges (research permit, respondent, and
research location). Researchers must pay attention
to eliminate the challenges to get data validity and
reliability.
AKNOWLEDGEMENT
The researcher thanks all participants who were
involved in the study.
REFERENCES
Agustine, T.E., Suparman, H.N., Prawira, M.Y., Jannah,
A.N., and Sadat, D.N. 2018. Problems of research
permit in Indonesia: a case study in Yogyakarta and
Jawa Timur Province. Retrieved from:
https://www.kppod. org/backend/ files/laporan
penelitian/problem-izin-penelitian-di-indonesia.pdf.
Gob, R., McCollin, C., and Ramalhoto, F.R. 2007.
Ordinal Methodology in the Analysis of Likert
Scales. Quality and Quantity, 41(5): 601626.
Graneheim, U.H., and Lundman, B. 2004. Qualitative
content analysis in nursing research: concepts,
procedures and measures to achieve trustworthiness.
Nurse Education Today, 24(2): 105112.
Haunberger, S. 2010. The effect of interviewer,
respondent and area characteristics on cooperation in
panel survey a multilevel approach. Quality and
Quantity, 44(5): 957969.
Hing, C.B., Smith, T.O., Hooper, L., Song, F., Donell,
S.T. 2011. A review of how to conduct a surgical
survey using a questionnaire. The Knee, 18(4): 209
213.
Indonesia Ministry of Research and Technology Affair.
2012. Procedures of research permit for foreign
university, foreign research development
Institutions, foreign business entities and foreigners
in conducting research development Activities in
Indonesia. Retrieved from:
https://www.kemlu.go.id/moscow/id/ layanan-
konsuler/prosedur-dan-
peraturan/Documents/Prosedur%20Penelitian%20Ba
gi%20Peneliti% 20Asing.pdf.
Janhunen, K. 2012. A comparison of Likert-type rating
and visually-aided rating in a simple moral judgment
experiment. Quality and Quantity, 46(5): 1471
1477.
Polit, D.F. and Beck, C.T. 2012. Nursing research:
generating and assessing evidence for nursing
practice. Nine Edition, Philadelphia: Lippincott
Williams & Wilkins.
Sahu, P. K., 2013. Research methodology: A guide for
researchers in agricultural science, social science
and other related fields. India: Springer.
Challenges of Data Collection for Research in a Developing Country
169