CHAPTER THREE

RESEARCH METHODOLOGY

3.0 Introduction

This chapter presents the research paradigm, research design adopted for the study, population, sample and sampling strategies, data collection methods and tools, data quality control, ethics, gender consideration, the limitations and delimiters of the study.

how to write methodology in research

3.1 Research paradigm

Research Data Management practices is a social real-life issue that can be explained using different theoretical frames. However, for purpose of answering the research question, a pragmatic philosophical stance is adopted. This is concerned with what works and provides solutions to an identified problem (Creswell, 2013; Patton, 2002. Pragmatism allows the researcher to emphasize the research problem and use all approaches available to address the problem. It is an approach that uses mixed methods. How “research data is managed” and why there has been a slow readiness to adopt and uptake of RDM practices” as constructed by respondents and its implications are explored. Thus pragmatism will give the researcher the freedom of choice of methods, techniques, and procedures of research that best meets the needs and purpose of the study (Creswell, 2013b.).

3.2 Research Design

Different research designs relate to philosophical assumptions and the research design is associated with a pragmatic paradigm (Creswell, 2013; Kothari, 2004). The research design associated with the pragmatic paradigm involves mixed methods (Creswell, 2012). The current study adopted a concurrent parallel design combining the survey design applied within a case study. The mixing of the two designs provided a better understanding of the research problem since it utilizes and is built upon the strengths of both quantitative and qualitative data (Creswell, 2008; Saunders, et al., 2012).

Methodology writing

3.2.1 Case Study Research Design

The case study is a method of study that focuses on in-depth rather than breadth. This research will use a case study design involving three health research institutes. The case study is a design in qualitative research an objective as well as a product of inquiry (Creswell, 2013). It is where the researcher explores real-life multiple bounded systems (cases) over time through detailed in-depth data collection involving multiple sources of information in which inferences can be drawn (Creswell, 2013). The case study design will be a basis to explore how research data is managed, the level of RDM awareness and readiness, given why a slow response to adopting research data management practices to undertake data-intensive research (Ritchie, et al., 2013).

3.2.2 Survey Research Design

Creswell (2012) defines survey research design as procedures in quantitative research in which investigators administer a survey to a sample or to the entire population of people to describe the attitudes, opinions, behaviors, or characteristics of the population. The current study adopted a cross-sectional survey design since the researcher shall collect data at one point in time and measure the practices, awareness, and readiness then (Creswell, 2012). Furthermore, survey research typically collects data using two basic forms: questionnaire and interview which will be applied in the current study.

3.3 Area of Study

The study area will include: Uganda Cancer Institute, Uganda Virus Research Institute and Joint Clinical Research Centre with its regional service centers in Mbale, Gulu, Mbarara, and Fort Portal assumed to be health research institutes in Uganda recognized by the Ministry of Health. Although each of the selected cases has its history, they are considered to be centers of excellence in their respective line of research within national, regional, and international collaborations

3.4 Population

The population is the entire set of respondents from whom the study sample with common observable characteristics (sample) will be drawn (Taherdoost, 2018). The population of the study is 490 respondents as shown in the table below;

Table A: Population of possible respondents

How to write chapter three

3.5 Sample Size and Sampling Strategies

A sample is any part of a fully defined population (Banerjee and Chaudhury, 2010). Here below the sample size and sampling strategies of the study are explained.

3.5.1 Sample Size

The ideal sample size for researchers as respondents was calculated using the Cochran formula at the desired level of precision, confidence level, and the estimated proportion of the attribute present in the population.

The Cochran formula is:

Where:

• e is the desired level of precision (i.e. the margin of error),
• p is the (estimated) proportion of the population that has the attribute in question,
• q is 1 – p.
• the z-value is found in a Z-table.

This gave a sample size of 212 Respondents. However, to allow a representative sample from each study site, Stratified sampling was used to generate an appropriate sample depending on the population of researchers available in each study site. Consequently, Cochran’s formula is modified to calculate a sample for small (Hypergeometric) populations, applied as shown below:

n=

Where:

n= Sample size

N= Population size

Z=z-score

e= Margin of error

P= Sample proportion (If unknown we use 0.5)

N=469, Z-score at 95% confidence level=1.96, e=5%, P=0.5

Substituting into the formula

n=

Sample size (n) = 212

For each stratum, a proportionate stratification shall be given by the formula below.

Where:            =Strata sample size,

=Strata population size,

= strata

However, to attain a balance in response from each of the study sites, the sample for the respondent for the questionnaires shall be as here below derived from the calculation:

Sample for UCI ( ) =          = 25

Sample for JCRC ( ) =       = 16

Sample for UVRI ( ) =       = 171

The sample size for key informants includes: Managers/Administrators of research, System Administrators/Data Officers, and Librarians/Medical Record Officers from each of the selected health institutes was one respondent per category per institute which makes the total of 9 Respondents. Thus the total sample size will be 221 Respondents for both questionnaire and interview.

3.5.2 Sampling Techniques

Three sampling techniques will be used to select the respondents as here below explained

3.5.2.1 Stratified Sampling

Stratified sampling is where the population is divided into strata (or subgroups) and a random sample is taken from each subgroup. A subgroup is a natural set of items. Subgroups might be based on company size, gender, or occupation (to name but a few). Stratified sampling is often used where there is a great deal of variation within a population. Its purpose is to ensure that every stratum is adequately represented. A sample for each of the study sites as a stratum was calculated.

3.5.2.2 Random Sampling

At each institute, the respondents for questionnaires will be selected using random sampling. The researchers’ list shall be requested from the Human Resources Office in each institute. The entire population of the researchers at each institute will be given a number code and the numbers will be put together in a bag and randomly select one by one with replacement until all the required sample at each site is selected. All those selected will then be contacted with a request to participate in the study by answering a questionnaire.

3.5.2.3 Purposive Sampling

Purposive sampling known as judgmental sampling is defined as selecting a relatively small number of respondents who can provide valuable information related to the research questions under examination (Teddlie and Tashakkori, 2009). The rationale for purposive sampling is its ability to enable the selection of informed persons who possess vital information, comprehensive enough to gain a better insight into the problem under study. Purposive sampling will be used to select key informants. These include Managers/Administrators of research, System Administrators/Data Officers, and Librarians/Medical Record Officers from each of the selected health institutes. The office-bearers in the identified categories above are assumed to be information-rich on issues related to research data practices within institutes by virtual of their roles, training, and skills. A list of office holders in those categories will be identified with the help of the Human Resources Office at each institute. Where there is more than one person in each category, the most senior will be selected. The selected respondents will then be contacted requested for an interview.

Methodology in research

3.6 Data collection methods and tools

Data will be collected using multiple data collection tools. The tools will include: questionnaires, interviews, and document reviews

3.6.1 Questionnaire

Questionnaires will be used to collect both quantitative and qualitative data for analyzing the RDM practices in health institutes. Questionnaires will be used to collect data from respondents who are currently enrolled as researchers at different professional levels at the institutes of study. These include senior researchers, middle and junior/Ph.D. student researchers. Researchers as respondents in this study are important since they play a critical role in the research data lifecycle. Questionnaires with options for selecting and measuring based on the Likert scale are considered appropriate for the study.

However, due to restrictions imposed by COVID-19, questionnaires have been designed in several formats to allow wider distribution to researchers and as a strategy to reduce the risk and improve the response rates. Thus, a print copy will be available at the office of the person responsible for research at each study site.  This will be accessible to identify researchers who may need a physical copy to fill. In addition, an electronic link will be sent via individual official electronic mail to all identified researchers with a request to answer the questionnaire prepared in Google form within a specified time of about 10 weeks with a weekly reminder if one has not already filled in the questionnaire. The questionnaire will be concurrently distributed to all study institutes. Thereafter all physically completed questionnaires shall be collected and input into the Google forms by the researcher and input into SPSS for analysis.

3.6.2 Interview

The interview is a data collection method in qualitative methods. It is where the respondents reply to questions asked by the interviewer verbal communicating and spoken narratives of insights constructing their social world (Ritchie, et al., 2013). This method will be used to collect verbal responses from respondents. Interviews will be conducted with Managers/Administrators of research, System Administrators/Data Officers, and Librarians/Medical Record Officers. Interviewing this category of respondents will reveal in-depth personal accounts and explore issues in detail about research data, its management, readiness, challenges experienced, and thereafter interviewees may make suggestions proposing a measure that could enhance adoption and uptake of RDM practices in health institutes and the readiness required to achieve this in compliance to funders and publishers requirement. Interviews shall be held at the respondents’ place of work or by telephone whatever is convenient to the interviewee. Interviews will be recorded on request and thereafter transcribed into textual data for analysis. The text will be assumed to be a replica of the verbal responses of the respondents.

 steps in writting research methodology

3.6.3 Document Review

Written documents are a pervasive socially constructed representation of reality in institutions (Patton, 2002). However, documents pose challenges among which are access to official documents, understanding how and why they were produced, and difficulty to determine their accuracy.  Nonetheless, as a data collection method, documents to be reviewed will include: legal and policy documents; strategic and annual plans; statistical and research-related reports of institutes understudy and collaborators, and oversight institutions’ reports. Despite the challenges, documents are an important source of data related to the institutional context that could be useful in understanding the RDM practices and readiness.  Therefore, document review will provide evidence of what is taking place reservee what is documented. Identified documents will be requested from the respective institute Administrators for review.

3.7 Data Collection Tools

Data will be collected using: Questionnaires (Appendix 1), Interview guides (Appendix 2a, 2b, and 2c and a Document review guide (Appendix 3).

phd research methodology

3.8 Data Quality Control

Data quality control are measures put in place to ensure data integrity and authenticity and to safeguard the quality of the research output.

3.8.1 Validity

Validity is how well an instrument measures what it is supposed to measure (Taherdoost, 2018). The goal of validity is to ensure accurate, objective, and neutral representation of the topic under study (Marwill and Rossman, 2011). A pre-test will be carried out to assess the face validity of the data collection tools to gauge the meaning and attributes of the questions both in interview and questionnaire (Ng ’ Eno, 2018). The pre-test shall also assess whether the data collection tools capture the information required for the study. It will also help to give confidence to the researcher in data collection and eliminate barriers such as resistance to recording and mistrust of the researcher’s agenda which shall consequently strengthen the study (Marwill and  Rossman, 2011).

3.8.2 Reliability

Reliability will be ensured by the data collection tools being subjected to respondents who were identified for the purpose. All researchers shall be subjected to a questionnaire across the study sites and interviews shall be conducted for pre-identified key informants in each institute. Finally, pre-identified documents shall be reviewed across the three institutes of the study.

3.9 Data presentation and analysis

Data shall be collected from each study site and analyzed. Thereafter, cross-case conclusions shall be made and the report written.  However, all collected data shall be under the custody of the researcher. Qualitative data will be presented using the themes derived from the interpretations analyzed by NVIVO software to give deeper research insights. The software will be used to create mind maps to quicken the process of analyzing qualitative data (Godau, 2014). Qualitative data will be presented in form of direct quotations from respondents explaining in detail their experiences and practices.

Quantitative data will be analyzed using Statistical Package for Social Scientists (SPSS) and presented using interactive statistical analysis.

writting research methodology

3.10 Ethical consideration

This research will be subjected to a series of ethical reviews; first, the proposal was submitted to the committee of Higher Degrees at the East African School of Library and Information Science for defense. Thereafter, the proposal was submitted to the Doctoral Committee at the College of Computing and Information Science, and finally, the proposal was submitted to the School of Health Sciences Research Ethical Committee, Makerere University for clearance.

The cleared proposal was submitted to the National Council of Science and Technology in fulfillment of the mandatory requirement for all Ph.D. studies in Uganda for clearance too.

The researcher protected all respondents’ confidentiality and privacy and health institutes by anonymizing the collected data such that it is not possible to identify the respondents and the institute in published texts and to allow free discussion of the data and the emerging issues from analyzing the collected data.

However, there are minimal risks directly associated with this study. The researcher will ensure compliance to Ministry of Health SOPs in all engagements with the respondents to protect all from the COVID-19 pandemic.

All the data collected for this study will be under the custody of the research, preserved digitally, and encrypted. However, it will be accessible to supervisors and examiners if they ask for it.

All respondents were given a refund token of 20,000/- to compensate for: time, data, and transport.

3.11 Gender consideration

The researcher gave equal opportunity for both male and female respondents to participate in the study in the course of collecting data.

3.12 Limitations of the study

The COVID-19 pandemic has disrupted nations and systems leading to the closure of institutions and nations across the world and has led to increased restrictions of access to Health Institutes. The current lockdown may deter full access to respondents and the required documents for the interview and review respectively. However, adherence to the new norms and SOPs put in place by the Ministry of Health is a must for the researcher and all respondents including sanitizing hands and tools regularly, wearing of face mask at all times, and social distancing at least two meters apart at the different study sites.

Where the researcher may find it hard to access the respondents physically for interviewing, an alternative scheduled telephone interview shall be arranged and if agreeable with the respondent, will be carried out.

The busy schedule of respondents particularly respondents who are working on different COVID-19 related projects may affect the response rate. However, the researcher in consultation with the Human Resources Officers of the respective research institutes will organize a two hours webinar for researchers as a means of creating awareness about RDM practices and the benefits of participating in the study geared at making proposals to enhance research data services. This it is anticipated will motivate researchers to answer the questionnaires. In addition, there will be a weekly email for about 10 weeks sent to respondents as a reminder asking them to the respondent to the questionnaire if they had not already done so. A print copy will also be available at a convenient place where the target respondents will easily access them for answering and returning.

methodology writting in research

The researcher is not sure whether the required documents for reviewing exist or are readily available within the selected Health Institutes since their existence is not necessarily guaranteed. The researcher will inquire about the existence of the required documents in the processes of asking for administrative clearance and from respondents during the interviews.

Researchers as respondents are usually very difficult to respond to data collection tools that usually target them. There are usually low response rates in such studies, to mitigate this challenge, the researcher shall request each of the study institute administration to allow him to conduct a 2 hours webinar via Google teams on “research data management”. This is anticipated to create awareness and thereafter request stakeholders to participate in the study for better RDM service planning.

3.13 Delimiters of the study

Despite the existence of different types of data in health institutes, the focus of the study will be limited to “research data” for health and biomedical research purposes only.

3.14 Mapping research questions to theoretical models

The table below shows the research questions, the theoretical model and key variables to focus on in the course of the research, the data analysis methods to be used, and the summary of the findings for each research question.

methodology writting in research