Qualitative Research Methods Toolkit
What is Qualitative Research?
Qualitative research methods seek to answer questions such as the why and how of human behaviour, opinion, and experience. The basis of qualitative research lies in the interpretive approach to social inquiry and in the lived experience of human beings. Approaches to social inquiry consist of sampling procedures, data collection and analysis tools and world views of the researchers. Therefore, qualitative research must look into the appropriateness of methods and theories suited to the research question, the perspectives of the participants and their diversity and the reflexivity of the researcher and the research (Flick, 2009:14).
When addressing the question of why, it is useful to the follow the hints below:
Why
- Why is this research relevant within the context of the study/intervention?
- Why is this research being conducted?
- Why do the participants feel, behave or experience phenomena the way they do?
When addressing the question of who, it is useful to the follow the hints below:
Who
- Decide on which person(s) you want to interview (case sampling) i.e. head of household, traditional healers, sex workers, health workers, housewife, farm labourer, fisherman, community leaders
- Decide on which groups these should come from (sampling groups of cases) i.e. men, women, mixed gender, young people, children, family living in household [may be extended-not nuclear], age specific, specific collective, religious, ideological groups etc.
- Ensure principles of consent and confidentiality are assured by the investigator [this can be found in the OAD IAU Ethics Guidelines].
When addressing the question of what, it is useful to the follow the hints below:
What
- Decide on themes you want to investigate for the survey: for example, community knowledge, attitudes, perceptions and beliefs related to the research question and specific objectives of the research. This could be related to some examples stated below:
- social organisation;
- culture;
- migration;
- conditions of family life;
- access to and utilisation of health services
- household income
- literacy rates and level of education
- Themes need to be put into categories and coded to help with sorting and analysis. This occurs for the quantitative tool (structured questionnaire) before the survey and for the qualitative tools (those producing text for analysis) after the data has been collected.
- Decide on questions you want to ask under each theme chosen.
When addressing the question of how, it is useful to the follow the hints below:
How
Choose the methods/ tools you want to use such as:
- Case study
- Comparative study
- Sample survey (pre-selection of groups representing community and theme under scrutiny
- Purposive survey (pre-selection of specific professions, people, places, subjects, education level etc)
- Participant observation
Choose ways of extracting data either through:
- Individual interviewing (key informant, head of household)
- Structured or semi -structured questionnaire (including text responses)
- Focus groups/ natural group
- Film
- Audiovisual recordings
- Systematic observation
- Collection of documents (incomplete drawings)
- Stories (open ended)
Sample size
There are no hard and fast rules for social data. Determining an adequate sample size in qualitative research is ultimately a matter of judgement and experience dependant on the quality and purpose of data collected.
It is more important to differentiate between the case sampling and the groups from which the case sample/ or unit should come. For qualitative information the size of the sample is less important than the quality of the sample. It is the representative nature of the sample that is of greater importance than the size. An adequate sample is one that permits by virtue of not (being too large), the deep case oriented analysis that is the hallmark of qualitative inquiry. This results in a new and richly textured understanding of experience.
Sampling
The logic and power of purposeful sampling lies in selecting information-rich cases for study in depth. Information-rich cases are those from which one can learn a great deal about issues of central importance to the purpose of the research.
Triangulation
Triangulation refers to the process of utilising several methods (data sources, theories, or methodologies) in the study of one phenomenon.
There are a numerous types of triangulation methods that can be used to produce robust findings. The different types of triangulation methods are data triangulation, theoretical triangulation, methodological triangulation, analysis triangulation and investigator triangulation (Hussein, 2009:3). Some of them are discussed below.
Theoretical triangulation makes use of a number of theories within the same study to generate support for accepting or rejecting findings from multiple perspectives (Hussein, 2009:3).
Data triangulation comprises of the gathering of data through multiple sources of information to improve the validity of the study.
Methodological triangulation involves a number of methods in studying the same phenomenon under analysis where it can occur at various levels, namely, at data collection or research design level (Hussein, 2009:3).
Reliability and Validity
Reliability and validity are benchmarks against which the quality and accuracy of research findings can be measured to provide rigour and value to the findings.
Lincoln and Guba (1985) proposed four criteria for establishing reliability and validity in qualitative research which are: internal validity (truth-value), external validity (applicability), reliability (consistency) and objectivity (neutrality). For Lincoln and Guba, establishing trustworthiness is paramount as it reflects the accuracy with which the investigator interpreted the participant’s experiences. The table below summarises the criteria.
Criteria for Trustworthiness
Criteria |
Procedures |
Credibility (internal validity) |
Prolonged engagement, Continuous observation, Triangulation (p.301) Ensures that the study measures what it actually intended
|
Applicability/Transferability (external validity) |
Thick description (p. 316) A thick description is where the researcher thoroughly describes the context in which the data is collected and presents it with adequate detail and precision to allow judgments about transferability to be made by the reader |
Dependability (reliability) |
Triangulation of methods, Dependability audit- examining the process of the inquiry (how data was collected; how data was kept; accuracy of data) (p. 317) |
Confirmability (objectivity) |
Confirmability audit - examines the product to prove that the findings, interpretations & recommendations are supported by data (p. 318-320) Can also be attained by mitigating researcher bias, admission of the researcher’s assumptions and beliefs, admission of the limitations of the study |
Source: Lincoln and Guba. (1985). Naturalistic Inquiry.
Bias in Qualitative Research
Bias affects the validity and reliability of findings. One's personal history encompassing of one's class (which may or may not include changes over the course of a lifetime), one's race, one's gender, one's religion etc. all have profound influences on how we think and can constrain the production of knowledge. To avoid error and bias, qualitative research demands some measure of scepticism, commitment and detachment and to look beyond one’s own world views. There are various types of biases that one can encounter in research.
Researcher Bias
Researcher bias refers to when a researcher’s personal beliefs influence the way information or data is collected and analysed. It can also refer to the researcher’s natural disposition to look for data that confirms his/her hypotheses or confirms personal experience, overlooking data inconsistent with personal beliefs.
Selection Bias
Selection bias can refer to the process of recruiting participants and study inclusion criteria. Successful research begins with recruiting participants that meet the research objectives. For example, recruitment bias could occur if participants were invited to participate in a survey posted on the internet, which automatically excludes individuals without internet access.
Some methods to avoid bias in qualitative research:
- A random sampling procedure provides the highest assurance that those selected are a representative sample of the larger group that ensures that the researcher has no control over the choice of informants increasing the chances of diversity within the sample. However, this may not always be possible.
- Triangulation to reduce effect of investigator bias.
- Admission of researcher’s beliefs and assumptions.
- Recognition of shortcomings in study’s methods and their potential effect on the research.
References
Flick, U. 2009. An Introduction to Qualitative Research. Fourth edition. London: Sage Publication
Hussein, A. 2009. The use of Triangulation in Social Sciences Research: Can qualitative and quantitative methods be combined? Journal of Comparative Social Work, 1-12.
Lincoln, Y.S. and Guba, E.G.1985. Naturalistic Inquiry. Newbury Park, CA: Sage Publications.
Patton, M. 1990. Qualitative evaluation and research methods (pp. 169-186). Beverly Hills, CA: Sage Publication.
Shenton, A.K. 2004. Strategies for ensuring trustworthiness in qualitative research projects. Education for Information, 22:63–75.