SOAD2021 Chap.5 Qualitative Research and Evaluation
Qualitative Research and Evaluation
Qualitative inquiry examines meaning, experience, practice and process in context. A strong design begins with a question that needs participants' accounts, interactions, documents or observed settings rather than treating interviews as a default.
Method choice should follow the kind of knowledge required and the relationships through which it can be produced.
Semi-structured interviews permit consistent topic coverage and responsive probing. Focus groups use interaction as data but require attention to confidentiality and group power.
Observation and learning walks reveal practice and environment, while documents show institutional accounts rather than direct experience. Sampling seeks information-rich variation, not statistical representativeness.
Analysis is an auditable movement from records to interpretation.
Familiarisation, coding, pattern development and review should be accompanied by memos that explain why codes changed and how counterexamples were handled. Quotations illustrate an interpretation but do not establish it alone.
Credibility grows through reflexivity, transparent decisions, attention to divergent accounts and conclusions bounded by the sample and setting.
Quality begins in the relationship between question and setting. A researcher studying why people disengage may hear different accounts if recruitment occurs through the service that participants left.
Sampling and recruitment therefore shape what can be known, not merely who appears in a participant table. During data collection, prompts should invite concrete events before abstract judgement, and field notes should distinguish observation from interpretation.
During analysis, compare material within and across participants, actively search for instances that complicate a theme, and document how the researcher's standpoint enters the reading. A theme is not a topic label repeated by several people; it is a patterned meaning that answers the question.
Reporting must show enough analytic reasoning for a reader to understand the move from extract to claim while protecting participants from identification. End with implications proportional to the setting and variation actually examined.
What this chapter covers
- 01
Qualitative questions
- 02
Semi-structured interviews
- 03
Focus groups
- 04
Observation
- 05
Purposive sampling
- 06
Coding
- 07
Reflexive interpretation
Worked example · free
Worked example: improve an interview guide
- 1Remove the assumed positive experience and ask: Tell me about what happened when you were referred.
- 1Add a sequence probe: What happened next?
- 1Add a meaning probe: What made that part feel easy or difficult for you?
- 1Add a counter-space: Was there anything different from what you expected or anything else that mattered?
Key terms
- Purposive sampling
- Purposive sampling is the deliberate selection of participants or materials because they can illuminate the question. Selection criteria should reflect relevant variation and be stated transparently.
- Coding memo
- A coding memo is an analytic record of developing meanings, decisions, relationships and questions during analysis. It creates a trail from coded material to interpretation.
- Divergent case
- A divergent case is an account or observation that does not fit the dominant pattern. Examining it can refine, qualify or overturn an emerging interpretation.
Qualitative Research and Evaluation FAQ
How many interviews are enough?
There is no universal number. Adequacy depends on the question, sample specificity, variation sought, depth of accounts, analytic approach and available resources. Justify why the material can illuminate the claim, and report limits instead of using a numerical rule as a substitute for design reasoning.
Are focus groups just efficient group interviews?
No. Their distinctive evidence includes interaction, agreement, disagreement and shared language. The group can also silence or expose participants, so topic sensitivity, existing relationships, facilitation and confidentiality require explicit consideration before efficiency is treated as a benefit.
Does coding make analysis objective?
Coding makes analytic handling more visible, but codes still reflect questions, concepts and researcher judgement. Credibility comes from clear definitions, comparison across material, reflexive memos, attention to divergent cases and a traceable explanation of how patterns became conclusions.
When should I use observation?
Use observation when practice, interaction, environment or sequence matters and self-report alone cannot answer the question. Specify the observer role, recording method, consent process and interpretive limit. What is visible in one setting may not represent private experience or other times.
Assessment move
Draft five interview questions for one evaluation purpose and label each as opening, descriptive, sequence, meaning or closing. Remove double questions and assumptions, then practise three neutral probes. Code a short paragraph twice: first descriptively and then interpretively. Write a memo explaining the change and locate one phrase that challenges your emerging pattern.
Review by matching each method to a question it can answer and a question it cannot. End by writing a theme claim, locating its best counterexample and qualifying the claim without reducing the counterexample to an exception. Check whether the supporting quotation adds evidence or merely repeats the theme label, and replace it if necessary.
