Lecture note on research methods design quality criteria: for Housing Studies students

Lecture note on research methods design quality criteria: for Housing Studies students

 

Describe three qualitative research methods design quality criteria and three quantitative research methods design quality criteria in the context of primary research methods used in part-time Housing Studies dissertation projects.

In housing studies dissertations, quality criteria differ between qualitative and quantitative paradigms because they address different notions of “good evidence.” Below are three core criteria for each, explained in the context of primary research on housing (e.g., interviews with residents, surveys of housing conditions, policy implementation studies).

Three qualitative design quality criteria

1. Credibility (trustworthiness of findings)

Credibility is the qualitative analogue of internal validity: it asks whether the findings accurately reflect participants’ experiences and the phenomenon studied (e.g., how public housing residents perceive recent regeneration policies).hopkinsmedicine+1

In a part-time Housing Studies dissertation, you might enhance credibility by:

• Using triangulation (e.g., combining interviews, focus groups, and policy documents).
• Conducting member checking (sharing summaries or themes with participants to confirm accuracy).
• Engaging in prolonged engagement with the field (e.g., multiple site visits to a housing estate over time).hopkinsmedicine

2. Transferability (applicability to other contexts)

Transferability corresponds loosely to external validity: it concerns whether findings can be meaningfully applied or “transferred” to other settings, not statistically generalised.hopkinsmedicine

For housing research, this often means:

• Providing thick description of the context (location, housing type, policy environment, demographic profile).
• Clearly describing your sampling strategy (e.g., maximum-variation sampling across different public housing blocks in Hong Kong or Singapore).
• Discussing how your findings resonate with or differ from existing housing literature in other cities.hopkinsmedicine

A reader can then judge whether your insights about, say, stigma in subsidised housing might transfer to similar estates elsewhere.

3. Confirmability (neutrality from researcher bias)

Confirmability addresses whether findings are grounded in the data rather than the researcher’s preconceptions—particularly important when studying sensitive housing topics (eviction, informality, discrimination).ncbi.nlm.nih+1

You can demonstrate confirmability by:

• Keeping an audit trail of decisions (how you selected cases, coded transcripts, refined themes).
• Practising reflexivity (a research diary noting how your position—e.g., as a practitioner or resident—might shape interpretation).
• Using peer debriefing (discussing emerging interpretations with supervisors or colleagues).hopkinsmedicine

Together, credibility, transferability, and confirmability form a coherent framework for judging the rigour of qualitative primary research in housing studies.ncbi.nlm.nih+1

Three quantitative design quality criteria

1. Internal validity (causal credibility)

Internal validity asks whether observed relationships (e.g., between housing quality and health outcomes) can be attributed to the variables of interest rather than confounding factors.ncbi.nlm.nih+1

In a part-time dissertation, you might strengthen internal validity by:

• Using clear operational definitions (e.g., specifying how “housing quality” is measured—objective inspection scores vs self-reported indices).
• Controlling for confounders in analysis (income, household size, tenure).
• Choosing designs that reduce bias (e.g., longitudinal panels rather than one-off cross-sectional snapshots where possible).hopkinsmedicine

2. External validity (generalisability)

External validity concerns the extent to which results from your sample can be generalised to a broader population or other contexts (e.g., from one district’s public rental housing to the city-wide stock).ncbi.nlm.nih+1

Practical steps include:

• Employing random or stratified sampling where feasible (e.g., sampling households across different housing types and districts).
• Clearly describing the population and setting so readers can judge similarity to other contexts.
• Replicating analyses in sub-samples or comparing with existing large-scale housing surveys.hopkinsmedicine

Even with convenience samples (common in part-time projects), you can discuss limits to generalisability transparently and situate findings within wider housing datasets.

3. Reliability (consistency of measurement)

Reliability refers to the consistency and stability of measurements if the study were repeated under similar conditions.ncbi.nlm.nih+1

In housing research, this might involve:

• Using standardised instruments (e.g., validated housing quality or satisfaction scales).
• Reporting internal consistency (e.g., Cronbach’s alpha for multi-item scales).
• Ensuring standardised data collection procedures (training enumerators; clear protocols for physical inspections).hopkinsmedicine

High reliability ensures that observed patterns are not artefacts of measurement error—critical when your dissertation informs policy-relevant conclusions about housing conditions or needs.

 

A collection of blog notes on using chatgpt for research purpose.