Unit of Analysis
Methodology & design
The 'what' being studied and about which conclusions are drawn — individuals, teams, organisations, transactions or documents — which must stay consistent from research question to sampling to statistics.
The unit of analysis is the entity your conclusions are actually about — the individual, the team, the firm, the country, or the document. It answers the question of what each row in your dataset represents, and it is decided at the design stage, not during analysis.
Getting it wrong produces the ecological fallacy, where findings about groups are wrongly applied to individuals, or the atomistic fallacy, where individual-level findings are wrongly generalised to groups. Both are serious inferential errors that examiners look for.
Complications arise when the unit of measurement differs from the unit of analysis — for example surveying individual employees to draw conclusions about their organisations. That requires justified aggregation, supported by inter-rater agreement statistics, or a multilevel model that handles the nesting properly. State your unit explicitly and keep every claim consistent with it.
Where it's used
- Aligning sampling and analysis levels in the methodology chapter
- Avoiding ecological fallacies when data and claims sit at different levels
Related concepts
Guides that use this concept
All guidesNot sure where to start?
Book a free 15-minute consult. We'll map your next three steps — no obligation.