Pile Sorting
Also known as: Pile Sort Task, Free Pile Sort, Card Sorting (ethnographic), Sorting Task
Pile sorting is an elicitation technique in which informants are handed a set of cards — one per item in a cultural domain — and asked to group them into piles of items that 'go together.' By recording which items each person places in the same pile and aggregating across many informants, the researcher builds a similarity matrix that reveals how the culture organizes the domain, which is then visualized with multidimensional scaling and clustering.
Key highlights
- Collects similarity judgments among many items very efficiently — far faster than rating all pairs or triads.
- Intuitive and low-burden for informants, including those with limited literacy when picture cards are used.
- Aggregates naturally into a similarity matrix that feeds directly into MDS and clustering for visualization.
- Reveals the emic category structure of a domain, complementing the inventory produced by free listing.
Intuition
This section is available to Pro members. Upgrade to Pro
How it works
This section is available to Pro members. Upgrade to Pro
When to use it
Use pile sorting when you have a defined set of domain items (often from a prior free list) and want to recover how informants perceive the similarity and grouping structure among them. It is excellent for moderate-to-large item sets (roughly 15–60 items) where exhaustive paired judgments would be impractical. It assumes items are concrete enough to be represented on cards and comparable on some basis. It is less suitable for very small item sets (where triad tests give finer resolution), for purely verbal or abstract items hard to put on cards, or when you need the specific reason behind each grouping rather than the aggregate structure.
Strengths & limitations
- Collects similarity judgments among many items very efficiently — far faster than rating all pairs or triads.
- Intuitive and low-burden for informants, including those with limited literacy when picture cards are used.
- Aggregates naturally into a similarity matrix that feeds directly into MDS and clustering for visualization.
- Reveals the emic category structure of a domain, complementing the inventory produced by free listing.
- A single free sort yields only binary same-pile/different-pile data, so individual-level similarity is coarse.
- Informants differ in how finely they split piles ('splitters' vs. 'lumpers'), which can confound the aggregate similarity.
- It records that items were grouped, not why, so the meaning of each cluster must be probed separately.
- Results depend on the item set presented; omitting key items or including ambiguous ones distorts the recovered structure.
Common pitfalls
This section is available to Pro members. Upgrade to Pro
Applications
This section is available to Pro members. Upgrade to Pro
Frequently asked
How does pile sorting differ from a triad test?
Both elicit perceived similarity, but they trade off resolution against scale. A pile sort lets one informant handle many items quickly but yields only coarse binary same-pile judgments. A triad test asks which of three items is most different and gives finer, fully ranked similarity data, but the number of triads grows rapidly with item count, so triads suit small domains and pile sorts suit larger ones.
Should I use a free sort or a constrained sort?
Free (unconstrained) sorts are most common because they let informants reveal their own number of categories, which is itself informative. Constrained sorts (a fixed number of piles) standardize granularity across informants and ease comparison, at the cost of imposing structure. Successive pile sorts add hierarchy by recording how informants split or merge piles, recovering nested categories.
How is the similarity matrix analyzed?
The aggregate same-pile proportions form a similarity matrix that is converted to dissimilarities and analyzed with non-metric multidimensional scaling to display items in a spatial map, and with hierarchical clustering to recover discrete categories. Property fitting or interpretation of the MDS axes then links the spatial dimensions to underlying cultural attributes.
Sources
- 1.Weller, S. C., & Romney, A. K. (1988). Systematic Data Collection. Qualitative Research Methods Series 10. Newbury Park, CA: Sage.ISBN 9780803930742
- 2.Borgatti, S. P. (1994). Cultural domain analysis. Journal of Quantitative Anthropology, 4(4), 261–278.
You have read it. What now?
Cite this page
ScholarGate. (2026, June 22). Pile Sorting. ScholarGate. https://scholargate.app/anthropology/pile-sorting