Exercise 8. Sorting

Name of exercise: Sorting

Tools: Pen, cards, objects (or names written on cards) to be sorted

Time needed: 1 – 2 hours

Description: Sorting means dividing objects or ideas into groups and subgroups, and then exploring the characteristics of each resulting set or subset. The technique allows participants to make explicit their reasons for clustering ideas and objects in specific ways.

How to do it: The items to be compared (and the domain to be explored) should be chosen through a process of free-listing, participant observation, group discussion or semi-structured interviews. If possible use real objects for the sorting exercise, or record the local terms for the objects or ideas on cards. Randomize the objects or cards by mixing or shuffling them each time before asking someone to sort them. Give the cards or objects to participants, who will sort them into groups freely, or according to specific criteria suggested by the interviewer. Respondents can carry out a dichotomous sort (dividing into two groups, then each of these groups into two subgroups, and so on) or a free sort (dividing into as many groups or levels as they wish). The sort can be exhaustive – continued until all objects or ideas are by themselves – or partial, leaving some items clustered in subsets. The interviewer or respondents can record the members in each group or subgroup as the sorting proceeds. Once the sorting is complete, ask participants to list the characteristics of the groups and subgroups, and their reasons for dividing the objects in the way that they did. If the technique is carried out by numerous individuals, then the results can be recorded in a matrix like the one shown below, assigning one point when objects are together in a group or subgroup (and 0 points if the two objects are not together). This approach summarizes the consensus or majority view about which objects go together. Alternately, the sorting can be carried out by a group of participants who together discuss the characteristics of the items and how they should be grouped together.

References:

IIRR. 1996. Recording and Using Indigenous Knowledge: A Manual. Silang, Cavite, International Institute of Rural Reconstruction. Pages 95 – 96, "Sorting and ranking".

Martin, G.J. 1995. Ethnobotany: A Methods Manual. London, Chapman and Hall. Chapter on anthropology.

Example: The fifth year students of the Department of Pharmacognosy and Pharmaceutical Botany of Khon Kaen University experimented with sorting by dividing 18 Thai fruits into groups according to overall similarity (participants were left to discuss what this meant). After the sorting was completed, each student explained his or her own criteria for grouping the fruits. The results were tallied in the following matrix, which gives an impression of the majority view on which fruits go together. For example, pear and apple (both in the Rosaceae) were grouped by 22 of the students, langsart and longong (both in the Meliaceae) by 24, and orange and pomelo (both in the genus Citrus in the Rutaceae) by 18. On the other hand, nobody considered that there was any similarity between some fruits (papaya and grape; pineapple and longong or grape, etc.). The students then split into groups to come up with explanations for the cumulative results; two explanations are presented below.

BACK