Ranking helps assess the relative priority of issues/needs/problems or the solutions to these problems (i.e. it can be completed before or after action planning). It explores the reasons people make specific choices and can be used to compare the preferences of different individuals/groups in a community.
a) choose a theme/issue b) ask participants to identify the most important problems (or solutions) related to the theme (this can be done in advance of the session) and c) produce a preference matrix which has identical alternatives across the top (x-axis) and down the left side (y-axis) – see below. Each cell represents a comparison of two items or alternatives.
Pairwise ranking matrix: ranking of reasons for girls not attending school in a village in Gambia. Source: RIETBERGEN-MCCRACKEN et al. (1998)
Ranking can be completed by individuals by write or drawing each problem/preference on a card and then asking the interviewee to identify the biggest problem/best solution from a pair, giving reasons for the choice. A second pair is then considered and this is repeated until all possible combinations have been considered.
The problems can then be listed in the order in which the interviewee has ranked them i.e. how many times each problem was judged to be bigger than the other problems. The activity is then repeated with other individuals and the results can be used to begin a discussion about possible solutions to the priority problems
Ranking can also be completed in pairs or groups (through consensus) by taking each cell in turn and asking participants collectively which is most important. Once there is agreement/consensus write the most important problem in the cell and repeat until all problems or solutions have been compared against each other.
Count how many times each problem appears - the one that appears most often is the most important for this group of people (see example below).
Table from problem ranking of reasons for girls not attending school in a village in Gambia. Source: RIETBERGEN-MCCRACKEN et al. (1998)