Ranking question component - how are rankings calculated?

Ranking question components allow respondents to prioritise a set of items. For each item, the answers from all respondents need to be aggregated into a single number that can be used to determine the overall popularity of that item. This number is the item's ranking.

Citizen Space calculates the ranking of each item as follows:

First, a weight is assigned to each possible ranking position. If there are five positions for the respondent to choose from, the first place (number 1 in the dropdown list) will be given a weight of 5, the second place will be given a rank of 4, and so on.

The weighted average score for each item is then calculated as:

(w₁c₁+ w₂c₂+ w₃c₃+ ... + w_nc_n) / t

Where:

• w₁, w₂ ... w_n are the weightings of position 1, position 2, etc
• c₁, c₂ ... c_n are the number of respondents who chose position 1, number of respondents who chose position 2, etc
• t is the total number of respondents who answered the question.  A respondent is included in the total if they ranked any item in the question, even if they did not select a position for this particular item.

This calculation is repeated for each item to be ranked.