MerQuR+Statistics Glossary

Statistics Glossary → Multiple-response set (MR set)

Multiple-response set (MR set)concept

A set of columns for a question where respondents can tick several options, with each option stored in its own column and the set treated as one variable in the analysis.

Questions such as 'Which social media apps do you use? (tick all that apply)' cannot be stored in a single column. The usual layout is one binary (selected/not selected) column per option; alternatively, answers can be kept in a delimited text column or in slot columns holding category names, such as 'first choice, second choice'. An MR set groups these columns into one question.

Frequency tables give two percentages: the percentage of responses (an option's share of all ticks, summing to 100%) and the percentage of cases (the share of respondents who ticked it, which can sum to more than 100%). Because one respondent contributes to several cells, the observations are not independent; so in MR × categorical tables the ordinary chi-square test cannot be applied directly to the whole table, and special tests such as simultaneous marginal independence, or adjusted option-by-option tests, are used. Comparing selection rates between options involves matched binary data and is done with Cochran's Q.

An everyday example

A café asks 200 customers which drinks they like, and each can tick as many as they want. There are 480 ticks in total, and 'filter coffee' is ticked 120 times: its percentage of responses is 120/480 = 25%, its percentage of cases 120/200 = 60%. For the owner, the second is usually the meaningful one: 60% of customers like filter coffee.

In MerQur

Offered as 'MR Set' in MR frequency analysis, the MR × categorical crosstab and Cochran's Q test, and as 'MR Set 1 (row)' and 'MR Set 2 (column)' in the MR × MR crosstab. A set is built in the Data > Multiple Response panel by grouping binary columns, splitting a delimited text column or grouping slot columns.

Analyses where this concept appears

MR Frequency AnalysisMR × Categorical CrosstabMR × MR CrosstabCochran's Q Test

Related concepts

Percent of cases and percent of responsesMultiple marginal independence tests (MMI, SPMI)