Survey-native data structures

Survey data is more than a flat table.

Multiple-response questions, Likert scales, categorical and ordinal types, grouped variables, nets, and survey metadata are handled as research data. Multi-selects, grids, weights, and multi-wave studies are ready to inspect as the survey structures they are.

Keep the meaning of the questionnaire intact.

AddMaple groups related grid questions, detects multi-select splits, respects value labels from SPSS, and treats opinion scales as ordered data for analysis. Schema files can guide column matching for complex survey exports, while nets and derived groupings stay connected to their source variables.

Let the analysis work the way researchers expect.

Generic spreadsheets break multi-selects, ignore scale order, and force manual relabelling before the work begins. With the structure intact, banners, significance testing, clustering, and key-driver analysis can start from the survey rather than a preprocessing script.

Preserve context for the next analyst.

Survey metadata is part of the evidence, not decoration around it. Keeping labels, grouped questions, and ordered scales together makes the project easier to review and reduces the risk that a later analyst has to infer what a column meant from its values alone. From there, the team can create scores or reshape the data while retaining the original structure.