Sprint #20

November 10 → 24

Automatic scripting of questionnaires: the AI now imports an identifier for each question, and it is this identifier that is used to import the filters. Also, if a variable name is not specified in Word, the AI will create one automatically and consistently.

Variable names: to ensure the proper functioning of the various types of comparison, they now become mandatory.
Field monitoring: the interview date filter is now based on the start date of the interview (and no longer the end date).
Emailing: the task of importing a new contact list has been reconfigured to ensure that it no longer stops, even when it is very long.

Powerpoint report: when a crosstab containing a coded question was displayed as a table, the “non-answer” item could be incorrectly formatted.
PowerPoint report: it was not possible to generate a graph comparing several codified multiple open-ended questions.
PowerPoint report: after codification, the word cloud of an open-ended question displayed codes instead of verbatims.
PowerPoint report: the basic option (raw or restated) was not taken into account in the cross charts.
Codification: verbatims with inconsistent names (“Checkbox xxx”) could slip into the codification.
Codification: the page did not load properly after moving the verbatims, giving an impression of being very slow.
AI codification: on multilingual studies, AI could encounter difficulties in proposing codes.

Automatic scripting of questionnaires: import of screen-outs and recodes.
Redesign of the study comparison: work is in progress, the first brick of this redesign should be put into production during the next sprint.
New questionnaire interface: the work is still ongoing, with the evolution of the questionnaire editor. On this project, everything will be put into production at once, towards the end of the quarter ⏳.