Usecases

Surveys Construction

A question has a title, a text, some scales and some items. Scale/item pairs can be named or anonymous and indicate in inversion of the used scale.

A survey is a list of items, among which are headers, texts, images and named or anonymous questions. The same question may appear multiple times.

Questions can be selected from a pre-existing list.

Types of questions

• Free text questions, 'finish the sentence ...' questions
• Free numerical/date/email/phone entry (possibly within range)
• Yes/No?, Agree/Disagree?, Checkbox+Description
• ++/+/0/-/--, labels for every option or at extremes, values 0...n or -m .. 0 .. m
• Ordering questions (order several items, either by giving them unique values or really ordering them)
• Fixed sum questions (divide amount n over several items)

Conducting Surveys

• Personalized questionnaire. A limited amount of surveys is conducted, instantiated in advance. They can have e.g. unique response URL's.
• Bulk questionnaire. Multiple respondents can fill in one questionnaire.

• A survey can be conducted several times (e.g. different response groups, different year)

• A facilitator can add properties/answers to a survey instance. Any metadata that is relevant for later understanding of the data. E.g. Veerkracht, instantiate a bulkQ for every manager, adding him as metadata. The link for that Q can be send to his lackies.
• Continuate a Q started earlier. Use cookies, generate a continuation URL that people can bookmark or something like that.

Reporting and data export

Data can be reported or exported for use in a statistics program. Anonymous variables will get random names. Together with an export some description is generated, indicating the questions and the codification of the answers.

It should be possible to select a 2D range and show it in a table, or a 1D range and show it in graph. The type of scale can guide the user in available operations on the data.

Data unification

An important goal is that survey data and game data are expressed in the same way, so their data series can be used together in statistics and reports. For RDF the Data Cube vocabulary seems fit for that. The system should be able to adapt for different structured game data, unknown in advance.

Data stores

Data is sometimes collected on tablets, offline sessions etc. Data can be easily merged into the general data store after a session or even during a session?