Manuals

Title Usable respondents (surveys)
Vsebina

 

The 'DASHBOARD' – 'Usable respondents' tab includes data about the quality of responses (variables) according to the record number of the respondent. Based on this data, key information is calculated – the usability of each questionnaire. Respondents and questionnaires are divided into three categories, with the rows of the table appearing in different colours:

 

  • Fully usable (white colour): questionnaires where the respondents answered all or almost all questions (more than 80% of the responses were usable).
  • Partly usable (yellow colour): questionnaires with a usability rate of 50% to 80%.
  • Unusable (red colour): empty questionnaires and questionnaires with a usability rate of under 50%.

 

The scope of the aforementioned categories can be customised, by setting the 'Bottom limit (unusable)' and the 'Top limit (usable)' above the data table. These two values are set by default to 50% and 80%. This means that questionnaires with a usability level of less than 50% are marked as unusable, questionnaires between 50% and 80% are marked as partly usable, and questionnaires with over 80% are marked as fully usable.

 

What determines the usability of a questionnaire? In addition to the above-mentioned criteria, the data table also includes other criteria, based on responses, non-responses and missing values in questionnaires:

 

  • Usable: the share of usable answers is calculated based on the number of questions the respondent answered and the number of questions the respondent skipped (non-response). Substantive and non-substantive answers (e.g. -96: none of the above, -97: inappropriate, etc.) are also taken into account. If the respondent could not answer because a condition (IF) was set and the question was not shown, this does not reduce the proportion of usable responses.
  • Valid: in the calculation of the proportion of valid responses, the number of questions the respondent answered is also taken into account. The key difference is that non-substantive answers are counted as valid.
  • Non-substantive: in the case of non-substantive answers, the respondent answered the question, but he chose a non-substantive answer, e.g. inappropriate (-97).
  • Non-response: the share of questions the respondent did not answer and which were displayed according to the (IF) conditions.
  • Break-offs: the share of questions the respondent did not answer, because the respondent left the questionnaire half-way through the survey.

 

The above described usability of the respondents, therefore, enables a more precise analysis of the status of units that are displayed in the ‘DATA’ – ‘Browse’ tab, after you turn on paradata, and separate the respondents into one the following groups:

 

  • Fully completed questionnaire (status 6 – finished survey)
  • Break-off during completion (status 5 – partially completed)
  • Initial break-off (status 3 – click on address, status 4 – click on survey)
  • Empty units (status 5l – partially empty, status 6l – completely empty)

 

Related content: 

Edit date 19.05.2015
HelpUser guide
Manuals
Print

Manuals (117)