|
<< Click to Display Table of Contents >> Navigation: Reference Manual > Questions > Responses |
A response is one of the list of possible answers to a single-coded or multi-coded question. The list of possible answers is called a "response list". See Entry details window.
In this documentation "Response" always refers to the answer to a single-coded or multi-coded question. "Answer" is used for all other question types.
Each response has:
•Response number - this is the number stored in the data file unless individual response locations are used
•Response text - this is the text that identifies this response
The response number is usually used in input to select the response.
There are many other optional settings for individual responses:
The text for a response will be shown alongside the response figures in tables.
Response texts can also contain headers and over-headers.
To include a header you add the header text to the front of the first relevant response text, followed by \. For example, you might use: Gender\Male.
When used as the rows a header will be shown as a separate row above the relevant responses.
When used as the columns a header will be centred above all the following columns until the next response with a header.
Headers are also used to control how the significance testing is done, see Significance formats. Using column identifiers only columns under the same header are compared with each other.
Here is a complete example showing 3 response texts including a header:
Age\16-30
31-50
51+
If responses have individual response locations then a data location and code is allocated to each response.
The program will allocate these automatically, but you can set them yourself, see Matching data locations.
If individual response locations are not set then the response number is what is stored in the data.
One response can be marked as a reject response.
A question with a reject response will never be empty because if there is no answer then the reject response will be selected.
An "embedded reject" is a normal response that will be used as the reject response. This means that this will be the selected response if no other response is chosen.
A "supplemental reject" response must be the last response in the response list. This response is not stored in the data file, but will be selected for analysis if no other response is selected.
Any response marked as "refuse input" is not a real response and will not be accepted as an answer to the question.
Such responses are place holders and are used to reserve responses for possible later use or to allow the last few responses to always be used in the same way. For example, if you always want 98 to be "Don't know" and 99 to be "No answer given", you can allocate all 99 responses and mark any unused responses as "refuse", see Reserve responses.
A filter entry can be applied to a response. The response will only be available if the applied filter is true.
Only one filter entry can be applied to each response.
A response can be marked as:
•Ineligible - any respondent selecting this response is not needed for this project, no more questions will be asked for such a respondent
•Close - no more questions will be asked for any respondent selecting this response
•Question routing - this has no effect, but the question selected will be printed alongside the response in a printed questionnaire
•Text routing - this has no effect, but the chosen text will be printed alongside the response in a printed questionnaire
Often there are a few responses at the end of a response list that are not part of the main list. This will include responses like "None", "Not applicable", "All of them", "Refused to answer", and reject responses. These settings should be used for these:
•Single response only - if this response is selected in a multi-coded question, no other responses can be selected
•Fixed position - if the responses are shown in a random order, do not move this response (keep it at the end)
•Ignore response restrictor - if response filters are applied to the question they will not prevent this response from being available
A score value is a number associated with the response that is used in mean score calculations, see Response analysis settings tool window. For example:
Response number |
Text |
Score value |
1 |
Much better |
2 |
2 |
A bit better |
1 |
3 |
The same |
0 |
4 |
A bit worse |
-1 |
5 |
A lot worse |
-2 |
6 |
Don't know |
|
The last response is not allocated a score value and will not be included in any mean score (average) calculations.
Other analysis settings that you may also want to set:
•You can mark a response to start and another response to end ranking, during analysis these responses and any in between them will be sorted so that the response with the highest count appears first, and the lowest appears last
•Zero suppress - in analysis do not show this response if it is empty
•New page - start a new page (overflow) for this response
•Base for following - this response count is to be used as the base for percentages on the remainder of the responses