Survey Question Types
When Creating a Survey users have the option to add different question types.
This guide will look to explain the different question types available and provide an example of when they might be used.
Considering which question type to use for specific questions is important as this impacts how users analyse can analyse their results later. For example a large set of 'Text' responses might be more difficult to analyse than if participants had answered a 'scale' or 'single choice' question.
Another consideration might be that organisations plan to Compare Survey responses to questions over time (e.g. Baseline / Follow Up) and so certain quantifiable question types need to be used and made required, such as Whole number, Scale, Yes/No, Single choice, Multiple choice, Count and Matrix questions.
Finally, it can be worth considering whether to create Standardised Questions. This allows organisations to create a 'bank' of questions that can be selected from when creating new surveys. These also have additional reporting capability.
Organisations will also find some pre-built Standardised Questions available under the heading of Upshot Global Questions. These questions come from validated and accredited surveys from the sector and can help ensure a consistent approach to measuring impact, rather than the need for organisations to create their own custom questions.
We are always happy to support organisations with creating their surveys on Upshot, so please do contact support@upshot.org.uk if you need any support.
- Text
- Whole number
- Decimal number
- Scale
- Yes/No
- Single choice
- Multiple choice
- Count
- Date
- Matrix
- Conditional Questions
Text
This type of question allows a respondent to give a free text response, helpful for questions that require added detail or to explain a previous answer. These are a powerful tool for qualitative analysis purposes, allowing the respondent to provide in-depth feedback.
E.g. - 'Explain how you feel the session could be improved?'
Whole number
This allows a respondent to state how many, or how much, of something there was.
E.g. - 'How many times a week do you eat chocolate?'
Decimal number answer
This allows a respondent to state how many, or how much, of something there was but also include up to two decimal points in their response.
E.g. - 'Please specify the weight (KG): '
Note: A respondent could still input a 'whole number' in their response here as well as a number with decimals. E.g. the answer to the above could be 18.00 or 18.25.
Scale
This allows respondent to give a rating/score based on their thoughts and feelings, or perhaps to illustrate how many times a week they completed something.
The scale can be set to between 0-10.
Scale questions can be particularly helpful for analysis purposes, allowing users to later work out an 'average score' for a particular question.
This is why in some instances, setting up a questions with a range of single-choice options such as 'Strongly Agree/Strongly Disagree', might work better as a scale question, as this would be easier to analyse. In these instances it would make most sense to set up the most positive answer 'Strongly Agree' as the highest score on your scale.
It is often helpful to provide additional help text to explain the scoring system on the scale.
E.g. - 'How would you rate today's session?'
'How many times a week do you exercise?'
'To what extent do you agree with the statement that 'people from your local area can be trusted''
Yes/No
This can be used for definitive opinion questions and are often used in conjunction with follow up text questions to allow an explanation for the response given.
E.g. 'Do you plan on attending the session again?'
Single choice
This allows respondents to state a preference out of some pre-determined response options. The theme/content of questions where single choice would be appropriate can really vary, but they can be a key way of allowing organisations to quantify responses around a set of predetermined options.
E.g. - 'What do you prefer to drink?'
'How often do you use public transport?'
Multiple choice
This allows respondents to choose multiple responses in relation to a question, allowing users to analyse how many respondents selected certain options.
E.g. 'Which of the following have you heard of?'
Count
This allows respondents to state the amount of time they completed certain actions in response to the options presented.
In instances like the example below, this can save users from having to create separate questions for all of the relevant options.
E.g. - 'How many times did you use each of these methods of transport to get to school in the last week?'
Date
This allows respondents to add a date. This can be helpful if the date of the event they attended was different to the date they are completing the survey, which will automatically be generated in the results.
E.g. - 'What date did you attend?'
Matrix
Matrix questions allow organisations to present questions that have the same set of responses to a series of statements, such as 'Strongly Agree/Strongly Disagree', in a grid format, which can help to shorten the length of the survey.
'Options' (columns) may represent answers such as 'Strongly Disagree - Strongly Agree' whilst the 'Rows' would represent a series of statements or criteria.
The matrix can be single or multiple choice, by default this will be single-choice but users can select 'Allow multiple responses per row?' when creating.
Questions can also be assigned a numeric value or ' weight' that can aid analysis later, by treating these similar to scale questions. When creating, the user can select 'Assign numeric value for answers?'
Please note that survey respondents would not see these 'numeric values', they would just be shown to Upshot users on screen and be applied in the background to aid users analysis.
E.g. - 'Please select the option that best describes your experience of each over the last two weeks.'
'For the following mobile phones, which features do you like?'
Conditional Questions
Users can choose to make survey questions ‘conditional.’
This means that questions are only displayed based on a participant’s response to a previous question.
This can simplify the survey for respondents, asking relevant extra detail of those that answer in a certain way and streamlining the survey for those that do not.
In the below example an organisation creates a conditional question in Q2, that asks users to give more feedback only if they have answered 'Q1 – How did you find today’s session?' with ‘Very good’.
The Display when . . . in the bottom right-hand corner sets the condition for when this question will appear.
Key Rules
- Conditions can be set based on previous responses to 'Whole number, Scale, Yes/No, Single and Multiple-choice questions.'
- A conditional question must always be placed after the original question it is conditional on. E.g. Q2 must always come after Q1 above.
- The conditional question does not have to immediately follow the question it is conditional on, it just has to be after. E.g. Q4 can be conditional on Q1.
- Conditional questions can be made required if desired.
- Multiple conditional questions can be dependent on the same original question.
- Surveys can have multiple conditional questions dependent on another. E.g. Q2 can be dependent on the response to Q1, and Q3 can be further dependent on the response to Q2. See example below:
- When exporting survey results the conditional questions will always appear in the download.
- When Upshot users view the survey in the system, complete a survey response on a participant's behalf or use the Printable Questions option, the condition will be written out below the relevant question.
