Crowdsourcing the survey analysis?

By the end of next week we will have had over 1000 responses to our survey about the future of the Carriageworks and Westmoreland House.

An idea for analysing it is to do a bit of crowdsourcing – i.e. making the data openly available for interested people to analyse.

So some questions:

Do you think this would work? What guidelines (if any) should we provide? What online tools might help? Should it be open to anyone or to an invited list of contributors?

If you have answers or other thoughts, write them in below.

For information: the data is a mixture of quantitative (e.g. select preferred options) and qualitative (open questions, text answers). It’s all been collected in Survey Monkey and will be downloaded as raw data.

3 thoughts on “Crowdsourcing the survey analysis?

  1. Crowdsourcing is best analysed by a selected group who are briefed to conduct the analysis in the same way. Quantitative data is fairly simply analysed by a group of analysts but qualitative data is a more protracted process and often calls for opinion forming. Therefore it is best carried out by a single analyst who maintains a singular thought process.

  2. In order to maintain integrity of the analysis it should be carried out by people (person) who understand at least basic statistical processes and can apply them.
    As Richard says, qualitative data usally takes longer and needs care to interpret and pull out relevant themes.
    Even quantitative data analysis is not necesarily a simple matter and you need to be mindful that the group of respondents are self-selecting.
    The analysis needs to be presented with more than just the results, but an explanation of the method used to make the analysis. This will allow for peer-reviewing and further analyses to take place.
    I migt be able to help, depending on timing…

  3. Thanks for your comments Liz and Richard. They were useful in helping us plan what to do.

    We are going to run two parallel processes. One is tightly controlled with a small group of well briefed people working to a specific methodology.

    The other process makes the data public and lets anyone interested devise their own methodology to do some analysis. You can find out more at

    It would be really great if you can help.


Leave a Reply

Your email address will not be published. Required fields are marked *