Question

Topic: Research/Metrics

Citing Different Survey Sources

Posted by Anonymous on 5/20/2014 at 12:16 AM ET 250 Points
My company wants to put out an industry report based on the results of our recent survey but want to use results from surveys from our business partners to support the findings. I'm cautioning against it given that a) we don't have access to the partners' raw data, and b) methodologies/question/question placement may differ, c) the surveys may have been administered at different times. I can go on, but I digress. Should they opt to move forward with the idea anyway ...IS there a good way to go about this?
To continue reading this question and the solution, sign up ... it's free!
SUBSCRIBE
Or, SIGN IN

RESPONSES

  • Posted on 5/20/2014 at 8:31 AM Moderator
    Sure. Just disclose that the results cited are from unverified sources, though you have no reason to question them.
  • Posted by Gary Bloomer on 5/20/2014 at 6:43 PM Accepted
    Although I agree with Michael I'd urge a tad more caution on using unverified sources —IF there's a risk that your readers could misinterpret your intent (which there may be scope to do). If your gut is to avoid the data that you can't verify, go with it (in as much as you're able to). State your concerns to management (or whoever), and if they insist, drop the issue. Why? Because if this is a decision that needs to be taken at several levels above your salary grade you may want to ask yourself if the view is worth the climb.
  • Posted on 5/20/2014 at 8:06 PM Moderator
    There are a number of ways to disclose the uncertain nature of the findings you cite. For example, "... The company's report of findings states ..." or "... the company's own analysis showed ..." etc.

    And any hard number should probably be footnoted to say that the research was conducted and reported by XYZ company, so it's clear that it's not YOUR number.

    Rarely will all the methodologies and base sizes be exactly the same as yours. Unless you have reason to doubt the accuracy, I would think that a simple disclosure should be sufficient.
  • Posted on 5/21/2014 at 8:58 AM Author
    Thank you both for your response. I guess I've always bee accustomed to reporting only on the findings of our surveys, not those of others. This report will be handled by a different department and written by a freelance writer. I've been asked to provide my input from a research perspective. I guess my concern is that they'll want to compare the 59% of whatevers from survey A to the 46% percent of whatevers from survey B and because they don't know the methodology for survey B they may end up comparing apples and oranges.
  • Posted by tcgren on 5/21/2014 at 10:31 AM Member
    I would disclose the links to the other sources, so that readers can verify the details. And if those surveys change, then readers can get whatever revised results without asking you for it.
  • Posted on 5/21/2014 at 11:09 AM Author
    I just wanted to clarify a couple of points. When I mentioned the comparison, I'm not referring to the reader, but the writer of the report - I meant comparing as in implying correlation between our results and the partners', or comparing their data points to ours within a graph/chart, etc.. Our org is considered the voice of the industry so I'm anticipating questions from readers and media.

    Gary you made a good point about the climb.

    Thanks again.
  • Posted on 5/21/2014 at 12:28 PM Accepted
    Since your company is publishing the report, and since you have a leadership reputation at stake, I would just make sure that there's a sentence or two of that discloses the possibility of mismatched data because you are using results from different surveys, with information gathered in different ways, at different times. That should be sufficient.
BACK TO TOP

Post a Comment

Related Discussions

Community Info

Top 25 Experts
(Research/Metrics)