Rules for meaningful workshop evaluations and other simple home-grown surveys

Updates: First, if you look over this list, please leave me a comment (whether you love it or loathe it or really didn’t need it…)
Second, feel free to repost giving this url or the general url www.itchybiscuit.net as a source. You may email me for specific citation information if you wish.
Third, I seem to have posted a draft of this list, and left it unedited for quite some time. Mea culpa. Many extra spaces and poor word choices are now deleted. Haha, I did not follow my own advice, haha…
Now without further ado:

Rules for meaningful workshop evaluations and other simple home-grown surveys

1. Make your survey clean, well formatted, and uncluttered. If you don’t know how to do this, recruit someone who does.

2. Spell check, proofread, and double up on both. Take serious time to do this. Many people will look over this document carefully, while you are asking them for feedback, and they will react to every error.

3. Use a single, sans-serif font. If you don’t know what that is, you definitely need help with #1.

4. All of side 1 should be agree/disagree prompts. All of side 2 should be open ended prompts and questions.

5. Use a consistent Likert scale. The options should be:
“Agree Strongly…Agree…Neutral/Not sure…Disagree…Disagree Strongly…N/A”

6. Use prompts that skew negative and positive.

7. Use more than one prompt about the same issue. If you have too many single prompt issues, you need to focus down on what you really care about.

8. Use at least a few loaded words and provocative prompts, like “The instructions on text and format seem arrogant.”

9. One sheet, double sided, is all you get, unless your workshop went past 6 hours.

10. Use all of both sides.

11. Ask positive and negative open ended questions. Demonstrate that you value criticism by using set-up prompts like “Please list a few guidelines that are confusing or poorly explained.”

12. Open ended questions fail if they can be answered with a simple “yes” or “no.”

13. When writing any prompt or question, think scientifically/experimentally. First, what do you want to know about the participants’ experiences? Then posit a few possible answers… Finally, craft a question or prompt that might serve to test which of those answers is accurate. This may seem confusing, or poorly explained. If so, you need to recruit someone with a strong social science background.

14. With questions, use all 5 W’s of journalism. 4? 6? Whatever.

15. Consider the data entry when you format your Likert. If the forms will be coded by hand/eye, make sure the bubbles/circles are distinctly visible.

16. Include an email address on the form (preferably tear-off) for further responses.

17. Leave a sufficiently large manila envelope out for collection.

18. Do not collect the evals personally, and don’t look through them until you have left the location!

19. If you code up or otherwise sum the data promptly (and you should), consider emailing a short summary to the participants, and inviting further feedback.

20. Think through common self-reporting behaviors and how they will impact your results.

21. Think ahead of time about what kind of results will influence what kinds of decisions.

22. Pay attention to length of responses on the back side, sloppiness of circles/bubbles on the front side, and your return rate; unless you’re a professional social scientist working with a professional statistician, these “data points” will be just as unscientific, and potentially just as useful for reflection, as the actual response content.

23. Make sure you have writing utensils for everyone.