When it comes to writing training survey questions, the most important principles is to write questions that seek responses on which you can take action. After all, isn't that the point of survey responses? You want to collect feedback so that you can improve your training course. Yet, too often, we write questions that no one answers or that seek responses on which we have no way of making improvements. Either way, we have survey data that is useless except that we can tell our stakeholders and check the box: Yes, we survey our learners.
With a little effort, we can improve the effectiveness of our surveys so that more get completed and we get the information we need to take action. Here are just a few ideas that can improve your surveys.
There is nothing that prevents people from completing survey more than a long survey. People are busy, so if you ask someone to complete a 20 question survey, especially a survey that requires people to click through several pages, you are increasing the chances that people will not complete your survey. I can't tell you how many times I have started a survey and abandoned it because it took too long. Sound familiar?
Keep your survey's short. There is no precise number of recommended questions, but for sure, let it be 10 or fewer questions. I'd suggest you strive for five questions.
Instead of writing this question: Did you find the course effective? Write the question in the form of a statement: I found the course effective. The first question begs a "Yes" or "No" response and does not allow for the possibility of a "Maybe" or an otherwise very strong or very negative response. Use statements and ask people to rate their level of agreement with each statement. This will give you a better idea of what people think of your training.
I have been in too many meetings in which people argue about whether to use a four, five, or a ten point scale in survey responses. There isn't one answer, but here is my advice. Stick with a simple five point scale. Ten is far too many. Seriously, what is the difference between a seven and an eight anyway? A four point scale forces someone to choose a positive or negative agreement even if they are neutral. This can skew the results. A five point scale is simple and allows for a neutral response.
Please, do not force people (make comments field required) to add comments, unless you want people to type in random text or useless comments just to complete the survey or skip your survey altogether. Personally, I just add a single period to required comments fields, if I don't have anything to say. Sometimes people do not have anything to say. If they do, they will add comments.
The worst possible statement you can make on a training survey is one on which you cannot take action. For example, we want to know whether people think the training was too long or too short. So we make this statement, The length of time was appropriate. If a respondent disagrees with this statement, what are you going to do about it? Does disagreement mean they thought the training was too long? Too short? There is no way to know. So this is a wasted question. Don't ask it or any another question like it.
Although a course survey is only the first level of evaluating the effectiveness of your training course, it is a valuable way to collection learner satisfaction data. If you genuinely value feedback, you can spend the time to write survey questions that collection data on which you can take action. I recommend reading this guide to writing survey questions.
Bill Cushard, author, blogger, and learning experience (LX) designer, is a human performance technologist (HPT) with extensive, in-the-trenches experience building learning organizations in start-up and hyper-growth organizations like E*TRADE, the Knowland Group, and Accenture. You can follow him on Twitter or on Google+.