survey, program evaluation

Better survey results: Presenting the results of a retrospective pre-post design

April 05, 2018 | Pieta Blakely


two men sitting at a table with books and pens


A retrospective pre-post design is one where you ask learners to evaluate their own learning after a training event. This is a great alternative to administering two surveys, one before and one after.

It is hard for people to rate their knowledge of a topic before they've been exposed to it. They don't know what they don't know. Asking them to rate their knowledge before the training and then after it can lead to confusing results where they actually say they know less after the training -- after they've realized that the topic is more complex than they had assumed. A retrospective pre-post avoids this problem by helping learners rate their prior knowledge in the context of what they know now.

With a retrospective pre-post, you don't have to worry about matching pre- and post- responses if the survey was anonymous. Here, they're both on the same piece of paper, so you can see whether an individual made progress.

You might ask two questions like this:

Before today's training, how would you rate your knowledge of facilitation?

I did not know anything
I knew very little
I had some knowledge
I was very knowledgeable


After today's training, how would you rate your knowledge of facilitation?

I do not know anything
I know very little
I have some knowledge
I am very knowledgeable


I used a bold font on "before" and "after", because they questions look very similar and this will help someone who is reading quickly recognize the difference between them.

You would probably ask several questions in this format. For example, how much do they know about the history of facilitation, cultural considerations in facilitation, and the practice of effective facilitation, followed by a question about their confidence that they would apply this learning in an upcoming meeting.  

On each of these questions, you will probably use a Likert scale for your answer choices, and that will have some implications for what you can do with the data. One thing you can not do is calculate a change from pre to post scores.

So here are some things you can do with the data.

Option 1: Summarize the important categories of the post data. Example: 90% of the participants said that they were knowledgeable or very knowledgeable after this training (compared to 10% prior to training)

Option 2: Report on the number of people who said that their level of knowledge had increased. Example: 70% of the participants reported that they were more knowledgeable about facilitation after this training than they had been before.

Option 3: Use a stacked bar chart

pre-post training-1

I've written more on using bar charts to present Likert scale results here.

If you try this, please let us know how it worked for you in the comments below!

Subscribe to my blog



Tags: survey, program evaluation