Have an idea?

Visit Sawtooth Software Feedback to share your ideas on how we can improve our products.

Progress bar - estimated %

I'm trying to understand the mechanics behind the estimated % for the progress bar where your survey contains loops. I was testing the progress bar against a loop question and the default progress bar method is misleading as the progress bar just sits in the one  position and treats the whole loop question like it is a one page question.

So going to the alternative method, I'm a bit confused with what % I should use. I understand it is an estimate and not an exact science.

Referring to the Sawtooth help ...

Question seen method:  This method shows progress based on (#questions seen)/(total number of questions based on Estimated % of questions each will see).  This method potentially works better for questionnaires involving loops and randomized blocks.

I'm also wondering how a loop question with say 5 pages  would be treated in this calculation? Would it be 1 (referring to the question) or 5 (referring to the number of pages displayed)?

Appreciate any assistance. Thanks.
asked Mar 12, 2014 by Paul Moon Platinum (91,280 points)
Hi - Did you figure this out?  I am realizing this question is very old.  I have a survey with a few loops, branching, and skip logic.  I get an error with both specifications of the progress bar and as you mention, the bar just stays stuck at 0% as the person goes through the survey.  What did you end up using?
Wow, this is an old post.

The progress bar is 100% accurate for your straight forward START to END surveys (where loops and randomised blocks / pages do not exist).

You simply have the progress bar setting set to "Question position in the questionnaire" and away you go.

When these elements do exist, you can select the other setting option "Questions seen" and set the %. Sometimes you test and fiddle with that % to get it just right. But this option works well when you know how many times you are going through a loop and how many blocks are being shown.

When you have loops and randomised blocks that are conditional, and you don't really know how many times the respondent goes through them (because the iterations are dependant on a previous question answer), determining the progress bar % setting becomes impossible.

I see 2 options at this point ...

1/ You can set the progress bar to a specified % using [%ProgressBarSet(Value)%] at any given question. You can use this multiple times throughout the survey. You may use this once you know how many times you are going through a loop. If your survey has many variations due to loops and randomised blocks, it can still be difficult.

Below is an example of setting the progress bar based on a question response using Perl Script ...
[%Begin Unverified Perl 

 if (VALUE("Q1")==1)
 elsif (VALUE("Q1")==2)
End Unverified%]

2/ Turn it OFF.
Thanks, Paul.  I might indeed just have to turn it off.  For the Perl script above, would that just be placed in the footer of the relevant question?
Yes to the Perl Script being placed in the footer of a question. Do a test and check it out.

You can also use SSI Script when no conditions apply ...


And turning off the progress bar is okay when you have those surveys where the length of the survey is impossible to determine given there are many combinations.

Your solution to the original question

Please only use this to answer the original question. Otherwise please use comments.
Your name to display (optional):
Privacy: Your email address will only be used for sending these notifications.
Anti-spam verification:

To avoid this verification in future, please log in or register.