Want to know more?
Why not contact us so that we can advise how
demographix could give you a competitive edge
The Demographix Research Blog
Random thoughts from the team at demographix, keeping you up to date with what's new in the world of online research. Use the year and month links below to navigate the blog entries.
all 2006 2007 2008 2009 2010January February March April July August
August 2010 blog entries:
Using weighted averages in survey analysis
5 Aug 2010 — Demographix has a highly flexible weighted average facility built into its analysis software. It means you can assign values to answer sets, and the system will calculate a weighted average.
For instance, if you have a set of satisfaction statements, you can assign values for each one (say "Very Satisfied = 10" through to "Very Dissatisfied = -10" with increments of 5 points). An overall weighted response of 4.33, would indicate that the respondents lean towards being Satisfied (which was given a value of 5), while a -5.6 would indicate the data set overall leaned towards being slightly more than Dissatisfied.
You are not in any way restricted in what values you assign. Demographix allows you to apply whatever values you wish to a set of statements. As examples, it could be any of the following:
Very satisfied=100 / Satisfied=75 / Neither satisfied nor dissatisfied=50 / Dissatisfied=25 / Very dissatisfied=0
Very satisfied=5 / Satisfied=4 / Neither satisfied nor dissatisfied=3 / Dissatisfied=2 / Very dissatisfied=1
Very satisfied=100 / Satisfied=50 / Neither satisfied nor dissatisfied=0 / Dissatisfied=-50 / Very dissatisfied=-100
Very satisfied=1 / Satisfied=0.75 / Neither satisfied nor dissatisfied=0.5 / Dissatisfied=0.25 / Very dissatisfied=0
...or any combination thereof.
Your interpretation of results, of course, will depend on these value sets. In the example above, the two products have been weighted precisely by the values selected (this is automatically done in selected slider questions within Demographix). The first product's average value indicates it has a far higher "excellence weighting" than Product 2.
Not only do we allow this flexibility, but value setting is also fully dynamic - you can change values after data has started coming in, and at the point of analysis. By editing the question, you can select the "Edit Average Values" option shown above, and apply new values sets. These will be used, the next time you clcik on the Analysis of Responses button.
Be sure to check back regularly. You can also sign up to hear about future demographix news.