What Are Weight Variables?

A weight variable provides a value (weight) for each respondent in a data set. Response data that have relatively large weights have more influence in the analysis than the data that have smaller weights. As default setting the imported Weight variable is checked and used. In addition to compulsory Weight variable, you can designate some other variables and use them as new/additional weight variables.

Weight example

In the example below you can see 10 respondents:

Based on this table you can see the following unweighted results:

Brand awareness: 2/10 * 100 = 20%

Brand awareness among women: 1/2 *100 = 50%

Brand awareness among men: 1/8 * 100 = 12,5%

When we need men and women to be representative, we might want to weight them to 50% men and 50% women. To do this, we add a weight factor:

In this table above we have 2 women, but both count as 2,5. This makes them counting for 5 in total. Every man has a weight factor of 0,625 which in total adds up to 5 too (8 * 0,625). Now we have weighted 5 women and 5 men.

Now the weighted results would become:

Brand awareness: ((1*2,5) + (1*0,625))/(2,5+2,5+0,625+0,625+0,625+0,625+0,625+0,625+0,625+0,625) * 100 = 3,125/10 = 31,25%

Which doesn't change the result within the women's group:

Brand awareness among women is (1*2,5)/(2,5+2,5) *100 = 50%

And doesn't change the result within the men's group:

Brand awareness among men is (1*0,625)/(0,625+0,625+0,625+0,625+0,625+0,625+0,625+0,625) * 100 = 12,5%

Weight variables selection in Forsta Visualizations

Selecting variables that will be used as weight is possible by accessing Weight variables tab, under Report objects section. Once you are on the weight tab the new screen will appear where you define the variables to use as weight .

To select a new Weight variable, check the box next to the variable. Change the default Weight by selecting a round button next to checkbox. The Code and Report text search are also present on this list.