The index tells you how much more or less likely members of your audience are to exhibit a particular trait compared to the base. An index above 100 means they’re more likely to exhibit that trait, while an index below 100 means they’re less likely.
How should I read the index?
From the crosstab below, we can say that 16-24-year-olds in the UK are 34% more likely to buy new tech products as soon as they become available. If the index were below 100, say 80, we’d say that 16-24-year-olds were 20% less likely.
When is an index meaningful?
Typically, indexes between 90-110 aren’t worth highlighting. However, they can still be useful when investigating large audiences and common traits. That’s because the extent to which an index is meaningful depends on the size of the audience and datapoint in question.
When looking at the cross tab shown above, an index of 134 may seem, at first glance, meaningful. However, less than 10% of the total universe feel that the statement in question describes them. In fact, there’s only a 3 percentage point difference between 16-24-year-olds and the total universe. Therefore, although 16-24-year-olds are more likely to buy new tech products as soon as they become available, it’s important to keep in mind that only a small percentage of them actually do so.
How is the index calculated?
The index is calculated by dividing the column % of the crossover cell by the column % of the total cell for the relevant datapoint, and then multiplying by 100. Returning to the example above, we can calculate the index like so:
When doing this, it’s important to be aware that the platform uses unrounded figures to make its calculations. As such, manual calculations won’t always align perfectly with what’s shown on the platform.
I’ve calculated the index manually and am getting a different result - why?
Some questions aren’t asked of all respondents and instead are asked in a follow-up survey called an extension module. These modules are completed by a representative subsample of the main sample and are weighted to represent the same total universe as the rest of the study. As far as indexes are concerned, this means that respondents outside of this subsample have to be excluded from the calculation. The platform does this automatically, so it’s not something users of our data need to worry about. However, it does mean that manual calculations won’t correspond with what’s shown in the platform in such situations.
Can I calculate indexes between two audiences?
Yes - by dividing the column % of one audience with the column % of the other, and then multiplying by 100. Using figures from the crosstab below, we can calculate how much more likely 25-34-year-olds are to feel that the statement in question describes them than 16-24-year-olds like so:
This figure should be read like a regular index. We can therefore say that 25-34-year-olds are 5% more likely to buy new tech products as soon as they become available than 16-24-year-olds.