# Comparing /Indexing Global Average of a variable to Country within the average

#### sdjw10

##### New Member
Hi All,

This feels like a pretty introductory and basic statistical question, but I have never been able to get a consensus.

Let's say I am comparing a country (USA) to a global average (which includes US data) to get an index to show where the US compares.

So lets say :
50% of the Global market (US/UK/FR/DE) likes chocolate
65% of US likes chocolate

Instead of doing a sig test difference comparing 50% to 65%, we want to do an index (65/50)*100=130, which, if we have an 80/120 thresholds would be a significant difference. In sum (assuming good sample size), we would say the US is more likely to like choc than the Global Market based on this.

My questions are:
- Is this the right way of doing the index? Or should i actually remove the US data from the GLOBAL when building the index? So dividing my 65 by wtv the GLOBAL is without US?
- Is there a case where i would not do an index like this? For example, let's say the GLOBAL average is 4% on a variable and all the countries have small ranges (like 4;4;2;6). In this case, it feels index would be on the 2 and 6 but the range is so small, it doesnt feel right. Are there rules to keep in mind? Like a minimum 15% range within the countries?

Thanks in advance. I could be thinking about this all wrong, just looking for some help.