More than a thousand hits is a very good sample size. The total population of the UK barely matters in this check for statistical significance. A very general

More than a thousand hits is a very good sample size. The total population of the UK barely matters in this check for statistical significance. A very general rule of thumb is that you want 10-100 samples in each category. In this case it's about a thousand children in one category and millions in the other category. That's more than enough to measure a 30% difference.