Your second sentence does not follow logically from your first though.
A randomly selected male might be half as likely as a randomly selected woman to be a victim of domestic violence, but what a man in the far smaller set of people who have googled that particular phrase?
I would venture to say the ratio might be a lot closer
You may be right (and that’s a thing I hadn’t considered, thank you!) but I worry that getting it wrong means you’re adding to noise and masking the message underneath a veneer of familiarity (how many compulsive gamblers see the “Gambling problem?” message at the bottom of casino billboards? How many smokers read the surgeon general’s warning?).
To get that calibrated correctly, you’d need some research; but the data on why a person searches a particular search term is definitely going to be skewed in this instance by a flood of people searching these two terms to compare the results.
Your second sentence does not follow logically from your first though. A randomly selected male might be half as likely as a randomly selected woman to be a victim of domestic violence, but what a man in the far smaller set of people who have googled that particular phrase? I would venture to say the ratio might be a lot closer
You may be right (and that’s a thing I hadn’t considered, thank you!) but I worry that getting it wrong means you’re adding to noise and masking the message underneath a veneer of familiarity (how many compulsive gamblers see the “Gambling problem?” message at the bottom of casino billboards? How many smokers read the surgeon general’s warning?).
To get that calibrated correctly, you’d need some research; but the data on why a person searches a particular search term is definitely going to be skewed in this instance by a flood of people searching these two terms to compare the results.