But that "professional developers" histogram is almost indistinguishable from "all respondents," so really your statement should be "the age distribution of Sta

But that "professional developers" histogram is almost indistinguishable from "all respondents," so really your statement should be "the age distribution of StackOverflow developer survey respondents is different from that of the overall human population." What's the sampling bias of this survey? How many pro developers over 40 were contacted for the survey but did not respond? Is StackOverflow usage higher among people aged 20-39 compared to 40-59, all else being equal?