The data you have is the data you have but not necessarily the ground truth. See, for example, San Francisco shoplifting statistics.
You can't just say "the data COULD be unrepresentative", then just use some random guys anecdotes on the internet as more valid than national statistics.