All students touched upon reasons why we might have seen such broad variability in response to the question "are you familiar with the digital divide?" Could this be an example of "measurement errors"? There are basically two causes of measurement errors - "random" and "systematic". In this case, it is likely due to the fact that the response groups are so different. Read the following link and offer a response your thoughts as you are able.
http://www.socialresearchmethods.net/kb/measerr.php