Last week, there was a much discussed story about the online study that reported that Americans choose to spend time on the Internet rather than having sex. The WSJ Numbers Guy makes an astute observation, that "people who answer online surveys aren't likely to be representative of Americans when it comes to online behavior." When he dug a bit deeper, he found that the survey was conducted with a panel of 1,011 online respondents, who responded to the survey during the week that it was open. Furthermore, the people surveyed were drawn from a pool of people that actually signed up to be surveyed. Obviously the results are skewed to users that actively use the Internet. When questioned, Ann Mack, JWT's director of trend spotting replied "The fact that the survey was conducted online may skew the results a bit." This just serves as a reminder to take studies or statistics that are presented to you with a grain of salt. In this case, the company that did the study was looking to create a new advertising category for people whose lives are so tied up with new technology -- so it's not surprising that the results that it found supported this categorization.