Statistics have always fascinated and repelled me. I hated my undergraduate statistics class and muddled my way through my graduate class with a far better teacher and book to help me along the way. If I had been more attracted to statistics, I might have pursued my PhD...
The fascination comes about with data and how the questions asked, methodology used, population studied, sample size selected, time period of the research, researcher bias etc. can all lead to different results.
The latest W I D E discrepancy in Internet statistics has to do with data on broadband and Internet access penetration. The General Accounting Office (GAO) in a recently published report (GAO 06-426, May 2006) reported that about 30 million American households have adopted broadband and that 41% of American homes do not have Internet access. Their data was compiled by analyzing Knowledge Network/SRI's "The Home Technology Monitor: Spring 2005 Ownership and Trend Report". This data is somewhat different then the Nielsen/NetRatings data (Press Release March 14, 2006) that shows that 74% of American homes have Internet Access (which I think would imply that only 26% DO NOT have Internet Access) and that of these 68% use broadband an increase of 13% over 2005. This represented 95.5 million active users from home.
I suspect some of the discrepancy has to do with the timing of the two study samples - Spring 2005 vs. Feb 2006. However, the numbers give me pause, as I look at current Government data showing a less evolved market and private industry data showing a rapidly evolving market with the subsequent consequences to marketing and communications planning.
Where do you think Internet and Broadband penetration are currently at?