« New R User Groups in Seoul, Denver | Main | Because it's Friday: How Machines Work »

August 27, 2010


Feed You can follow this conversation by subscribing to the comment feed for this post.

Web surveys can be extremely biased, because you have no additional information about your respondents (e.g. not likely to be a random sample); meanwhile, I believe it is a good practice to report the sample size when quoting a percentage from a survey -- you could also say "half of respondents considered switching to different software" when there were actually two respondents in total and one of them considered a switch. Personally I hate Newspaper-like titles...

Anyway, I'm happy to see the victory of WPS against (the proud) SAS.



Hahh!!! Another possibility to get biased samples in web: I was not able to post the comment under Firefox 3.6.8 (Ubuntu) until I switched to Google Chrome! The 'Post' button was disabled no matter what I filled in. What if I don't have Chrome installed?

Thanks for letting me know about the posting problem Yihui. This blog is run by Typepad, I'll see if they have a solution about the Firefox/Ubuntu issue.

surprising results...and I agree that it would be a good idea to disclose survey details. The online tool SoGoSurvey has good features and an impressive survey bank.

The comments to this entry are closed.

Search Revolutions Blog

Got comments or suggestions for the blog editor?
Email David Smith.
Follow revodavid on Twitter Follow David on Twitter: @revodavid
Get this blog via email with Blogtrottr