Here are two sampling techniques that often lead to biased samples.

• Convenience sample: A convenience sample is only made up of easily available members of a population. For instance, standing outside of a grocery store and asking everyone going into the store, "What is your favorite grocery store?" Obviously, people who don't like this grocery store will be underrepresented in this sample, and those that do like this store will be overrepresented. So, the sample is not truly representative of the population.
• Self-selected sample: A self-selected sample is only made up of people who choose themselves by responding to an appeal. For instance, a television station might ask its viewers to go to a website and select the candidate for whom they are going to vote for in an upcoming election. Usually only people who have a strong opinion will take the time to respond, which will result in a sample that is not truly representative of the population.

If you're interested in learning more about the 1936 Literary Digest poll, check out this report on Scribd. The report gives a background on the poll and an explanation of why the poll was so flawed.

Sample answer:

In 1993, presidential candidate Ross Perot asked Americans to complete a survey in TV Guide and mail it in. The sample was biased for two reasons. First, response to the survey was voluntary so it could not guarantee a representative sample. Second, the questions on the survey were biased.

To read more about this infamous 1993 survey, you can visit http://www.hillsdalesites.org/personal/dmurphy/Kclasses/math105/PerotPollStory.pdf.