Log-in Polls

A log-in poll is an unscientific poll that typically is conducted by news and entertainment media on their Web sites to engage their visitors (audiences) by providing them an opportunity to register their opinion about some topic that the media organization believes has current news or entertainment value. Typically two choices are given for someone to express her or his opinion. One choice might be for those who agree with the issue and the other might be for those who disagree. For example, a log-in poll question might be to indicate whether a Web site visitor agrees or disagrees that Congress should impeach the President.

These polls are not accurate measures of public opinion on the topic. The people who choose to register their opinion on the Web site represent no known target population, and as such, the media organization cannot know to whom the findings generalize. Most often, response options such as “undecided” are not given as choices. This further invalidates the data, as anyone who holds an opinion other than what the limited response choices reflect or holds no opinion on the topic (e.g. has no previous knowledge of the topic) cannot indicate that on the poll. Furthermore, oftentimes the Web site allows people to register their opinion more than once, by exiting the Web site and then returning at another time, which is a further indication of the unreliable data these pseudo-polls generate. In fact, the data these polls generate do not even necessarily represent the opinions of visitors to the Web site since the method of sampling is self-selection, which is unreliable. Although log-in polls may provide some entertainment value for the media organization and its visitors, especially those who register their opinion, they should not be considered accurate news.

A real danger these pseudo-polls can create is when their results are disseminated uncritically as though they had some degree of validity. For example, a talk-radio program host might cite the “latest results” from the station's Web site poll to stimulate conversation about the issue from call-in listeners. As the on-the-air conversation about the poll topic proceeds, new listeners who were not tuned-in when the log-in poll results were first mentioned may have no sense how the topic of discussion started. Additionally, some listeners may assume the topic is a controversial one among large portions of the public, and the topic may get further discussed and disseminated via other uncritical news outlets and by word of mouth. Thus, what started out as a pseudo-poll finding with no basis of scientific support can take on a life of its own far beyond what it should have received.

Log-in polls also are used by organizations other than the media to gather information about visitors to their Web site. But again, the participants are self-selected, and the data are unlikely to provide the organization with reliable information about their population of visitors.