Most polling is likely still conducted on people with a land line telephone, what affect if any, do you think has on polling in the coming years?
Most polling is likely still conducted on people with a land line telephone, what affect if any, do you think has on polling in the coming years? The New Hampshire presidential primary during the winter of 2008 was fascinating. Coming off of his victory in Iowa, every poll showed Obama with a substantial (double digit lead in many) advantage heading into New Hampshire. Yet, Clinton stunned pollsters/media by winning. Provide some plausible explanations as to what possibly happened. One of the more likely scenarios for this election was likely polling overreacting or over correcting after Obama’s win in Iowa. Prior to the Iowa caucus, Clinton led polling in New Hampshire by 20 points, but after Iowa had fallen behind by 13 points. By the end of the primary Clinton had won by approximately 3%. The results in Iowa versus the results in New Hampshire, are often incongruent with one another. Part of this is because Iowa is a caucus, versus New Hampshire being a primary. At the Iowa caucus a candidate must earn 15% of the vote at the venue in order to be considered a viable candidate. If the 15% threshold is not met, then the delegates must find a viable candidate to support. New Hampshire is a primary with secret ballot voting as most everyone is accustomed to. Interestingly enough most of the main candidates, Obama, Clinton, and Edwards all ended up with similar numbers of delegates from both states events. The other likely culprit was the amount of undecided voters that could not be categorized properly, that ultimately made a choice. As the chart below indicates over time the undecideds chose their candidate. Also note the linear regression of these polls. As noted from the textbook “In general, it is dangerous to use the fitted regression function to predict Y values beyond the range of the available data.” (Hanke and Wichern, 2009, p. 229). While we have empirical data points, there is a good chance that a lot of the polling was predictive based on the previous few months of polling. As the demographics were examined some of the groups that supported Obama in Iowa, instead supported Clinton in New Hampshire. More women and voters over 40 supported Clinton in this primary, that had supported Obama in the caucus. Over the last few years we have noticed that polling, especially exit polling has become almost unpredictable. Too many news organizations care more about being first with something rather than being correct. Lastly, there is a level of backlash against the news and polling where it is likely that some of those being polled simply do not want to discuss or be judged for their answers, so they deliberately obfuscate their true answers. One final note that the tin foil hatters and conspiracy theorists always bandy about, the Diebold machines were obviously hacked and rigged to register vote counts in order to make the race more interesting. One of the other candidates in the race Dennis Kucinich asked for a recount to make sure that the ssystem had counted everyone’s vote. He cited polling data that showed Obama the clear favorite before the election. “In a letter to Secretary of State William Gardner, Kucinich claimed there were “serious and credible reports, allegations and rumors” about the integrity of the primary results.” (Associated Press reports, January 2008). Question: Most polling is likely still conducted on people with a land line telephone, what affect if any, do you think has on polling in the coming years? Reference: Polling Chart is from wikimedia commons. Hanke, J. E., & Wichern, D. W. (2009). Business forecasting (9th ed.). Upper Saddle River, NJ: Prentice Hall. Associated Press. (2008) Retrieved from The Boston Herald. Retrieved 12 November 2019. https://web.archive.org/web/20080311134002/http://news.bostonherald.com/news/2008/view.bg?articleid=1066072