Statistics Essay, Research Paper

1)

From the reading that I?ve been doing on polling it seems that everything has some effect on the results. It seems like most of the polling biases come from the question and the questioner. If the question leads you to think about something else, whether it be positive or negative, it can change the results of the answer. If the person conducting the poll has an opinion of the question they can shape the outcome. Also the gender of the person conducting the poll can change the answer. Also, sometimes if people aren?t completely sure they might just say what?s on the top of their mind without giving it any serious consideration. Even something as minute as the time of day can effect a poll. If you call between 5:30 and 9:00 you leave out everyone that has to work during that time. Many people don?t really want to take the time to answer polls so the results only come from those who are willing to participate.

2)

In order to remove the bias pollsters have adopted many new polling techniques. They try to conduct most of their polls over the telephone so they don?t have physical contact with the person they are polling. This is good because then any preconceptions about race, or lifestyle are not brought into play. It is a disadvantage sometimes because you rely on the person on the other end of the line to provide accurate information about themselves, or they might answer differently depending on whether or not it?s a man or woman asking the questions. Also with telephone polls they use a random number generator to get phone numbers. This is good because it provides the most random group of phone number, but at the same time it can provide numbers of people who may be very indifferent of the subject or do not speak English well enough to answer the question. Pollsters try and arrange the questions so that one does not effect the other. This is good because it helps cut down on spur of the moment responses. The disadvantage is that unless you only have one question it can be nearly impossible to arrange questions in an order that is totally unbiased.