Don T Be Fooled By Bad Statistics

There are several types of bad statistics that can be seen when looking at statistical data. According to the video “Don’t be fooled by bad statistics” (2010), there are three basic types of bad data consisting of poorly collected data, leading questions, and misuse of center. Poorly collected data can produce misleading results. For example, when a publishing company conducted a phone survey of popular magazines but did so during business hours when stay at home moms were most likely to participate in the survey, this led to only two types of magazines being selected. Thus, the results are not representative of the population and as such are inconclusive (Dressler, 2010). This led to a waste of time and money on part of the research company. Poorly collected data thus can be considered a sampling bias where the sample is not an accurate representation of the population (Miethe, Gauthier, 2008). A sampling bias occurs when there are flaws in the sampling process as in the example illustrated above.
Questions must be properly worded in order not to elect a bias response. For example, a researcher asked participants in a survey their thoughts about a correlation between cell phones and cancer. The researcher told participants that researchers at a leading university found that cell phone use leads to an increase in cancer, and then asked if the participants agreed. Most people would tend to agree with data by a leading researcher, and thus it leads the question (Dressler, 2010). Instead, it should be asked if the subjects believe something without leading them. Leading questions causes errors in the data collected as the data collected is influenced by the researcher’s leading question. Leading questions can be a form of human fallibility and even a political influence