Statistics give the numbers that add credibility to research findings, proposals, ideas, conclusions, and recommendations. Despite the significance of statistics, accepting them without understanding how they have interpreted or without critically looking into the methods used to acquire the statistics nullifies the credibility of the idea or report supported by the statistics (Dominitz & Manski, 2018). Deception in statistics is common, and as such, it is imperative to understand the ways in which the statistical data might be erroneous. Three forms of bad statistics are presented in the assignment video. The video first explains the importance of statistics in our daily lives and our reliance on the numbers from news outlets (Dressler, 2010). This paper explains the forms of bad statistics and why they should be avoided.
The first form of bad statistics mentioned in the video is poorly collected data (Dressler, 2010). Poorly collected data might consequently produce entirely wrong results. An example of poor data collection was provided in which researchers seeking to find out which magazine was read the most made calls during the working hours. This meant that there was a high likelihood of only stay-at-home mothers giving the responses to the research, thereby limiting the research findings only to that group (Dressler, 2010). This form of data collection provides results because not all the population groups are involved in the research. According to Adams & Lawrence (2018), the underlying cause of poor data collection is the lack of proper research design or subversion of the research methodology. For effective data collection, the sample population should represent the entire population, be proportional to the population and be specific to them. The consequences of poorly collected data include but are not limited to misleading related subsequent research, misappropriation of resources, compromised decision, and the inability to validate and accurately answer research questions. It is therefore important to conduct data collection appropriately so as not to invalidate the research findings or further related research.
Delegate your assignment to our experts and they will do the rest.
The second form of bad statistics highlighted in the video is the use of leading questions (Dressler, 2010). This includes designing the research questions in a way likely to prompt certain specific responses or guiding the respondent in providing his/her responses. For instance, in a research seeking to find out the public’s opinion on whether cell phones cause cancer in which the researcher suggests a case of examples relating cell phones to cancer, most respondents would positively respond to the question regardless of whether they believe if the cell phones cause cancer or not. On the contrary, simply asking the respondents if they believe cell phones can cause cancer would generate truthful result not driven by the researcher’s suggestive inclination.
In an article published by the Pew Research Center, leading questions are described as either close-ended or open-ended. The research sought to find out what issues mattered the most to the electorates in the election of a president in the 2008 presidential elections. The closed-ended category questionnaire consisted of six categories from which the respondents were to choose from. From the results, 58% of the participants in the open-ended group answered that the economy was a major factor of consideration; while only 38% of the participants in the close-ended category answered that the economy was the main issue. The participants in both categories had been evenly sampled (Suh, 2015). This example demonstrates how the structuring of questions affects the responses and results of a research project.
The third bad statistic highlighted in the video is the misuse of the measures of centers (Dressler, 2010). In statistical analysis, the mean is usually used in determining the average of a set of data. However, alteration by outliers can result in discrepancies, and the median can be used in place of the mean. Outliers are data sets that fall outside the pattern or trend of the other data in the set. Although outliers are part of the accurate data set, including them in the statistical mean calculation may lead to a misleading result. If the outlier is of a lower value than the rest of the data, the resulting mean will lower and vice versa. The median, on the other hand, is not affected by outliers (Stone et al. 2012). While the average of a statistical data set is determined by taking the sum of all the data and dividing it by the number of values in the set, the median is determined by finding the middle value of the set of data. Inclusion or exclusion of outliers will insignificantly affect the median value. Using the median as a measure of central tendency for a statistical data set gives a more accurate representation of the data pattern.
The example provided demonstrates how outliers may affect the measures of central tendency. In the example, Michael Jordan’s salary is depicted as an outlier when graphically compared with the income of his fellow Geography graduates. Jordan’s high income is from his profession as a basketball player, unlike his fellow graduates whose income is dependent on their Geography professions. This example proves why the man may not always be used to determine the average of a set of statistical data. In the example provided in the video, the mean annual income of the 1986 Geography graduates from the University of North Carolina was $250,000 whereas the median was $22,000 (Dressler, 2010). Another example in which outliers can prejudice the conclusion is that of a man who drowns in a pool with an average depth of four inches. The pool might be four inches everywhere, but at the center, there is a hole ten feet deep. Most likely, the man might have drowned in the ten-foot-deep hole, but using the average, but the measures of the statistical center will mislead the conclusion.
The Michael Jordan Fallacy, as hinted in the video, demonstrates how the inclusion of outliers can result in mean values that do not depict the accurate representation of the statistical data. Michael Jordan graduated from the University of North Carolina in 1986 with a Geography degree. A statistical analysis of the average salary of the 1986 UNC Geography graduates was $250,000. This raised the question as to why the UNC graduates were earning so much money yet the national Geography graduates were starting at $22,000 a year. The floated average resulted from including Michael Jordan’s salary in the calculation (Dressler, 2010). Michael Jordan earned a lot more money than his fellow graduates as a professional basketballer. If the median had been used to determine the center of UNC graduates’ salary, the value would have been closer to the average national starting salary of Geography graduates.
Statistical data might be accurate, but if they are incorrectly collected or misanalyzed they might lead to misinterpretations and misleading conclusions. Statistics are a part of the daily life activities as they occur in most aspects of life and as such, proper methods of collection and analysis should be used to ensure the validity of the information they support or substantiate. Misleading statistics are a result of poor data collection methods, showing bias and the use of leading questions, and the misuse of center.
References
Adams, K. A., & Lawrence, E. K. (2018). Research methods, statistics, and applications . Sage Publications.
Dominitz, J., & Manski, C. F. (2018). More data or better data? Using statistical decision theory to guide data collection. LSE Business Review .
Dressler, E. (2010, February 27). Don't Be Fooled By Bad Statistics. Retrieved September 06, 2018, from https://www.youtube.com/watch?v=jguYUbcIv8c
Stone, B. K., Scibilia, B., Pammer, C., & Steele, C. (2012). Using the Mean in Data Analysis: It’s Not Always a Slam-Dunk. The Minitab Editor. Retrieved September 06, 2018 from http://blog.minitab.com/blog/michelle-paret/using-the-mean-its-not-always-a-slam-dunk
Suh, M. (2015, January 29). Questionnaire design. Retrieved Septembe 06, 2018, from http://www.pewresearch.org/methodology/u-s-survey-research/questionnaire-design/
Ott, R. L., & Longnecker, M. T. (2015). An introduction to statistical methods and data analysis . Nelson Education.
