Discussion 1

docx

School

University of Maryland Global Campus (UMGC) *

*We aren’t endorsed by this school

Course

650

Subject

Statistics

Date

Feb 20, 2024

Type

docx

Pages

Uploaded by MajorMoleMaster406

1. Working in a clinical research setting, I see and use descriptive statistics regularly when organizing and summarizing our data results or when reviewing and studying other research studies. In my line of work, it is really important to both present the data accurately and critically analyze the descriptive statistics from our own data in hopes of answering the clinical research question and supporting our hypothesis. Often, I am responsible for collecting the data, organizing it, and presenting it via graphs and/or numerical charts to be submitted for publication or presentation. It is also equally as important to critically analyze the descriptive statistics displayed in other research studies as it can be instrumental in determining if additional research is required surrounding the clinical question presented. This is often something I complete when finding supporting scholarly articles or references. In my personal life, I’m aware that there are many times I see and refer to descriptive statistics but most commonly and on the regular basis, I check the weather trends and forecast’s presented. 2. A common misuse of statistics occurs when only a portion of the data is presented. This was the case with a Colgate advertising slogan from 2007. Colgate claimed that “More than 80% of Dentists recommend Colgate” (Lebied, 2018). This falsely led to readers believing that 80% of dentists would recommend Colgate over any other brand, while the remaining 20% would instead choose a competitor’s brand of toothpaste over Colgate. But, what failed to be presented was the fact that when surveyed, the dentists were allowed to select more than one brand of toothpaste and ultimately, another competitor brand toothpaste was recommended just as highly as Colgate (Lebied, 2018). This type of advertising is not uncommon and is still seen in commercials or billboards today. 3. Chart junk is adding visual graphics or elements to a chart or graph that is not necessary for the data to be presented. In other words, chart junk is anything presented that is not data or a scale/label (Kosara, 2010). Often times, though, chart junk is used to spark the reader’s interest. As shown below, the graphic of a football player, is more appealing to the eye, than a bar graph and therefore, use of chart junk to pull the reader in, is useful (Data Lass, 2017).

Chart junk, when used in newspapers, web articles and news reports, can help engage the reader/viewer. But, although useful as a way to attract an audience, chart junk should likely be avoided when the audience is already engaged and is looking for supporting data or information. For example, when reviewing a scholarly clinical research article or performing data analysis, chart junk would likely distract from the data instead of complimenting it. Unfortunately, some of the worst data I have seen displayed is in the media when viewing potentially biased news reports. In 2012, Fox News displayed the following image to show the change in the top tax rate if the Bush tax cuts were to expire in 2013 (Flowing Data, 2012). At first glance, it looks as though the top tax rate would increase by more than three times the current amount. But, when looking more closely, the scale begins at 34% instead of 0% and therefore skews the visualization of the graph. As shown below, when scaled appropriately, it’s much easier to see the difference is 4.6% instead of 3-4 times more (Flowing Data, 2012).

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version

Access to all documents
Unlimited textbook solutions
24/7 expert homework help

This example is a way to display data badly as described by Wainer’s Rule # 5, “Graph Data Out of Context.” An arbitrary scale can lead to create skewed perceptions of the data and this type of manipulation can easily be used to press an inaccurate idea about a certain data point. 4. After looking at the data broken down into subset categories, it reveals that Hospital B is actually superior in cardiac surgery. As pointed out, Hospital B has a higher survival rate in all three categories of patients (Fair, Serious, and Critical) but when averaged all together, Hospital A has the higher overall survival rate. This is because Hospital B has a much higher percentage of critically ill patients. Critically ill patients are less likely to survive than fairly ill, or seriously ill patients. Therefore, since Hospital B sees more critically ill patients, the total average for Hospital B’s success rate is lower than Hospital A. Examining data in subsets allows you to avoid the Simpson’s Paradox phenomenon in which a trend appears in several different subsets of data but disappears or reverses when the subsets are combined (Carlson, 2010), as demonstrated with the Hospital A/B example.

Carlson, B. (2010). Simpson’s paradox . Britannica. https://www.britannica.com/topic/Simpsons- paradox Data Lass (2017, October 23). Data Visualization Design, Part 4: Removing Chart Junk. https://medium.com/@LauraHKahn/data-visualization-design-part-4-removing-chart-junk- 28b3bdd0faa1 Flowing Data. (2012, August 6). Fox News continues charting excellence. https://flowingdata.com/2012/08/06/fox-news-continues-charting-excellence/ Kosara, R. (2010, April 22). Chart Junk Considered Useful After All. Eagereyes. https://eagereyes.org/criticism/chart-junk-considered-useful-after-all#:~:text=There%20is %20almost%20universal%20agreement,detrimental%20to%20understanding%20the%20data . Lebied, M. (2018, August 8). Misleading Statistics Examples – Discover The Potential For Misuse of Statistics & Data In The Digital Age. Datapine. https://www.datapine.com/blog/misleading- statistics-and-data/ McCready, R. (2020, April 17). 5 Ways Writers Use Misleading Graphs To Manipulate You . Venngage. https://venngage.com/blog/misleading-graphs/ Wainer, H. (1984, May). How to Display Data Badly. The American Statistician, 38 (2), 137-147. https://doi.org/10.2307/2683253

I found my research on Simpson’s Paradox to be very interesting and revealing about how data can seem skewed based on the level of aggregation. Dr. Trefor Bazett (2020) explains how important it is to find casual relationships and know when it’s appropriate to aggregate data or not. Bazett, T. (2020, July 08). How Simpson’s Paradox explains weird COVID-19 statistics [Video]. YouTube. https://www.youtube.com/watch?v=t-Ci3FosqZs The use of percentages. Imagine that Politician A says, “Unemployment decreased by 25% during our term.” Politician B says, “No, it only decreased by 2%.” These are interpretations of the same data. Imagine that out of a city of 100 people, 8 are unemployed. Later, only 6 out of the 100 are unemployed. Politician A interpreted the data as the unemployment decreasing from 8 people to 6 people, which is a 25% decrease as he correctly stated. Politician B interpreted the data as the unemployment decreasing from 8 people out of 100 to 6 people out of 100, which is a 2% decrease as he correctly stated. It’s scary to think that this could be happening right under our noses.

Your preview ends here

Eager to read complete document? Join bartleby learn and gain access to the full version

Access to all documents
Unlimited textbook solutions
24/7 expert homework help

Discussion 1

Related Documents