Replay this Initial post:Policy makers, public health officials, scientists, business leaders and others use data analysis to make informed decisions. Unfortunately, data and statistics can also be used to provide misinformation. The numerical and mathematical nature of a statistical analysis might convince an unknowing audience that the information being presented is factual when in reality it is not. Your task is to find an example of a misleading statistic, graph, chart, etc. and describe why the information being presented is misleading or incomplete. News organizations with obvious political leanings and social media are good places to begin your search.
Select Reply to post your response; initial post by Tuesday, continue the conversation with your peers by Friday.
comment on this post:Chris Spann
RE: Week 1: Bad Data Analysis
COLLAPSE
In a US news roundup published on May 23, 2021 on USA Today’s website, a headline claimed that “months of isolation and anxiety brought on by the pandemic have led to a spike in eating disorders.” The analysis referenced on the website looked at electronic medical records from “about 80 U.S. hospitals” and found a 30% increase of cases of eating disorders starting after March of 2020.
Although it is true that the pandemic created a variety of conditions that could have led to eating disorders in people who did not previously have one, the statistic presented on this website is misleading for several reasons. First, the opening statement claims that months of isolation “led” to the spike in eating disorders but analyzing on only 80 hospitals electronic records cannot determine the cause of the increases. It is not reasonable or accurate to pin the cause of the increases to be due to the pandemic alone, as suggested in the write-up, because there could be a number of lurking variables not mentioned or measured that could have also impacted the number of cases where an eating disorder is flagged in the record.
Another misleading part about the presentation of the statistic is how the increase is explained. The write-up states that the “analysis of electronic medical records data from about 80 U.S. hospitals found a 30% increase in cases of eating disorders starting after March 2020.” It is unclear, at least to me, what is meant by the 30% increase after March of 2020. At the beginning of the write-up, the author mentions that eating disorders, under “normal circumstances,” are estimated to cause about 10,000 deaths a year in the United States. The 30% increase just references cases, so the reader has no way to look at the trend to understand what the increase refers to. For example, did the number of cases a month increase 30% and sustain itself at that increased average after March of 2020, or did the total number of active cases increase between two points in time? Without a clear understanding of the meaning, it is difficult to digest the true impact of what is happening with eating disorders in the United States.
Lastly, the statistic presented is misleading because it is only based on electronic hospital records from “about” 80 hospitals in the United States. A quick Google search shows that there are a total of 6090 hospitals in the United States according to the American Hospital Association. We have no way to know how the 80 hospitals and their records were chosen, or if that sample even represents the characteristics of the population of all United States hospitals. 80 divided by 6090 is only 1.314% of the entire population of US hospitals, so it is entirely possible that the sample is too small to accurately represent the population; moreover, the selection for what hospitals had records available may have been biased or not random, which would skew the interpretation of the results.
Overall, claiming that the anxiety and isolation brought by the pandemic caused a spike in eating disorders is misleading because this type of analysis cannot definitively determine the cause of something. In addition, the small sample size and lack of information on how the sample was obtained makes it hard to know if the statistic is a representative estimation of the population. The statistic is incomplete because it is missing an explanation for what the analysis is exactly tracking (increase compared to what), how the sample was formed, and if the sample is even representative of the population in question due to its size.
Source of statistic:
New US cases fall to numbers not seen in almost a year; ‘massive increases’ in eating disorders, expert says: Latest COVID-19 updatesNew cases have tumbled to rates not seen in more than 11 months, sparking optimism that vaccination campaigns are working. Latest COVID-19 news.USA TODAY
Source of number of hospitals:
Fast Facts on U.S. Hospitals, 2021 | AHAThe American Hospital Association conducts an annual survey of hospitals in the United States. The data here, published in 2019, are a sample from the 2017 AHA Annual Survey (FY 2017).American Hospital Association
Also comment on this:I would like to share two articles from two different online platforms that have misleading information and bad charts.
This article published by The New Daily on March 08, 2021, about the Gender Wealth Gap among homeowners. This article has a couple of charts of which there is a pie chart that shows Australia-wide ownership proportions by gender.
In the chart above, there are different segments and values associated with it. However, the chart is misleading the reader with the numbers as the total % sums up to more than 150%. In my opinion, a reader will only look at the percentage of each segment and assume that this is how the properties are owned across Australia. The pie chart should always have equal segments with a total % of 100.
Source:
https://thenewdaily.com.au/finance/property/2021/03/08/gender-wealth-gap-home-owner/
In June 2020, an Argentinian media called C5N published an article showing how many COVID-19 tests are done per million population. Just by looking at the graph, it would seem like Argentina is testing slightly a greater number of people than Brazil (Brasil). On the other hand, it would look like the United States of America (EEUU) is testing just slightly a greater number of people than Argentina and so on. This graph is so misleading for someone who cannot interpret the data properly. Brazil has tested 258 people per million and Argentina has tested 330 people per million. The difference, in this case, is only marginal.
However, in this graph, when Argentina is compared with the United States of America (EEUU), the numbers throw off a huge difference as the USA tested about 7000 people per million which is 20 times more than what Argentina is doing. The difference is significant, and this graph does not show that realistically.
Source:
https://www.facebook.com/C5N.Noticias
Nicholas Kahn INSTRUCTOR MANAGER
Week 2: A Mock Empirical Research Paper
COLLAPSE
Comment on this?
Roughly speaking, an empirical research paper is a research paper that includes data analysis. You won’t be writing an empirical research paper in this course. However, if you were going to be writing an empirical paper, what topic would you be interested in researching? What variables would be most relevant for your analysis? How would these variables be measured (i.e. continuous, discrete, qualitative)? What type of graphs or descriptive statistics would be most useful for describing your dataset?
Select Reply to post your response; initial post by Tuesday, continue the conversation with your peers by Friday.
Comment on this post
COLLAPSE
Hello everyone,
I would have chosen to do empirical research on topic “Will the customer pay the credit card bills upon approval of credit card application and subsequent usage”. As a bank or credit provider institution, I want to know what are the chances that the customer will pay the credit card bill on timely manner.
The variables I could think of customer’s age, employment status, average monthly income, average monthly expenses, accommodation type if living in own house/ apartment or renting or living in RV or similar fashion, number of dependents, number of vehicles owned/leased in total household, state in which the applicant is currently living. I would study how many of these variables are significantly impacting the outcome.
The continuous variables would be – age, average monthly income, average monthly expenses,
The discrete variables would be – number of dependents, number of vehicles owned/leased
The categorical variables would be – employment status, accommodation type (owned or leased), state
The data can be studied by using descriptive statistics like: Mean, Median, Mode, Minimum, Maximum, Variance, Standard Deviation, Q1, Q3 values, IQR, standardize value, and generating graphs like line graph to see the trend, scatter plots to find the relatinoship between two variables, box plots – to study the spread, Histogram – to see the distribution.
Thanks and Regards,
Yatin Pawar
COMMENT THIS ONE AS WELL
Alicia Saputra
RE: Week 2: A Mock Empirical Research Paper
COLLAPSE
The topic that I’ll be interested in researching will be the relationship of social media and mental health in the United States. Nowadays, social media is a big part of our lives and it’s being used on a daily basis. We can provide any updates, share any pictures, even make money on social media. However, mental health is also a big issue. A lot of times, people tend to look at mental health the other way around. The discussion that I wrote last week stated that there is a high correlation between mental health and social media. It means that a lot of people are getting anxiety, depressed, etc with high usage of social media. I’m interested to see what are the correlations between social media and mental health.
The most relevant variable is my analysis would be:
Based on these variables, I will make an analysis and how each variable is related / impacting each other.
In terms of how I would measure the variables:
Two types of numerical variables:Discrete: noneContinuous: age, monthly incomeCategoricalNominal: Gender, type of employment, current mental health conditions, pre – existing health conditions, type of social mediaOrdinal: Life satisfaction
The type of graphs that will be most useful in describing my dataset will be histogramsince it has a relative frequency and frequency. It is also very useful in describing the center and spread of a continuous variable (age and monthly income). Another type of graph will be a bar chart. This will be helpful for describing categorical data (gender, type of employment, etc.) because it has horizontal and vertical axis. Horizontal axis will be used to measure the categories. Vertical axis will be used to measure the values of the outcomes of interest / relative frequencies of the categories.
For descriptive statistics, I will use mean, median, mode, box plots, variance, standard deviation, quartiles, and IQR to measure and describe my dataset.
Source:
https://www.mcleanhospital.org/essential/it-or-not-social-medias-affecting-your-mental-health
Essay Writing Service Features
Our Experience
No matter how complex your assignment is, we can find the right professional for your specific task. Achiever Papers is an essay writing company that hires only the smartest minds to help you with your projects. Our expertise allows us to provide students with high-quality academic writing, editing & proofreading services.
Free Features
Free revision policy
$10Free bibliography & reference
$8Free title page
$8Free formatting
$8How Our Dissertation Writing Service Works
First, you will need to complete an order form. It's not difficult but, if anything is unclear, you may always chat with us so that we can guide you through it. On the order form, you will need to include some basic information concerning your order: subject, topic, number of pages, etc. We also encourage our clients to upload any relevant information or sources that will help.
Complete the order form
Once we have all the information and instructions that we need, we select the most suitable writer for your assignment. While everything seems to be clear, the writer, who has complete knowledge of the subject, may need clarification from you. It is at that point that you would receive a call or email from us.
Writer’s assignment
As soon as the writer has finished, it will be delivered both to the website and to your email address so that you will not miss it. If your deadline is close at hand, we will place a call to you to make sure that you receive the paper on time.
Completing the order and download