Making Data Visual: Chapter 2 From Questions to Tasks
All visualization begins with asking questions. We are curious about an issue or thought and look to explore more about it. The question may be far off in regards to its connection with data, but looking into visualization of the data can help correlate the answer to the question.
Operationalization – the process of reducing a complex set of factors into a single metric (Fisher & Meyer, 2017).
It is also important to know who needs to know the answers to these questions. With Fisher and Meyer’s example of “Who are the best movie directors?”, there is a vast audience that may want to know about the data. It could be relatable to a film student, a hiring manager or a journalist.
Proxies – partial and imperfect representations of the abstract that a researcher is interested in (Fisher & Meyer, 2017).
Fake News: What It Is and How to Spot It
Fake news has many definitions and forms. Fake news stories can be falsely written to emphasize a certain viewpoint or they can be partially true with exaggerations. Lack of fact-checking can lead to fake news stories, as well.
One thing that is so dangerous about fake news is how fast it spreads. Spreading like wildfire, fake news can grace a timeline and then BAM, ten thousand shares later and that news is everywhere.
How Finland is Fighting Fake News – in the classroom
Finland has become one of the most resilient nations fighting against fake news, and they are using education to do it. Regulating the information that is spread to a nation is part of this process, but only a small amount can be done due to free speech. Education seems to be the best option for a large-scale intervention of fake news.
Research skills and critical thinking are big parts of education in relation to combating fake news. According to FactBar, Finland’s fact-checking organization, there are three areas to recognize when analyzing something as fake news or not.
Misinformation (defective information or mistakes)
Disinformation (hoaxes)
Malinformation (stories intended to damage)
“Widespread critical-thinking skills and a coherent government response are key to resisting fake-news campaigns,” Marin Lessenski, the Programme Director for European Policies at OSI-Sofia, said (Charlton, 2019).
The Ultimate Guide to Data Cleaning
Incorrect data can cause a lot of issues. For example, in the business world, having incorrect information in a database of customer information can result in wasted time and money for the company or a lost customer.
Data cleaning involves based on the type of data that you need and can let go of.
“Incorrect data is either removed, corrected or imputed,” (Elgabry, 2019).
Irrelevant data is data that you don’t need. For example, if you were looking at the statistics of a soccer team, you would not need to know how many siblings the players have. In addition, duplicated data should be removed as it will alter the dataset’s values.
Assignment Mod 3
Geographical Data – How One High-Risk Community in Rural South Carolina is Bracing for COVID-19
The first data set I collected was from FiveThirtyEight. The data the media/statistical company collected is about high-risk communities in the U.S. that are fighting against COVID-19. The article about the data set goes into detail about a rural town in South Carolina and how the town is bracing for COVID-19 by looking at the percent of individuals in metropolitan areas that are at a high risk on becoming ill from COVID-19. The data collected also shows the number of high-risk individuals per hospital in those areas. The data was collected from the CDC’s Behavioral Risk Factor Surveillance System.
Questions:
- Are there other factors involved in the data that shows higher numbers of at-risk people in larger metropolitan areas?
- For the health conditions that qualify people to become “high or at risk”, is there one that has shown correlations to rising COVID-19 numbers. (Example – does something like chronic bronchitis have a larger impact on someone suffering from COVID-19. Or what do these same numbers look like for someone with chronic bronchitis?)
- Is there a trend related to the number of high risk individuals in hospitals in certain areas as compared to the total number of COVID-19 cases in that area.
For this data, I believe a horizontal bar chart would work best. There are a lot of metropolitan areas involved, but with very few categories. The horizontal bar chart would be able to fit the metropolitan areas and give viewers the ability to compare the data set.
Data from a Reliable Public Data – How Urban or Rural is Your State?
The second data set I found was also from FiveThirtyEight. This data set is about urban versus rural places in the country and how those cities/places are going to affect the upcoming election. The teams at FiveThirtyEight calculated the average number of people living in a five-mile radius of every census track and then created an “urbanization index”.
FiveThirtyEight compared partisan lean with urbanization indexes of all fifty states. Partisan lean is the average difference between how a state votes and how the country votes overall, with the most recent presidential election weighing 50%. This data shows how important the urban-rural divide has become – the more urban states like New York, New Jersey and California fall on the blue side while the rural states like Wyoming, Montana and South Dakota are more red.
Questions
- How does the urbanization index break down when it comes to race/ethnicity?
- What is the average education level in in these areas?
- How many people are registered to vote in these areas?
For this data, I believe a bar chart would be best to show the divide of urban and rural areas in the country when it comes to elections. On the other hand, a map of the U.S. could also be used. The colors red and blue would be used to show partisan lean, and a lighter shade of the red and blue could be used to show a rural area versus an urban area.
MAGGIE PRUITT 