Project ID: 18

Project Topic and Exploration Questions

For our project, we are examining COVID deaths in the US Midwest region (Indiana, Ohio, Michigan, Wisconsin, Illinois, Minnesota) during 2020. With this data, we are looking at how different age and racial groups were affected disparately by the disease. Similarly, we will incorporate information regarding lockdowns and reopenings to understand how government reaction to COVID affected COVID deaths.

Question 1: What age group had the highest death rate?

Question 2: What age group had the lowest death rate?

Question 3: What states were hit hardest/least hard by COVID?

Question 4: What races in the US were hit hardest/least hard by COVID?

Question 5: How did political control of the state affect COVID-19 Deaths?

Question 6: How did a state's lockdown policy affect different age groups and their COVID mortality rate?

Question 7: Do states with a Democratic governor or states with a Republican governor have better results regarding COVID? If so, are there reasons that certain states and political parties fared better?

Question 8: Did young and old people fair the same regarding COVID no matter what state they lived in, or did location matter for a specific age group's death count?

Data Sources

https://www.statista.com/statistics/910817/illinois-population-share-age-group/

https://www.statista.com/statistics/1022301/indiana-population-share-age-group/

https://www.statista.com/statistics/1022332/michigan-population-share-age-group/

https://www.statista.com/statistics/1022674/minnesota-population-share-age-group/

https://www.statista.com/statistics/912099/ohio-population-share-age-group/

https://www.statista.com/statistics/1022770/wisconsin-population-share-age-group/

https://www.nbc4i.com/community/health/coronavirus/timeline-2-years-later-covid-19-waves-tell-story-of-ohios-pandemic/

https://ballotpedia.org/Documenting_Ohio%27s_path_to_recovery_from_the_coronavirus_(COVID-19)_pandemic,_2020-2021

https://ballotpedia.org/Documenting_Minnesota%27s_path_to_recovery_from_the_coronavirus_(COVID-19)_pandemic,_2020-2021

https://ballotpedia.org/Documenting_Illinois%27_path_to_recovery_from_the_coronavirus_(COVID-19)_pandemic,_2020-2021

https://ballotpedia.org/Documenting_Wisconsin%27s_path_to_recovery_from_the_coronavirus_(COVID-19)_pandemic,_2020-2021)

https://ballotpedia.org/Documenting_Indiana%27s_path_to_recovery_from_the_coronavirus_(COVID-19)_pandemic,_2020-2021

https://www.statista.com/statistics/910817/illinois-population-share-age-group/

https://ballotpedia.org/Gubernatorial_elections,_2020

https://www.census.gov/quickfacts/fact/table/WI,OH,MN,MI,IL,IN/PST045222

https://time.com/6235636/life-expectancy-black-americans-progress/

Data Journey and Caveats

At the beginning of this project, we intended to include data regarding the entire United States. However, when looking at how large the dataset was and how unmanageable this data was (Python could not process the initial, entire CSV), we decided to focus particularly on the Midwest. We chose the Midwest not only because the states have a diversity in political orientation, but also because it is applicable to us at Notre Dame in Indiana. After speaking with Professor Kumar about how to refine our data and focus our project, we decided to orient our analysis to the ramifications of political policy on COVID deaths.

Because of the limited scope of our dataset, there are a few caveats that must be considered. Firstly, because we are only looking at the Midwest, this data may be correlated to the general trend of the US, but it cannot be taken as representative of the entire country. Similarly, there are a multitide of factors that we would have liked to focus more on (like income, access to healthcare, and geographical location within the state), but because of the size of the project, we could not incorporate this data. Thus, when considering the insights of the project, remember that there may be alternative factors that we have not fully taken into account. On that note, we have done our best to only limit our insights to what we can extrapolate from our given data.

Also, our state population data uses information from the 2021 census. However, the age demographic data for states uses a source using data from 2021-2022 (there was no similar data available for 2020, the year that we are focusing on). This may create marginal discrepancies between the age demographics in our visualizations and what they were in reality. However, these discrepancies are slim and will not drastically affect our visualizations or insights.

About the Project Members

Will Mansour

I am a sophomore history major from Dallas, TX. I focus my historical research and courseload on issues regarding slavery, race, and the economy in the 19th century US. In my free time, I like to watch and play sports, exercise, and watch movies with my friends. This project interests me because I am passionate about public policy and how to lessen the negative effects of disease.

Zach Cali

I am a current sophomore and resident of Keenan Hall, studying architecture. I am from Hasbrouck Heights, NJ, a suburb of New York City. In my leisure time, I like to play my guitar, run, watch movies and tv shows with my friends, as well as draw. As I hail from NJ, my state was hit very hard at the beginning of the pandemic because of how densely populated it is. As I know many friends and family members impacted by the pandemic, I am very interested to see how the virus affected other parts of the country.

Visualizations

Visualization 1: COVID-19 Deaths by Age Group and Ethnicity by Week (2020)

Insights

Generally speaking, and as expected, the older age brackets were much more likely to fluctuate as a result of their age complications impacting their overall health. Through the line graph, we can tell that there was an uptick in Covid Deaths around week 13 or week 14 of 2020 which corresponds to the middle of March, which is when everything began to shut down. At the end of the year, many of the ethnic groups of the age group in 50-64 actually had higher Covid Deaths than their 85 and older counterparts which defies the age trend you would expect to see and had seen up until that point. Perhaps the variant of the end of 2020 was more deadly to middle-aged folk.The gap between the deaths of white people vs. other races increased as these groups got older. For example, at week 16, while approximately twice as many white people died from COVID as black people in the 75-84 age bracket, almost triple the amount of white people died as black people in the 85+ age group.The gap between white people and other races became larger as the age group increased. This may be explained by the longer life expectancy for white Americans as compared to black Americans. While the life expectancy of black Americans is approximately 71, the life expectancy for white Americans is over 77. Thus, this increasing gap between the races as age increases may simply be due to the fact that white people generally live longer than black people.

Visualization 2: COVID-19 Deaths per 100,000 by Midwestern State with Gubernatorial Affiliation

Insights

*Before beginning, it must be understood that the deaths counted in this visual are deaths from solely COVID-19, in other words, not those with other health complications.* Ohio has the highest Covid-19 death rate per 100,000 people of the midwestern states. Illinois and Wisconsin have the closest death rates per 100,000 people of the midwestern states. Minnesota has the lowest Covid death rate per 100,000 people of the midwestern states. There is a general consensus and stereotype, whether it is based in fact or not, that Democratic politicans are more hard on COVID and enforce COVID policy more strictly than Republican politicians. However, when looking at the governors for these Midwestern states, there seems to be a correlation between gubernatorial party and COVID deaths per capita, with states having a republican governor suffering higher death rates than those with a democratic governor. Of course, this is simply a correlation and there are several other factors, specifically the actual policies enforced to combat COVID (which will be mentioned later), but it is insightful when understanding why certain stereotypes arise regarding certain parties and their treatment of COVID-19.

Visualization 3: Choropleth of COVID-19 Deaths per 100,000 by Midwestern State

Insights

This choropleth shows the differing deaths per capita in different states. Ohio had the highest deaths per capita from COVID in 2020, with approximately 423 people dying from COVID out of 100000. Ohio's deaths per capita was 30 points higher than the next closest state, Indiana. Interestly, both of these two states have Republican governors. Ohio's governor reopened his state especially quickly when compared to the other Midwestern states, as his heavily protested mask mandate was quickly dropped, while restauarants and bars were reopened during the early summer months. The state with the lowest deaths per capita was Minnesota, with 267 deaths per 100000 people. Their governor, a Democrat, was very cautious while reopening during the summer of 2020, as he held off on reopening restaurants, gyms, and other public spaces for longer, and his reopening took longer. This may have contributed to the smaller deaths per capita in Minnesota.

Visualization 4 & 5: US COVID-19 Deaths by Age: Logarithmic and Linear Interpretation

Insights

For these two visualizations, we chose to represent the same data both linearly and logarithmically to highlight specific insights. By using a logarithmic interpretation in the visualization, we can hover over each group to actually understand what each group's death count was. The logarithmic visualization appears to be almost linear in its growth, suggesting that older age groups had exponentially higher death counts than younger ones. The linear interpretation can be used to understand just how disparate these different age groups' experiences were with COVID, as the deaths of the older age groups are massive while the youngest age groups can barely even be seen on the graph. The two largest jumps in deaths come at the 18-29 age group and the 50-64 age groups. This suggests that children were least affected by COVID, and then once adults entered that late adulthood/pre-elderly stage, their death rates skyrocketed. Less than 100 children age 0-4 died of COVID, and just over 100 children 5-17 died, suggesting just how little these younger age groups were affected by the disease.

Visualization 6: Total Covid-19 Deaths by Age and State (with underlying conditions)

Insights

As expected, the higher mortality rates in older age brackets can be attributed to the natural decline in immune function associated with aging. However, a noteworthy deviation from the anticipated trend is observed in Michigan, Ohio, and Indiana, where the 75-84 age bracket reports more deaths than the 85 and above category. This peculiar pattern calls for a closer examination of the interplay between age-related vulnerabilities and other contributing factors in these regions. Additionally, the paradoxical situation in Ohio, which exhibits a higher number of deaths despite having a smaller population compared to Illinois, raises questions about the influence of various factors such as healthcare infrastructure, socio-economic determinants, and regional susceptibility patterns. This calls for a nuanced analysis to unravel the complexities underlying these disparities and enhance our understanding of mortality patterns across states.As expected the older age brackets have more deaths since aging causes weakened immune systems. The trend of this graph is similar to the graphs that do not include underlying conditions such as respiratory illness and diabetes, yet most of the trends are similar. This suggests that states that struggled to contain COVID also had large numbers of people that had these underlying issues, and more people with these underlying conditions were dying of COVID. This helps to understand that policy is not the only factor to consider when looking at COVID deaths, but also underlying conditions are crucial. However, the similar trends may simply suggest not that these states had more people with underlying conditions, but that these people were more frequently dying from COVID, which could be a result of more lax COVID regulations.

Visualization 7: How Age Groups in Different States Comparatively Suffered (DPC = Deaths per Capita/100,000)

Insights

As we have seen previously, the deaths per capita increases as people increase in age. For the 85+ age bracket, deaths per capita hover around 0.4-0.7, whereas it is less than 0.1 in all states for people aged 55-64.Starting with the 55-64 age bracket and for every older age group, the Republican-led states, Ohio and Indiana, had the highest Deaths per Capita. The Ohio governor reopened quickly in the summer, and the Indiana governor also reopened in the early summer months, which may explain for their increased deaths per capita. Michigan and Illinois tended to be the Democratic states with the highest deaths per capita. In Illinois, their governor was not as quick to reopen as in states like Ohio, yet he did reopen more quickly than the very cautious governor in Minnesota. Michigan also reopened very quickly during the summer months, which led to a spike, yet their governor eventually reinstituted mandates that saw COVID deaths drop in the late summer/fall months. In general, this graph suggests that the Republican states had higher deaths per capita than Democratic states (when I say Republican or Democratic, I am referring to their governor, as he or she often made the most important policy orders during COVID regarding mandates and reopenings.)This disparity between party's is related to the Republican leaders' willingness to reopen restaurants, shops, and schools, as well as lift mask mandates, earlier than Democratic governors. Even in states where Democratic governors were faster to reopen (like Michigan), we saw that they had higher deaths per capita than other Democratic states.

Visualization 8: Percentage of COVID-19 Deaths by Age Group in Each Midwestern State

Insights

This pie chart breaks down the percentage of deaths in each Midwestern state based on age group. In every state, we saw that the majority of deaths were among 85+, which goes along with our other visualizations that suggest older age groups were hit the hardest. Minnesota and Wisconsin had the highest percentage of deaths in the 85+ bracket, whereas Ohio, Indiana and Michigan had the lowest. This may be due to the fact that states like Minnesota and Wisconsin were far more cautious with reopening, while Ohio, Indiana, and Michigan reopened things like restaurants and schools more quickly. Younger people are more likely to frequent these types of public spaces, so it makes sense that states that allowed for these to reopen quicker had less of their COVID deaths attributed to people in the oldest age bracket (who go to schools and restaurants less). Within the 0-24 age group, Indiana had the highest percentage of deaths when compared to other states. Indiana reopened their schools very quickly, and much more quickly than other Midwestern states, which suggests that this reckless reopening did indeed lead to more deaths being in young people. Just like the graph before, this graph suggests that policy does, in fact, influence how different groups are affected by COVID. States in which governors were less cautious about COVID saw the composition of their deaths skewed away from elderly people and more made up of younger people, as younger people tend to go out in public spaces more than older people. Similarly, states that opened schools up earlier had a larger percentage of their deaths be children and young adults.

What We Learned

Throughout this project, we learned what kinds of groups were more likely to die from COVID as well as how policy can have a major impact on COVID death counts. These states have a variety of different races and age groups, and understanding how different groups were disparately affected by COVID was crucial to answering our guiding questions (which we do in our conclusion). The clearest observation we have made is that the older a person is, the more likely they are to die from COVID. This observation holds true across all the Midwestern states and across all races. Our learnings about how race affects COVID deaths suggested that race is generally not an incredibly important factor, and that not one race was noticeably affected by COVID in the Midwestern states (except old white people, yet this is simply because white people live longer than other race groups). This exception suggests that there is some lingering inequality that may affect COVID deaths, yet it was not noticeable in our data. When considering partisanship in state politics, it was clear to us that partisan affiliations do affect COVID deaths and deaths per capita. States that have Republican governors are more likely to impose uncautious, reckless COVID regulations (or no regulations at all), which leads to higher COVID death counts. As states become more and more Democratic-oriented in their policy, reopening slower and mandating harsher regulations, their deaths per capita tend to drop. Furthermore, the type of policy can even affect what age groups are dying at higher rates between states. In states where schools opened up very quickly (like Indiana), we see that these states have a higher percentage of 0-24 year old deaths. In states that did not reopen restaurants and other public spaces quickly, we see that a higher percentage of deaths was elderly people, as younger people (who visit these public places at higher rates) were not dying as often and skewing the percentages away from the oldest age groups. Therefore, policy did indeed affect how many people died from COVID in the Midwest and what groups of people died from COVID in the Midwest.

Conclusion

What age group had the highest death rate?

In general, people aged older than 85 had the highest death rates, regardless of state or race.

What age group had the lowest death rate?

Young people aged 0-4 had the lowest national death rate. As the age of specific groups increased, their deaths from COVID increased almost exponentially.

What states were hit hardest/least hard by the pandemic?

In the Midwest, Ohio, Illinois, and Michigan were “hit hardest,” meaning that they had the highest total deaths. However, these states have high populations. Thus, when considering deaths per capita, Ohio and Indiana were hit the hardest, as their deaths per 100000 people were the highest. Minnesota and Wisconsin had the lowest total death counts and the lowest deaths per capita, suggesting that their response to COVID prevented high death rates, unlike some other states in the Midwest.

What races were hit hardest in the US by COVID?

Our data and visualizations suggested that the effects of COVID on different races was not incredibly significant. White people had the highest death counts when compared to other groups, but that is understandable because the most common race in these states is white. Our visualizations have left us with questions about how targeted economic policies and general racial-economic trends could have possibly changed this data, but overall, it seems that COVID posed a similar threat to all races.

How did political control of a state affect COVID deaths?

Generally, states with Republican governors suffered higher deaths per capita than states with Democratic governors. This can explain why, even if their COVID health policy may not directly reflect their political affiliation, certain political stereotypes have arisen regarding the different parties’ treatment of COVID.

How did a state’s lockdown policy affect different age groups and their COVID mortality rates?

States that had stricter, harsher lockdown restrictions generally had lower COVID mortality rates across the board, no matter the age group. Moreover, states that had stricter policies, specifically regarding the reopening of public spaces like restaurants and schools, had lower percentages of their deaths in younger age groups that often attend these public spaces in high volumes. In these states, a higher percentage of deaths were among elderly people, suggesting that states with harsher regulations prevented excess deaths of young people and therefore skewed the percentages towards elderly people.

Summary

Throughout this project, we have both learned a lot not just about how COVID affected our country and the region surrounding Notre Dame, but also about how government policy affected COVID. We both now have a greater appreciation of how policy can be crucial to be preventing unnecessary deaths in public health crises. Similarly, we both have a greater appreciation and understanding of how to make important, insightful visualizations that can explain certain trends in society. We created multiple visualizations that were not useful to our project, yet this helped us to discern what was necessary to get our points across. Overall, this project reshaped our understanding of how data can be used to demonstrate public health trends and can hopefully inform people about how public health policy can have drastic effects on our lives.

Demo Video