Problem
Was limited English language proficiency creating significant disparities in COVID-19 vaccination rates across the city of Chicago in 2021?
Method
To explore that question, we are going to use maps to visualize the following data:
measuring COVID-19 vaccination rates and limited English proficiency by Chicago zip code;
then compare that to counts of COVID-19 vaccination sites by Chicago zip code.
Insight happens by contrast. That means our analysis needs three different steps.
(1) Do vaccination site counts by zip code mostly coincide with vaccination rates by zip code? That would show a possible association between more vaccination sites in a given area and better vaccination outcomes.
(2) If so, are areas with limited English proficiency breaking that pattern, presenting worse vaccination outcomes than areas with higher English proficiency and a similar amount of vaccination sites?
(3) Areas with similar amounts of vaccination sites, different English proficiency levels and different vaccinations outcomes could suggest reduced access to COVID-19 vaccines due to linguistic diversity. That could reveal a possible bottleneck to widespread vaccination beyond just proximity to vaccination sites.
If these three different steps are not met, it would be difficult to differentiate among our different variables of interest and reach any meaningful insight.
It's important to note the limitations of this approach.
We are not trying to use statistical techniques to confirm a causal relation between English proficiency and vaccination rates. Instead, our objective here is to use visualization to explore those relations.
Context
Zip codes 60623 and 60608, more closely associated with Little Village and Pilsen, are indeed characterized by higher percentage of residents with limited English proficiency compared to other neighborhoods. That also seems to be shared by most of the North- and South-west sides, where we encounter other largely immigrant neighborhoods in Chicago.
Little Village and Pilsen have more COVID-19 vaccination locations than most of the South Side, but with less spots when compared to the North Side.
It is important now to look at the vaccination rates by zip code so we can start to explore (1) its possible correlation with the presence of vaccination sites.
Step 1: Vaccination Site Counts and Vaccination Rates
Map 2 mostly confirms our previous prediction in (1), regarding a possible association between concentration of COVID-19 vaccination sites and adult vaccination rate.
Areas associated with Little Village and Pilsen, zip codes 60623 and 60608, as most of the Southwest region of Chicago, have higher vaccination rates than the South Side, but lower than the North Side or downtown.
Such association makes sense, but for our purposes it raises other questions. How can we further explore (2), whether areas with similar number of vaccination sites but different levels of English proficiency rates have different adult vaccination rates?
Additionally, the lower presence in the West and South Sides of vaccination sites itself might be interpreted as a summation of various social forces, including systemic racial, ethnic and potentially linguistic discrimination. Those are our confounding factors.
Step 2: Vaccination Sites and English Proficiency
On Map 3 we can take a closer examination at Little Village and Pilsen, represented here by zip codes 60623 and 60608, and compare them to their immediate surroundings.
On the left side of the image, we can see a map of the COVID vaccination site counts per zip code. Zip codes 60623 and 60608, plus 60632 and 60609 directly south of them, stand out for a relatively higher number of vaccination sites in their boundaries, in addition to a higher proportion of residents with limited English proficiency.
Unfortunately, we are still left with little clarity about the specific role of linguistic access to COVID vaccination rates in Chicago.
Step 3: Insights?
In this report I tried to explore the role of linguistic barriers to healthcare in explaining differing adult vaccination rates across the city of Chicago. To achieve that, I focused particularly in the largely Latin American neighborhoods of Little Village and Pilsen.
Throughout the analysis, however, it became clear that it was hard to differentiate linguistic access from other variables with the data and the methods at hand.
Sometimes the data available does not provide the contrasts we need to draw meaningful insights. The most straightforward variable of interest at hand remained the counts of vaccination sites by zip code, which is arguably itself enmeshed in issues of racial, ethnic, class and language discrimination.
Here you can download a complete version of this report:
Comentarios