What Can We Learn About Mexican Immigration From Restaurant Names? (Part 2)
In part 1, I described the goal of determining the home regions of Mexican immigrants in the United States. I provided context for the question and gave an outline of how I would attempt to answer this question. I explained why I chose these cities in particular and why restaurants can (partially) display the origin of the people living there. In this blog post, I will examine the results and discuss their significance in helping us understand why immigrants settle where they do.
Analyzing the Results For Each City
The results were gathered in a Pandas Dataframe and compared between each city. Each match was checked using wikipedia to make sure that the matches were legitimate (e.g.: “gonzalez” almost definitely refers to a last name and not a town of 2,000 named “Doctor Gonzalez”). As we can see below, there were many matches in Dallas, and very few in Chicago. It seems that the number of matches in a restaurant name to a region is quite variable and random.
Returning to my original question, there were clear differences in the regions represented by restaurants in each city, both in terms of percentage and number of restaurants. The most prevalent region in any city was Puebla, which accounted for 46% of all matching restaurants in New York City (which is to be accepted when there are over half a million Pueblans in NYC!). The next most prevalent regions were Michoacàn with 31% of Dallas restaurants and 27% of Chicago restaurants.
For a broader view of the relative differences in the regional representation of the Mexican restaurants in the U.S. cities studied, I created a map. Here, it is quite obvious that the dominant region for each of the U.S. cities is different in every case! What is clear, is that most of the regions represented are from the Bajio and Central Mexico, a point I will come back to later.
Another question one might have is: which U.S. city is the most diverse in terms of the regional representation of their Mexican inhabitants? In order to answer this question, I picked 30 restaurants at random for each city, and counted how many regions and culinary regions were represented by those 30 restaurants. This was repeated 5 times and the average was taken. Since San Francisco had so few matches, it was excluded from this analysis. As we can see below, the cities are all quite diverse, with Dallas having the most regions represented and Chicago having the most culinary regions represented.
Analyzing the Combined Results
We have looked at the results for each city individually. How do the results look if we consider all of the results together? In other words, of all the Mexican restaurants in Chicago, Dallas, New York City and San Francisco, which regions in Mexico send the most immigrants? Turns out, Michoacán, Jalisco, and Puebla are the most represented regions in this sample of restaurants (all in the darker shades of red, left). This data is very well correlated with the total remittances sent to Mexico (of which, 94.1% comes from the U.S.): Michoacán, Jalisco, and Puebla are all in the top categories in the two graphs. The connection between Mexican restaurant regional connection and actual immigration is clearly correlated.
The restaurants in this project show a clear trend in terms of Mexican culinary regions: the southern and Gulf regions of Mexico were not well represented (only 15% when combined). The most populated region was the Bajio, with the other regions (Central, North and North Pacific Coast) all coming in at around 15% of the total. Interestingly, a majority of the restaurants representing North and North Pacific Coast Mexican cuisine where found in San Francisco and Dallas. This shows a potentially different migration from Mexico to the South/Southwestern United States for people from Northern Mexico. Overall, however, we see a somewhat diverse view of the origins of Mexican cuisine in the United States today.
Another interesting fact emerged from the data: the connections between Mexico and the restaurants in this sample is extremely specific to a region. For example, five restaurants had “Tulcingo” in their name in NYC but it didn’t appear in any of the other cities. When looking into this, I determined that the (Pueblan) town “Tulcingo del Valle” has around 9,000 residents within Mexico! In another small Pueblan town, the mayor is quoted as saying: “maybe three out of four of my constituents live in New York.” These cases show the drastic impact chain migration has on where people live.
Setbacks of this Project
This project is meant to be an initial look into the regional origin of Mexicans in select U.S. cities. This project, of course, rests on a few assumptions to prove this:
- The Mexican restaurants in this study are representative of the owners/workers in that restaurant
- Mexican immigrants of all regions of Mexico will open restaurants in equal proportion
- All restaurants have an equal likelihood of containing a name that is identifiable with a region
All of these assumptions, of course, are not rock solid. Thus, my conclusions about the origin region of Mexican immigrants if one started from a different assumption/data source (remittances to regions of Mexico would be an excellent one). Thus, my results should be understood as an educated guess to the origins of Mexican immigrants within the U.S.
The words matching restaurants and regions is also a source for error. I, a non-native Spanish speaker from the U.S., am clearly not the right person to decide where a match is legitimate or not, but needed to do so in the context of this project. While I tried to be thorough in checking for connections (using wikipedia to do so), there were still judgment calls that needed to be made. For example, some names referred to towns in two regions. Which to choose? I chose the town with the larger population, but this is potentially incorrect without a closer look. Additionally, the formerly-Mexican U.S. states of California and Texas made deciding between a legitimate match and an incorrect one slightly more difficult. Did a match refer to a part of the U.S. or Mexico? For example, a potential match in San Francisco included “Monterey.” Is this referring to the city in California or the state in Mexico? Luckily the spelling (only one “r” in the California city) made this clear, but many other cases were less so. It is my hope that despite these potential errors, overall, the same picture would emerge under someone else’s analysis.
Conclusions
Mexican immigration to the U.S. has a long and important history to the development of both countries. Mexican labor helped make the U.S. what it is today while also helping immigrants move up the social ladder. Additionally, the money immigrants made also helped, and continues to help, develop Mexico. By analyzing the distribution of regions represented in Mexican restaurant names in different U.S. cities, we have one small way of measuring where Mexicans in the U.S. came from. We see that the origins of Mexican immigrants is quite varied across all U.S. metro areas. Certain regions are more represented in some U.S. cities than others (e.g.: Puebla in NYC) but immigration from that region is not necessarily limited to that destination. At a more micro-level, there are cases where chain migration has created pockets of Mexican villages within the U.S., creating a second home for that village in the U.S. Overall, central Mexican regions are the most represented in the U.S. (and received the most remittances). Keep this in mind next time you are eating at your local Mexican restaurant!
To get more in the weeds of this project, check out the github repository. There, you can find my Jupyter notebooks, and the final list of restaurants (let me know if you disagree with my links to Mexican regions!), among other things. The map of regional connections with a U.S. city can be found in a Tableau Dashboard, too.
If you want to learn more about Mexican immigration to the U.S., I encourage you to read this blog post. Additionally, if you are interested in this idea of restaurant counts and their relation to an ethnic group, be sure to check out my blog series “Are Ethnic Restaurants in Ethnic Neighborhoods?” (Part 1, Part 2).
On March 12th, 2021 this article was published in El Diario NY.
