Ethnographic research in Mexico and the United States

This ethnographic, longitudinal study draws on participant observation, interviews and surveys conducted during numerous fieldwork seasons in rural Puebla and in the United States. Long-term engagement with Pahuatecos/as and Zapotitecos/ as allowed us to forge enduring relationships with men, women and children who invited us into their homes on both sides of the border to share their experiences. Our multi-sited, transnational research allowed us to situate these experiences in broader economic and political transformations that were dramatically changing life circumstances for people in Mexico and the United States (Marcus, 1995). Qualitative work allowed us to unveil the tensions and inequalities that underlie the formation of this new class of precarious workers, both in relation to capital, as well as outside the production process itself, taking into account the intersection of this process to social reproduction. Our strategy allowed us to follow the course of our subjects’ movement across borders, not only in a single direction in terms of migration-settlement from South to North, but also the instability and contingency of migrants’ “illegal” insertion.

While the book draws primarily on research conducted since 2010, we had extensive research experience in Pahuatlán and Zapotitlán before the economic crisis. This allowed us to identify patterns of change over time in mobility and the meaning and experience of migration for a region that became highly dependent upon transnational migration beginning in the 1990s. For example, Lee conducted fieldwork in Zapotitlán for 20 consecutive months from January 2003 to August 2004 and in New York City from September 2004 to February 2005 as part of her doctoral thesis research. In the village, she lived with two families whose male head of household was in the United States at the time of the research, participating in special events such as religious rituals, school presentations, civic ceremonies and birthday parties as well as mundane, routine activities such as meal preparation, taking children to school and food and clothing shopping. She conducted interviews mostly in people’s homes. In New York, she talked with people at the restaurants where they worked (usually during the lull between lunch and dinner) and conducted interviews in migrants’ apartments or in restaurants and delis.

From 2007 to 2010, D’Aubeterre and Rivermar conducted an investigation exploring migration and education in Pahuatlan. This project started after Regina Cortina, a researcher at the University of North Carolina Chapel Hill, shared her research with them. She had observed a rapid increase in the number of children from rural and indigenous towns in the municipality of Pahuatlan attending public schools in Durham. In the 2007-2008 school year, D’Aubeterre and Rivermar surveyed students in their final year of middle school and high school in four towns in the municipality: Pahuatlan de Valle (the county seat), Atla and Xolotla (Nahua towns) and San Pablito Pahuatlan (an Otomi town).⁶ They conducted focus groups with students, teachers and school administrators to investigate the repercussions of migration in the education trajectories of youth. Their research provided valuable information about household composition, remittances from North Carolina since the 1990s and the migration and occupation of parents in Mexico and the United States. Additionally, it allowed D’Aubeterre and Rivermar to identify the configuration of a migration habitus during three generations, beginning with the Bracero Program (1942-1964).

After the crisis unfolded, D’Aubeterre. Rivermar and Lee initiated fieldwork to collect the data upon which this book is primarily based. We drew on our understanding of regional conditions and reactivated our local contacts to carry out three phases of fieldwork between 2010 and 2014. In the first phase, we applied a modified version of the Mexican Migration Project’s (MMP) Ethnosurvey to 135 households in Pahuatlan de Valle (20 percent sample) between August and December 2010 and 170 households (a 25 percent sample) in Zapotitlan in May and June 2011.⁷ On account of our long-standing relationships with numerous contacts in the communities, the refusal rate in both villages was less than 5 percent. In fact, conversations often extended well beyond the questions in the survey. Our research team—consisting of our undergraduate and graduate students—made many brief notes in the margins to follow up on later. In both villages, we used a systematic, spatial sampling method, administering the Ethnosurvey to every fourth (Zapotitlan) or fifth (Pahuatlan) house on each housing block (Schensul & LeCompte, 2013, pp. 301-302). The Ethnosurvey provided basic demographic information about each member of the household along with their migration and work histories. The data was coded and analyzed using SPSS software; it forms the basis for the graphs and tables in the book.

In the second phase (February and April 2011 in Pahuatlan; May and June 2011 in Zapotitlan), we identified individuals (27 in Pahuatlan; 29 in Zapotitlan) who returned between 2007 and 2010 from the results of the Ethnosurvey and through participant observation in the communities. With these return migrants we conducted semi-structured interviews in returnees’ homes with questions designed to elicit information about employment and work conditions prior to migration, border crossings, working and living conditions in the United States, family relations and remittances, reasons for return and future plans. Interviews lasted about an hour. However, some extended to two hours or more, as interviewees discussed some topics in greater depth than we had initially anticipated.

Finally, in the third phase we identified 16 households in each community to follow for an 18-month period to identify and assess the strategies employed to weather the economic crisis. The households were divided into four categories and four households were assigned to each: (1) households with at least one voluntary return migrant; (2) households with at least one forced return migrant; (3) households with at least one migrant still in the United States; and (4) households with no active migrants in the past five years. Group 4 was a control group, since we believed that households in this group would be the least affected by changes in migration experience due to the economic crisis because they were not dependent upon remittances. We developed the interview schedules, which varied slightly because of the different household situations, based on the responses we received in Phase 2 of the research and other insights we gained as the research project developed and matured. We scheduled six interviews with household members (usually the female and/or male heads of households) at approximately three-month intervals. Inspired by Hirsch’s methodology (2003), we believed that revisiting the same households over time would increase rapport and make it easier for household members to discuss sensitive issues with researchers. This was largely the case. Research assistants transcribed digitally recorded interviews, and these were coded using NVivo software.

Although not initially contemplated in the original study design, Mario Macias, a member of the research team, conducted several interviews with Zapotitecos/as in New York during July and August 2014. D’Aubeterre and Rivermar conducted research in Durham among Pahuatecos/as they had originally met in Mexico and their extended family members in October 2013 and May 2014. They interviewed members of 13 households, three of them linked by kinship. Furthermore, they informally interviewed acquaintances of these groups, or people close to them, during casual encounters in shops and family-owned restaurants. This strategy allowed them to collect information about everyday life, work, routines and social environments and to observe consumption practices and visit businesses owned by Pahuatecos, which have flourished in Durham.