The fact that I was able to reach 100 percent of the families is a strength of the follow-up study, as is the fact that I gained the cooperation of all of the young adults and most of the parents. Nonetheless, even this small study was labor intensive. In the end, I conducted nearly forty follow-up interviews. Then I arranged to have the interviews transcribed; next came coding and analyzing the data; and finally, the struggle to write up my findings. In short, the project was a major undertaking in which I invested significant time and energy. Nevertheless, I am fundamentally dissatisfied with the data set. Because my criticisms are conceptual they are relevant for other studies.
Social scientists see a longitudinal follow-up to an ethnography as having many potential virtues, including the ability to assess the degree to which original theoretical conclusions are sustained over time. The in-depth interviews I conducted were revealing, but they provide few surprises in terms of the youths’ and families’ trajectories. Over time, the inequalities in family life grew, rather than shrank. Although some of the working-class and poor families made important gains, the power of social class remained considerable. Hence, the follow-up supports the basic argument of the original study. These are interesting findings, and they may help satisfy readers’ curiosity about what happened as the youngsters profiled in the text grew into adulthood. But, like all other longitudinal studies based solely on interviews, the follow-up has important limitations. The research design precluded collecting the deeper, richer, and, I believe, ultimately more valuable data that come from participant-observation of the rituals of daily life. The original study, because it involved participant-observation across multiple settings, embedded the families and kids in a social context. The longitudinal follow-up isolated the young adults and families from the social context.
This is a crucial methodological difference.8 With no observations of daily life (and no interviews with educators or other key people), the longitudinal follow-up lacks the critical institutional information and triangulation of data that characterized the original study. This severely limits its value. The lack of institutional checks weakens the interview findings. There was no way for me to confirm the young adults’ portrayal of events; and it was impossible to ascertain the accuracy of information conveyed by family members about key life transitions. And, particularly compared to the observational data collected in the original study, the interviews shed less light on a fundamental point: that differences in social class matter because they provide unequal advantages in key institutions.
WHAT I WISH I COULD HAVE DONE
In hindsight, I wish I had visited the kids when most were in their senior year of high school to do observations; gather school transcripts, SAT scores, and college applications; and conduct interviews with key teachers, coaches, and counselors. But even as wishful thinking, it is hard to imagine. It would have been too formidable a task, for a variety of reasons. The nine youths featured most prominently in the book attended eight different high schools; the full sample of twelve covered ten different high schools. In the years since I began the original study, the paperwork requirements for doing research have escalated. Applications to the committee for the protection of human subjects (commonly known as an Institutional Review Board, or IRB) are much more detailed than in earlier years. The IRB must review and approve researchers’ interview guides, consent forms, letters of solicitation to recruit participants, etc. Moreover, the application process for research in schools is cumbersome; the permissions alone typically take many weeks. Likewise, negotiating access with districts, principals, and families involves countless hours and a wide array of challenges.
As Mitchell Duneier has pointed out, many of the best ethnographic studies have been doctoral dissertations.9 These projects were not carried out by seasoned researchers but by novices who, despite their inexperience, have tremendous advantages. One is that frequently they are at a life stage that allows them to spend an enormous number of hours in the field. This immersion is often crucial for the establishment of rapport with participants and the subsequent development of rich theoretical insights. In the youths’ last year in high school (2001–2), I was still writing Unequal Childhoods. The following year, although I continued to think about beginning another round of observations and interviews for all of the youths, I faced many professional and personal obstacles to undertaking a labor-intensive study. I had significant teaching, advising, and professional responsibilities.10 The normal obligations of family life were heightened by the upheaval associated with the unexpected need for immediate and extensive home repair work. In addition, I faced personal challenges that year, with the deaths of my mother and a close family friend. Qualitative research intrudes further into the researcher’s personal life than does quantitative research, in that vital aspects of qualitative research are interpersonal rather than distanced. Despite my desire to reconnect with the twelve kids and to gather information about the key institutions in their lives, the prospect of launching another major study seemed overwhelming.
As with the initial study, a key problem presented by the follow-up was that the project was too big. With ethnographies, the more typical approach is to focus on one site—or even one family.11 With twelve families in the original study, Unequal Childhoods was too ambitious; subsequently, the longitudinal follow-up was also overly ambitious, which created complications at every point. If there had been only three families in the original study, I might have managed to gain access to the schools, follow kids around, and re-immerse myself in participant-observation. Doing that kind of follow-up with twelve (or even nine) families was not feasible.
There are, fortunately, some aspects of the longitudinal follow-up about which I remain enthusiastic. It was very helpful to examine the youths’ trajectory over time. The evidence of continuity rather than deflection in the trajectories is striking. Still, as I have tried to make clear here, there are very significant differences between the information yielded by interviews and the information yielded by participant-observation. Given the labor-intensive nature of participant-observation and the increased institutional demands on researchers, interviews are more common, but, despite some valuable features, interviews are inevitably less revealing about the rituals of daily life than are observations.12
THE COST OF RESEARCH: REACTIONS TO THE BOOK
The process whereby I learned what families thought about Unequal Childhoods unfolded over several months. As previously noted, the longitudinal follow-up consisted of in-depth interviews with all of the young adults, their parents, and, in most cases, one sibling. In some cases, such as with the Marshall family, I finished all of the interviews with the family members in the summer and then dropped off a copy of the book in the fall. Other times, I brought the book to the interview, and then when I came back to do another interview, I listened to how the families felt about the book. Some of the time, I heard what family members thought during telephone conversations (usually when I was calling to arrange another interview). Other times we were face-to-face. Sometimes I simply dropped by unannounced a week or so after having dropped off a copy of the book, just to see what the family thought. I usually brought food, such as a cake or a pie, when I came.13 In a few cases, I brought a tape recorder and taped the family members’ reactions to the book. “I want to be sure that I understand exactly what you are saying,” I would say. “Is it okay if I tape?” I also told the families that the second edition was going to include a new section, where I would summarize their reactions. In asking their permission to tape, I explained that I wanted the new section to be as accurate as possible.14