71
experiences from the very individuals who lived and experienced them, all of which requires a relevant project design.
72
be flexible to allow for exogenous changes, including political or legislative ones. Accordingly, I neither entered the field with a particular theoretical angle, nor a hypothesis to test (e.g. Atkinson, 1998). I was aware of challenges, but uncertain as to how they would factor into my study.
Additionally, I was uncertain about the theoretical angle I would eventually take to explore this phenomenon. I acknowledge, however, that flexibility and open-mindedness should not be confused with “objectivity.” No matter how open or flexible one attempts to be, even the best researcher cannot remain entirely uninfluenced or unprejudiced by past experiences in current research (Mosse, 2006).
4.2.2 Pre-Planning Fieldwork
When I began my Ph.D. in September 2011, my first priority was to contact individuals and
organizations in the U.S. to cultivate relationships and pre-plan fieldwork with key gatekeepers who could assist in accessing interviewees. Through oral and written communication, I discussed research goals, my background, and answered any questions gatekeepers or potential interviewees before entering the field. Having my primary research location in Denmark while researching 1.5GUY in the U.S. does present challenges, but with consideration of limits, planning, and efficiency, these
challenges can be mitigated. Further, I understand that “the ideal site for investigation of the research problem is not always accessible. In that event, the researcher accepts and notes the limitations of the study from the onset” (Fetterman, 1998, p. 32). Thus, I acknowledge the challenges and limitations of my research within the next few sub-sections.
With a research base in one continent and the field in another, processes of entering and exiting the field are distinct temporal and physical processes that do not easily allow for re-access if sufficient data is not collected. I needed to plan ahead and focus on recruitment. Further, I did not want to rigidly limit myself to particular demographics to the detriment of sample size and saturation. I initially planned a pilot project from February – March 2012 mainly to connect with organizations and establish contacts, whereby laying the foundation for a longer, more extensive visit the following fall. However, I managed to conduct eighteen interviews during this time and conducted a second round of fieldwork during September – December 2012.
73 4.2.3 Geographic Location
Based on my analysis of existing research on the 1.5GUY, I observed the particular geographic gap in research outside the state of California. The geographic and ethnic diversity gap within research has also been recognized by Cebulko (2014), who wrote “in sum, we are short on theory and data on the material and nonmaterial consequences of…1.5-generation immigrants who are not of Spanish-speaking Latin-American origin and do not live in California” (p. 145). Due to the documented geographic gap, I focused my efforts mainly in the Northeast in Connecticut, Massachusetts, New Jersey, New York, and Pennsylvania, but also conducted fieldwork in Texas due to access through a contact. I aimed to conduct fieldwork in a variety of geographic contexts to uncover a diversity of experiences—also a documented gap. I am originally from the Northeast, so the geographic focus also fulfilled a personal interest. When discussing research location with friends, family, and colleagues, I was often confronted with surprise; the response was often “are there really illegal immigrants in [northeast state]?” The public gap in knowledge of undocumented immigration in these areas helped fuel personal and academic interests.
While it is important to outline some state dynamics and demographics, I emphasize that neither
making a cross-geographical comparison nor in-depth context of reception analysis are the focus of this dissertation. States are diverse in a number of ways: the concentration of the undocumented
population; growth or decrease of undocumented populations in recent years; in-state and non-instate tuition for higher education; and in country of origin. Three of these states are amongst the most highly populated by the undocumented population in 2012: Connecticut (3,5%, 14th greatest share); New Jersey (5,8%, 4th greatest share); and Texas (6,3%, 3rd greatest share) (PEW Research Center, 2014).
The top five countries of origin for undocumented populations are also diverse in each state.20 While the undocumented population grew in Pennsylvania between 2009 to 2012, it decreased in
Massachusetts and New York (PEW Research Center, 2014). Half of these states allow undocumented residents to pay in-state tuition rates at public universities with proof of residency: Connecticut (since 2011), New York (since 2002), and Texas (since 2001) (NCSL, 2014). I initially anticipated focusing on experiences related to higher education, hence the attention to equality amongst in-state tuition
20 Data from Migration Policy Institute (2013).
74
policies. However, I decided against this focus, particularly as these challenges are amongst some of the most documented.
Figure 3 Top 5 countries of origin in each state where fieldwork was conducted21
21 Source: Migration Policy Institute (2013).
0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
Mexico Guatemala Ecuador Brazil El Salvador Brazil Brazil Guatemala El Salvador China Dominican Republic Mexico El Salvador Honduras Guatemala India Mexico China Ecuador Dominican Republic Guatemala Mexico Guatemala Ecuador India Dominican Republic Mexico India China Guatemala Dominican Republic
Connecticut Massachusetts Texas New York New Jersey Pennsylvania
Top 5 Countries of Origin for Undocumented
Population
75
Figure 4 Interviewee’s Country of Origin
While I attempted to have equal representation from each state, this was not always possible.
Furthermore, migration is rarely a completed act—a point made evident by the fact that many of the youth I interviewed have internally migrated within the United States. This is particularly the case amongst 1.5GUY who have moved across state borders, if not across the country, to attend a four year private university. Thus, locations indicate residency at the time of the interview. I acknowledge that contexts influence experiences, but as my aim is not to draw cause-effect conclusions, I maintain that the limitations presented by geographic selections are within the aims and methods of my research.
4.2.4 Accessing “Hard-to-Reach” & “Vulnerable” Populations
There are a number of stigmas associated with undocumented immigration, (e.g. Abrego, 2011;
Cebulko, 2014; Chavez, 2008; Coutin, 2005; Suárez-Orozco et al., 2011) and researchers have conceptualized undocumented individuals—and especially children or youth—as vulnerable (e.g.
Abrego, 2006, 2008, 2011; Abrego & Gonzales, 2010; Buff, 2008; Bhabha, 2011; Capps, Fix, Ost,
Brazil 18% (6)
Chile 9% (2) Colombia
3% (2)
Ecuador 12% (4)
El Salvador 3% (1) Mexico
46% (15) Peru 6% (2)
Philippines 3% (1)
Interviewee's Country of Origin
76
Reardon-Anderson, & Passel, 2004; Carens, 2010; de Genova, 2004; Gonzales, 2007, 2009a, 2009b, 2010, 2011; Olivas, 2009; Perez, 2009; Perez, Espinoza, Ramos, Coronado, & Cortes, 2009; Seif, 2011; Suárez-Orozco et al., 2011; Thomsen, 2012). These stigmas and vulnerabilities present additional challenges for researchers, including time and cost (Gonzales 2011). Gonzales (2011) has argued that “today’s anti-immigrant climate and localized immigration enforcement present challenges to finding respondents” and that “until very recently, it has been difficult to study undocumented young adults…their numbers have been prohibitively small” (p. 606).
In other fields, researchers have paid special attention to the concept of “vulnerability.” In a study on crowding and population, Loo (1982) wrote that relative to the majority, vulnerable people “have less power, opportunity, and freedom to determine outcomes in their lives or to make decisions that affect their situation due to their age, physical or mental condition, race, economic, or political position, or a captive status” (p. 105). In a study on nurses and illnesses, Moore and Miller (1999) argued that vulnerable individuals “are those who lack the ability to make personal life choices, to make personal decisions, to maintain independence, and to self-determine” (p. 1034). Legal scholar Fineman (2012) defined vulnerable peoples as individuals who lack autonomy and independence; who are likely to be monitored, disciplined, and supervised due to external perceptions that they are deviant, dangerous, at risk, or in need of control; and who are designated as vulnerable due to poor choices made in the past or those likely to be made in the future. These statements raise, but do not necessarily answer, questions as to how vulnerable populations navigate being approached for qualitative research, including their ability to say no to such research.
Various scholars have suggested that “gatekeepers often heighten their vigilance” (Moore & Miller, 1999, p. 1036; see also Berg, 2004; Feldman, Bell, & Berger, 2004; Wanat, 2008) when protecting research access, settings, and potential interviewees from within vulnerable populations. Going into the field, I was optimistic, but realistic about these challenges. I encountered the greatest roadblocks in communication with adult gatekeepers; in cases where an adult university administrator was in charge of an organization at a university, my access to potential interviewees was completely blocked. I was told that students’ personal information could not be shared due to university protection policies, nor that my details could be forwarded. I understand the legal limits and desire to protect individuals, but wish to problematize this protection and outsider’s perceptions of vulnerability. I agree that 1.5GUY
77
can be potentially hard-to-reach and vulnerable, but not all youth are in hiding, afraid of discussing their ULS and the challenges it brings, or wish to remain hidden or anonymous. That I found some 1.5GUY via newspaper articles prominently featuring their names, faces, or stories at length illustrates some 1.5GUY’s desire to raise awareness and share their struggles, rather than remain hidden. In turn, this raises questions about the value and validity of gatekeeper protection without consent.
Fineman (2012) explained “we are born, live, and die within a fragile materiality that renders all of us constantly susceptible to destructive external forces and internal disintegration” (p. 119). She (2013) later wrote “we will be dependent, weak, in need, as well as empowered and strong at different
developmental stages in our lives” (p. 120). Vulnerability is universal, though experiences are unique.
The relation between vulnerability and access undoubtedly affects the aforementioned lack of qualitative understanding. However, if we are to have access to, expose, and raise awareness of vulnerable populations and their circumstances, we must have access to individuals who are able and willing to share their stories. Treating all individuals as potentially vulnerable, but not incapable of making decisions includes letting them decide if and what to share.
4.2.5 Interviewee Diversity in an Exploratory Study
Knowing about the challenges of accessing vulnerable populations ahead of time, I used a “big net”
approach, “mixing and mingling with everyone at first” to ensure “a wide-angle view of events before the microscopic study of specific interactions begins” (Fetterman, 1998, p. 32, 31). Qualitative
researchers “typically use an informal strategy to begin fieldwork, such as starting wherever they can slip a foot in the door” (Fetterman, 1998, p. 35) and my approach attempted to balance between diversity, flexibility, and access. In communication with gatekeepers and potential interviewees, I stressed the three criteria guiding my study: ULS; arrival in the U.S. at age twelve or earlier, so as to focus explicitly on the 1.5 generation; and a current age of sixteen to twenty-five, so as to focus on youth, rather than fully transitioned adults. I made no further stipulations in terms of country of origin, current location, ethnicity, educational status etc., as I did not want to make accessing a hard-to-reach, stigmatized, and vulnerable population from abroad even more difficult, thus reducing my
opportunities for data collection. I did not ask about ULS until I met with an interviewee in-person, discussed the interview process, and obtained informed consent. Thus, out of thirty-eight total youth, thirty-two are undocumented; two are “allies” or U.S. citizens sympathetic to and engaged in
78
immigration reform; two have Temporary Protected Status; and two have other forms of legality, one of whom is on a temporary student visa and the other is a youth who had been undocumented for most of her life, but now has legal residency. The narratives represented in this dissertation come from thirty three of these individuals: the thirty two 1.5GUY, and the one who was undocumented most of her life.
My recruitment and interviewee demographics are in accordance with an exploratory and
phenomenological study. Stebbins (2013) suggested that the “most efficacious approach is to search for…understanding wherever it may be found, using any ethical method that would appear to bear fruit” (p. 6). Furthermore, Laverty (2003) wrote that the aim “is to select participants who have lived experience that is the focus of the study, who are willing to talk about their experience, and who are diverse enough from one another to enhance possibilities of rich and unique stories of the particular experience” (p. 29; see also van Manen, 1997). Gonzales and Gleeson (2012) stressed we “have not uncovered the diverse sets of undocumented experiences” (p. 3): my aim was diversity. My expectation was that the more diverse the interviewees and the contexts, the greater the potential for diversity of perspectives, experiences, and coping strategies. I recognize that differences in nationality, ethnicity, race, contexts of origin, context of reception, religion, sex, gender orientation, socio economic status, etc. influence experiences. I acknowledge that restricting interviewees to particular characteristics could have allowed for a different focus, if not research outcome, including an in-depth contextual analysis. While context likely shapes experiences, I wanted to explore these experiences, rather than attribute them to a particular cause.