Not Under My Roof: Parents, Teens, and the Culture of Sex (39 page)

These ABCDs of adolescent sexuality call for the bridging of oppositions that have characterized the politics of adolescent sexuality in the United States in the post-1960s-era—oppositions that have resulted, in part, from different interpretations of and responses to the changes in the relation- ships between parents and children, men and women, and citizens and governments since then. To overcome this adversarial political climate, some of which has been driven by a tug of war over whether teenagers are to be treated as children or as adults, we need to recognize that adolescents, like all human beings, need strong social bonds. To provide youth with such bonds—at an interpersonal and societal level—is the work of us all.

Methodological Appendix

Despite several obvious differences between the two countries—in size, in ethnic and racial diversity, and in geographic range—the United States and the Netherlands share some surprising commonalities.
1
Both countries were early republics with a highly decentralized state. In fact, the American and Dutch struggles for independence had much in common.
2
Indeed, it has been argued that the American Declaration of Independence was mod- eled after the Dutch Act of Abjuration (
Plakkaat van Verlatinge
) from Spain.
3
In addition, the middle classes have historically had a powerful position in both countries. In the absence of a strong aristocracy, commercial elites in the Netherlands and the United States achieved an early and unusually strong hold on the countries’ culture and politics.
4
Finally, in both coun- tries Protestants long formed the dominant religion but exerted their in- fluence within the context of religious pluralism, with Roman Catholics constituting a significant religious minority.

Building on these cross-national commonalities, my goal for this re- search project was to pursue samples consisting of three cells in each country—parents, boys, and girls—that would be as similar as possible in terms of socio-demographic characteristics. I focused on the secular or moderately Christian white middle class, recognizing that in both coun- tries religious conservatism significantly impacts attitudes toward sexual- ity, and that social class and education also affect childrearing practices.
5
To avoid interviewing only members of the upper middle class, I used a broad definition of middle class—including both those at the “lower” and “higher” end of the spectrum: salespeople, for example, with little or no college education and professionals with advanced training. Avoiding po- litical and geographic extremes, I sought out medium-sized cities and bed-

room communities, neither in large metropolitan centers nor in remote rural towns.

Yet, in search of perfectly cross-nationally comparable samples, I learned much about the political, economic, and cultural differences be- tween the two societies. First, it proved extremely difficult to gain access to high schools in the United States—a significant stumbling block since high schools served as the entry to reaching almost all teenagers, and, through them, in turn, many of the parents I interviewed. Describing my study as one about “growing up in the United States and the Netherlands,” and list- ing among the topics “school, work, alcohol, relationships with parents, relationships and sexuality, and adulthood,” all of which I covered in the interview, I found few schools, even those located relatively near the San Francisco Bay Area, responsive to my queries. Illustrating poignantly the political battles brewing around teenagers and sex education, one prin- ciple said he wanted to avoid all controversy and that “all it takes is one parent.”
6

The second, perhaps more surprising, challenge I encountered was finding a community that would, at least according to its descriptive de- mographic statistics, promise to render a sufficient pool of “middle-class” interviewees. Using census data, I found that most towns, suburbs, and medium-sized cities either appeared decidedly below or decidedly above “middle-income” if measured in terms of median family income. Espe- cially in Northern California, communities were separated into upper- and lower-income extremes, reflecting the hollowing out of “the middle of the middle” during recent decades.”
7
When I gained entry to schools in Corona, a medium-sized Northern California city, I found that interview- ees often lived either in spacious upper-middle-class houses or in house- holds with some economic instability.
8
And even in the town of Tremont, where the cost of living was lower by comparison, it was hard to recruit lower-middle-class parents in part because their children seemed more re- luctant to refer them.
9

Third, although I set out to exclude evangelical and fundamentalist Christians from my sample, I was not always successful. As we saw in the book’s introduction, this group of Christians differs from the rest of the American population in categorically opposing sex outside of marriage. While a much smaller proportion of the overall population in the Nether- lands, conservative Christians also have markedly different sexual attitudes than do the rest of the population. Thus, it made sense to try to exclude them for maximum comparability across nations. But especially in the small town of Tremont, I found that the screening questionnaire failed to

adequately identify a potential interviewee’s church as “conservative,” and only in the course of the interview did I discover that he or she could be categorized as such. Thus, a couple of the American interviewees do, in fact, live in Christian conservative families.
10

My experience searching for interview sites and soliciting interviewees was quite different in the Netherlands. First, almost all schools that I ap- proached were welcoming. Sometimes through a personal connection and sometimes without, I gained easy access to classrooms. That I was engaged in postgraduate research and affiliated with a major research university in the United States sufficed in vouching for my legitimacy. This easy access bespeaks a feature of Dutch society of the 1990s, which I discuss in chap- ters 2 and 4, namely a trust in elites, including medical, scientific, and edu- cational elites. This trust—and indeed confusion when confronted with a suggestion of lack of trust—became evident in an interview with a some- what older and less well-educated Dutch couple. When, after giving me a friendly welcome, they received the consent form describing the research— its “risks” and “benefits,” as required by the research protocol—the hus- band was a little indignant: “We agreed to do it, why do we need to read this?”

The second difference was that it was easy to find the group that proved elusive in the United States. In the Dutch lower-middle-class families in which I did my interviews, at least one parent had typically received some vocational education in addition to a high-school degree and had a white collar job. Parents’ educational credentials ranged from high-school de- grees with on-the-job vocational training to degrees from vocational (non- academic) colleges. They worked as salesmen, bank clerks, insurance work- ers, and middle managers. The solidity of the interviewees’ class position had perhaps less to do with the specifics of their education and occupa- tion than with the comforts and security that characterized the rhythms of their everyday lives. Fathers typically worked nine-to-five jobs, while moth- ers were homemakers or worked part-time. Typically, such families would take at least one lengthy vacation a year, often abroad.
11
And while having children in postsecondary education might require some budgeting, it was certainly within reach.

As in the United States, in the Netherlands I also interviewed physicians, teachers, civil servants, upper-level managers, and psychotherapists.
12
But in the Netherlands, parents with university degrees and lower-middle-class parents with little or no college were almost equally represented among the interviewees, whereas in the United States the parents with four-year degrees outnumbered those without by almost three to one. Consequently,

the American parents were, on the whole, more likely to be highly educated and hence, one might expect, more liberal. At the same time, the American parents and teenagers were more likely to have some religious affiliation and practice. There were similarities between the Corona and Norwood interviewees and the Dutch interviewees—both were usually Catholic or nonaffiliated. But in Tremont, parents and teenagers were typically Protes- tant and reportedly active in their church communities.

There were also several common patterns across the two countries. In soliciting interviews, I invited both parents to be interviewed as a couple. But, in practice, fathers were present and participated in only about a third of the interviews, usually with their wives, but occasionally alone. I did not set out to analyze parents by their gender, nor did the small numbers permit any systematic comparison.
13
In keeping with the experiences of previous researchers, I also found teenage boys somewhat more difficult to recruit than their female counterparts—though not necessarily more dif- ficult to converse with.
14
Finally, in both countries, teenagers were some- what more likely than their parents to live in families in which neither parent had a four–year degree or a professional occupation.
15
One reason for this discrepancy between teenagers and parents is that about half of the of the parents were recruited through informal and formal parent networks such as PTA’s, in which highly educated parents were overrepresented. I approached the remainder of the parents after interviewing their children and receiving the latter’s permission to contact them. Because I interviewed more teenagers than parents and because I recruited parents through mul- tiple methods, only fourteen of the thirty-six American teenagers and fif- teen of the thirty-six Dutch teenagers lived in families in which I also inter- viewed one or both of their parents.

Most of the recruitment of teenage interviewees took place through schools. In the United States, I selected public high schools which draw a substan- tial proportion of their students from middle and upper-middle-class neighborhoods. In the Netherlands, most of the high schools I selected included the pre-college tracks (
HAVO
and
VWO
) disproportionally popu- lated by students from lower- and upper-middle-class families, although two Dutch interviewees were students in the vocational track,
MAVO
. I so- licited interviewees after entering tenth-grade classrooms and explaining the interview and the topics it included. Usually, about half of the students in a classroom wanted to be interviewed. Interested students filled out the

screening questionnaire. They received a consent form for themselves and one for their parents. I interviewed students who met the sampling criteria and whose parents gave permission for the interview.

All interviews were semi-structured and in-depth. For such interviews to be successful and positive experiences, interviewees must feel comfort- able talking about personal matters. I took a number of steps to create an atmosphere in which the interviewees would experience such a level of comfort. I asked them to select the place where the interview would take place. Most teenagers chose to be interviewed at home, in a private office at school, or in a café or public library. Virtually all interviews with parents took place in their homes. To give interviewees a sense of what to expect from the interview, the consent form, which they read and signed before the interview, explained the topics that would be covered in each interview and how confidentiality would be maintained. Before each interview— which was tape-recorded—I emphasized that interviewees could decline to answer any question they did not want to answer.

Each person or couple was interviewed once. The interviews themselves were structured to gradually establish rapport. With the teenagers, I started with a number of concrete, easy-to-answer questions on their day-to-day schedules, school, friends, and jobs. Only gradually did I move into top- ics such as drinking, parental rules, and sexuality, topics which they might find more difficult to address. I concluded each interview by asking teen- agers questions of a less personal nature: their ideas about adulthood and their own future plans. After the interview, I asked them permission to con- tact their parents for an interview. With the parents, I started each inter- view with a number of open-ended questions about raising children and the adolescent phase in particular, only gradually moving into the more emotionally loaded territory of parent-teenager conflicts, drinking, rela- tionships, and sexuality. Again, I concluded each interview with questions about the parents’ wishes for their children’s futures, questions intended to be more general and easy to answer.

The interview questions were designed to elicit explicit opinions and at- titudes. But equally if not more important were the concepts, expressions, and mode of reasoning which interviewees used, often entirely unselfcon- sciously, to express implicit cultural assumptions and beliefs. To gauge ac- tual practices—which are sometimes different from stated intentions and beliefs—I asked questions about interviewees’ behavior in multiple ways at different times throughout the interview. In an attempt to get as close as possible to actual behavior, I asked interviewees to reflect on what their ac- tions were or would be in concrete situations—either real or hypothetical.

Thus, the interviews sought to establish a number of different “truths”: that is, the interviewees’ opinions, their taken-for-granted cultural languages and conceptions, their purported behavior, and finally, their actual prac- tices and lived experiences. The tensions and contradictions between these different truths themselves constitute valuable sources of information.

Each sociological interview constitutes a balancing act between accu- rately establishing the particulars of one individual’s perceptions and ex- periences and eliciting the kind of information that can be meaningfully compared across multiple cases. To balance the objectives of depth and systematic comparison, I sought to flexibly combine different types of in- terview questions. The first type of question lent itself for open-ended an- swers in which the interviewees had free rein to use their own language and convey meanings which would allow me to discern qualitative patterns in cultural meanings and practice. The second type of question was concrete and required clear-cut, often yes-or-no answers. For each important sub- stantive topic area—alcohol, parent-teen relationships, sex and relation- ships, adulthood—I included two or three of the latter.
16
Using the same language, I tried to ask all interviewees these standardized questions.