Introduction

Anxiety about Jewish survival has been a constant feature of American Jewish life, shaped even in the last decade by memories of the Holocaust. Through the 1970s, the locus of these concerns was anti-semitism and threats to Israel. Beginning in the 1980s, however, attention shifted increasingly to the perceived threat of assimilation and, in particular, intermarriage. Debate over the nature of the threat and the appropriate response has since become a prominent feature of scholarly and communal discourse. The present paper examines Jewish intermarriage in the context of Taglit-Birthright Israel (Taglit), a program that sends Diaspora Jewish young adults on fully subsidized ten-day educational tours of Israel. The goal is to assess Taglit’s potential contribution to Jewish demographic vitality in the United States and derive lessons about the dynamics of intermarriage that can help shape communal policy responses.

Intermarriage

In a 1963 American Jewish Yearbook article, sociologist Eric Rosenthal noted new, higher than expected estimates of intermarriage rates in Iowa and Washington, DC and commented that the findings “cast doubt on the doctrine of the persistence of religious endogamy in American life” (Rosenthal 1963, cited in Berman 2008, p. 47). The response was swift, with several rabbis and community leaders pronouncing an “intermarriage crisis.” Debate ensued among sociologists about the meaning of the new figures. In a study of assimilation published the following year, Milton Gordon “celebrated the fact that Jews could freely marry non-Jews…and extolled the virtues of a system in which identity was based on individual volition” (Berman 2008, p. 48). What is good for individual Jews, however, might not be good for the Jewish collective, and sociologist Marshall Sklare warned: “It is precisely the ‘healthy’ modern intermarriages which raise the most troubling questions of all to the Jewish community” (Sklare 1964, cited in Berman 2008, p. 48).

The first national data on intermarriage were collected as part of the 1970 National Jewish Population Survey and published in the American Jewish Yearbook. The investigators reported that 31.7 percent of recent (1966–1972) marriages of people raised as Jews in the United States were to non-Jews (Massarik and Chenkin 1973, p. 292). Shortly thereafter, a full page advertisement in the New York Times, placed by the Board of Jewish Education of Greater New York, warned, “If you’re Jewish, Chances Are Your Grandchildren Won’t Be” (Berman 2008, p. 49). But these bursts of concern aside, intermarriage became the focus of sustained communal attention only after publication of the 1990 National Jewish Population Study, which reported that 52 percent of recent (1985–1990) marriages were intermarriages (Kosmin et al. 1991). Jewish continuity commissions were created, and a host of programs to counter assimilation and intermarriage were launched. Although “52 percent” became the focus of discussion, there was disagreement about the actual figure (see Cohen 1994). Not disputed, however, was that since the end of World War II, the intermarriage rate in the United States had risen from about 10 percent of all marriages to approximately 50 percent (see United Jewish Communities 2003).

Notwithstanding the precipitous increase, the current rate of intermarriage is not high in comparison to other ethnic and religious groups in the United States (Phillips and Fishman 2006; Rosenfeld 2008), and its impact on the size and vitality of the American Jewish community is not obvious. Intermarriage is associated with lower levels of participation in various Jewish arenas—in synagogue life, philanthropy, Jewish education, and visits to Israel (see Cohen 2006); but it is unclear whether lower levels of engagement are primarily an effect of intermarriage or its cause, since Jews with weaker Jewish backgrounds and community ties are far more likely to marry non-Jews. In part, because of the difficulty disentangling the causes and effects of intermarriage, research has shifted from a focus on intermarried adults to the trajectories of their children. Recent research has demonstrated that children of intermarried parents who are raised exclusively as Jews have similar levels of Jewish ritual practice, Jewish identity, and engagement with the Jewish community as the children of inmarried parents, after controlling for Jewish education (Chertok et al. 2008; Phillips and Chertok 2004). However, the current rate at which intermarried parents raise their children as Jews is unknown. Most researchers put the rate at which intermarried families raise Jewish children in the United States at lower than 50 percent (Cohen 2006; Fishman 2004), in part basing their conclusions on the findings of the National Jewish Population Study (United Jewish Communities 2003). Although estimates based on these data have been challenged (see Kadushin et al. 2005), if correct, they imply that the Jewish community is losing population due to intermarriage. But this is not the case in all communities. Jewish population studies in Boston (Saxe et al. 2006a), Cleveland (Raff 1998), and St. Louis (Tobin 1995), among others, reported rates for intermarried parents raising their children as Jews as over 50 percent.Footnote 1

These complexities help to explain why there is substantial disagreement about the implications of a historically high intermarriage rate and the appropriate policy responses. Since publication of the 52 percent figure, the dominant perspective has been that marriage between Jews and non-Jews is a severe threat to Jewish demographic vitality (see, e.g., Cohen 2006; Fishman 2004; Wertheimer 2001, 2005). Intermarriage, in this view, inevitably leads to lowered rates of Jewish identity and engagement. One commentator notes: “Candor requires acknowledging that there is very little good news about mixed marriage. The facts indicate that it means minimal Jewish identity. Children of mixed marriage report even less affiliation than their parents, and grandchildren almost none at all” (Bayme 2006). From this perspective, the best response is for Jewish leaders to advocate inmarriage or conversion by non-Jewish partners, and for the Jewish community to increase its investment in Jewish education (see Fishman 2004; Wertheimer and Bayme 2005). Because this point of view focuses on reinforcing community boundaries and investing in the education of those who are born Jewish, it is often referred to as the inreach perspective.

The alternative view, known as the outreach perspective, treats intermarriage as an inevitable feature of contemporary life (McGinity 2009, p. 20). In this view, Jewish demographic continuity can only be assured if all are welcomed, diversity is celebrated, and barriers to participation in the community are reduced (see, e.g., Dorff and Olitzky 2007; Mayer 1991; McGinity 2009). “The true problem lies not in our choice of life partner or living in a multicultural society,” writes an advocate of this position, “but in trying to find Jewish institutions that will fully embrace our decision to lead meaningful Jewish lives once married ‘out’” (Bronfman and Wertheimer 2009). Intermarriage, in this view, is problematic only if the community fails to engage non-Jews who are members of Jewish families. If the Jewish community embraces intermarried families, intermarried parents are more likely to make Jewish choices for themselves and their children.

Taglit-Birthright Israel

Notably, advocates of both inreach and outreach concur regarding the value of Jewish education and, in particular, informal Jewish education. Accordingly, one response to concern about Jewish assimilation in the Diaspora was the creation, in 1999, of Taglit-Birthright Israel (Beilin 2009; Saxe and Chazan 2008).Footnote 2 Founded by U.S.-based philanthropists, Taglit’s aim was to strengthen Diaspora Jewish identity by enabling large numbers of young Diaspora Jews to have an experiential educational experience in Israel. During the program’s first year, nearly 10,000 Diaspora young adults left for Israel to participate in 10-day educational tours. By 2008, that number had increased to nearly 40,000 per year (Shoshani 2008). In nearly a decade, almost 250,000 18 to 26-year-old Jews from around the world have participated. The largest group of Taglit participants comes from the United States, but more than 50 countries are represented, sending participants roughly in proportion to the size of their Jewish communities.

A Taglit trip involves 10 days in Israel with a group of 40 or so Diaspora peers, led by a guide and educators (see Kelner 2010; Saxe and Chazan 2008). The itineraries are somewhat different by group, but all involve visiting sites that reflect Israel’s Jewish history, as well as its modern development. Historic and geographic sites notwithstanding, the focus is on developing relationships within the group and with Israelis who participate. All groups now include a small cadre of Israeli peers (most of whom are soldiers) who join a trip for five to ten days (see Sasson et al. 2008, in press).

Prior to Taglit, educational tours of Israel were typically tailored to already-affiliated pre-college adolescents (Chazan 2002; Cohen 2009; Mittelberg 1999). Taglit focused on an older, more intellectually mature and less Jewishly-affiliated population. Those who apply to participate in Taglit, particularly from North America, come from a broad swath of the Jewish population—from highly educated and engaged young Jews to those who grew up in nonobservant homes with little or no formal Jewish education.

Evaluation studies of Taglit have consistently shown that the program is highly valued by participants and has substantial impact on their attitudes toward Israel and their Jewish identities (Saxe et al. 2002, 2004, 2006b, 2007, 2009b). Previous research on the impact of Taglit identified significant differences in the importance participants and nonparticipants placed on marrying a Jew and raising children as Jews. An open question, however, was whether these differences in attitudes predicted later behavior, in particular decisions about marriage. Prior research, in the 1990s, by Mittelberg (1994; see also, Mittelberg 1999) found an association between Israel travel and endogamy. Inmarriage was not, however, an explicit goal of Taglit. Because contemporary young adults marry later than their predecessors, it has taken nearly a decade to be able to assess marital choices among program applicants. Ten years after the program’s inception, alumni from early cohorts are now embarking on their mature adult lives and making choices about their Jewish identities and, in particular, marriage and family.

Given the number of Taglit participants, program effects can have a potentially significant impact on the size and character of the American Jewish community. The present study examines Taglit’s effects on intermarriage, conversion, and attitudes regarding raising children as Jews and compares these effects to other educational interventions and background factors. Beyond the specific impact of Taglit, the paper seeks to address two unresolved questions regarding intermarriage dynamics, specifically: (a) Whether the intermarriage rate is an unchangeable feature of contemporary life (as many outreach advocates assert) or susceptible to influence (as many inreach advocates claim); and (b) Whether the rate at which intermarried parents elect to raise their children as Jews is an unchangeable fact of life (as many inreach advocates assert) or susceptible to influence (as many outreach advocates claim).

Methods

In order to ascertain Taglit’s impact both directly and in comparison to other factors, several surveys were analyzed. The first, a 2009 survey of U.S. young adults who applied to participate in a winter Taglit trip between 2001 and 2004, is used to understand the long-term impact of the program on marriage, conversion, and child-raising. Applicants to older cohorts were chosen to allow an assessment of the long-term impact of Taglit and maximize the number of individuals likely to be married and be parents. Two additional surveys, conducted with summer 2008 applicants three months prior to and three months after the trips, are also examined. Designed as pre- and post-trip evaluation instruments, the summer 2008 surveys did not explore the actual marital choices of program applicants. These surveys did, however, pose questions regarding attitudes on marriage and child-raising and, in contrast to the survey of 2001–2004 applicants, asked about a wide range of educational experiences and background factors. The surveys therefore provide a basis for analyzing Taglit’s impact in comparison to other factors.

The three surveys include Taglit applicants who did not participate in the program. These nonparticipants form a natural comparison because there has been substantially more demand for the program than available slots and, for the most part, the reasons that some applicants participated and others were placed on waiting lists were random. Thus eligible nonparticipants have been very similar to those who participated (Saxe et al. 2009a).Footnote 3 By comparing participants to those who applied but did not participate, one can isolate the effect of program participation.

Taglit applicants’ backgrounds cover the spectrum of Jewish experiences, from those who grew up completely estranged from Jewish life to those raised in highly engaged households. Overall, Taglit applicants come from somewhat more engaged backgrounds than those who do not apply to the program, and the sample, while extremely diverse, is not fully representative of young adult American Jews. This is particularly so for those who applied for 2001–2004 trips. Overseas travel in general, and visits to Israel specifically, were seen as dangerous by many applicants and their parents during those years and attracted some of the most highly engaged (Saxe et al. 2004). Later, the proportion of Orthodox applicants decreased, while the percentage of applicants with limited Jewish education and those who came from disengaged households increased (see Table 1).

Table 1 Selected applicant characteristics
Full size table

Survey of 2001–2004 Applicants

The sampling frame consisted of participants and nonparticipants who applied for Taglit winter trips during 2001–2004.Footnote 4 Included were those who went on any one of the winter trips, as well as those who applied but did not go on the trip for which they applied or on any subsequent trip.Footnote 5 Some nonparticipants had, however, traveled to Israel since applying to Taglit, either on their own or as part of another organized Israel program. In all, approximately 22,000 individuals who applied for these trips were eligible for inclusion in this study (more than 85 percent of the total number in the registration database for the cohorts included).

The sample was stratified by age, gender, round (year of application), and participation in Taglit (see Saxe et al. 2009a for details). Older individuals (age 30 and above) were oversampled in order to increase the number of respondents likely to be married and raising children; also, participants were oversampled to increase the reliability of estimates about those who experienced Taglit and to allow analyses of subgroups of participants. A total of 2,387 cases were selected from the sampling frame. Some cases were subsequently determined to be ineligible, and the final number of cases included in the sample was 2,266.Footnote 6

Surveys were completed by 1,223 eligible respondents, and the response rate (AAPOR RR4) was 62 percent for Taglit participants and 42 percent for nonparticipants.Footnote 7 Overall, the response rate (weighted because participants were oversampled) was 55 percent.Footnote 8 Relatively few individuals explicitly refused to take part in the survey, although the rate for nonparticipants (8.4 percent) was almost double that of participants (4.5 percent). The cooperation rate was 94 percent for participants and 80 percent for nonparticipants. Overwhelmingly, nonrespondents were individuals who could not be located due to a lack of valid contact information. The absence of valid contact information was particularly a problem for nonparticipants.

The survey used telephone and web modes (with a small number of mail surveys). Because many email addresses—recorded in Taglit’s registration database five to eight years earlier—proved unusable, most interviews were conducted by telephone. Researchers utilized email messages, phone calling, data enhancement services, and extensive internet searching to obtain up-to-date contact information for potential respondents. Locating respondents was facilitated by social networking sites (e.g., Facebook, LinkedIn). Telephone calls were made by student interviewers. Full telephone interviews averaged 10 minutes, but some were as long as half an hour. Most interview questions were close-ended, with an open-ended question asked at the end of the survey. Field operations began in February 2009 and ended in July 2009.

The only significant differences between participants and nonparticipants in these rounds were with respect to age, with participants being slightly younger at the time of the trip. Gender, high school ritual practice, Jewish education, parental intermarriage, and denomination raised in did not differ significantly between participants and nonparticipants (Saxe et al. 2009a). In addition to design weights developed to account for the differential probabilities of selection as a result of the stratified sample, post-stratification weights were created using initial applicant data on age, Jewish denomination, round, and gender. These weights correct for differences between the distribution of known characteristics of the respondents and known characteristics of the sampling frame.

The central findings are based on logit or ordinal logit regression models. The models only control for variables where systematic differences existed between participants and nonparticipants (age at time of trip and at time of survey), where such variables were significantly associated with outcomes, and variables that had a significant interaction effect with participation. Variables that did not differ between participants and nonparticipants, and variables that did not interact with program effect, were excluded from analyses.

Survey of 2008 Applicants

In March 2008, all 37,983 eligible applicants to the summer 2008 round of Taglit were emailed a link to a web-based pre-trip survey. Nonrespondents received up to three email reminders. There were only 54 explicit refusals to complete the survey, implying a low likelihood of bias due to systematic refusals. The overall response rate to the pre-trip surveys was 59 percent (AAPOR RR2), 67 percent for participants and 41 percent for nonparticipants.

In October 2008, post-trip surveys were sent to participants in the summer 2008 round and to eligible applicants who had not participated and had not reapplied to the winter 2008–2009 round. A total of 37,168 individuals were sent invitations to complete the post-trip survey. Up to four email reminders were sent to nonrespondents. A sample of nonrespondents also received follow-up phone calls. The overall response rate to the post-trip survey was 30 percent (AAPOR RR2), 37 percent for participants and 20 percent for nonparticipants. Respondents were weighted to match the entire population on known characteristics (age, gender, country of residence, student status, employment status, denomination at time of application, and trip organizer).

Analyses of the impact of factors other than Taglit on attitudes to marriage and child-raising were conducted using ordinal logit regression. Participation in Taglit was included as a predictor to control for the association of other variables. Pre-trip measurements of the importance of marrying a Jew and raising Jewish children were not included as controls, as the variables of interest were causally prior to the pre-trip measurement and the inclusion of a pre-trip measure resulted in underestimates of effect size. By contrast, models of the impact of Taglit on marriage and child-raising included pre-trip measures of the importance of marrying a Jew and raising Jewish children as a control (i.e., an untreated control group design with dependent pre-test and post-test samples; see Shadish et al. 2002).

Results

The findings are reported in three sections: First, drawing on the survey of 2001–2004 applicants, the impact of Taglit after five to eight years is examined with respect to choice of spouse, conversion, relationship status, and the importance of raising Jewish children. Second, drawing on responses to the open-ended question posed in this survey, the findings are contextualized in relation to the subjective experiences of Taglit participants. Third, drawing on data from the 2008 surveys, Taglit’s influence on the importance of marrying a Jew and raising Jewish children is compared to the influence of other educational and background factors.

Impact of Taglit

Choice of Spouse

Participation in Taglit was associated with significantly greater probability of non-Orthodox participants being married to a Jew (Table 2). Virtually all of the married applicants in the study sample who were raised Orthodox married another Jew (see Lazerwitz et al. 1998 regarding the impact of denominational affiliation on intermarriage). Accordingly, analyses of the effect of Taglit on marital choice are restricted to respondents who were raised non-Orthodox. Overall, holding rates of parental intermarriage at their means, the predicted probability of inmarriage for married non-Orthodox participants was 72 percent compared to 47 percent for married non-Orthodox nonparticipants.Footnote 9 Thus, expressed in terms of odd ratios, the odds of a non-Orthodox participant being married to a Jew are expected to rise by nearly 200 percent (see Long 1997).Footnote 10

Table 2 Coefficients of logit regression of intermarriage on selected variables
Full size table

The effect of Taglit was largest for non-Orthodox participants with intermarried parents. The odds of a non-Orthodox participant with intermarried parents being married to a Jew are expected to increase by more than 700 percent.Footnote 11 (The predicted probability of marrying a Jew for non-Orthodox participants with intermarried parents is 56 percent compared to 13 percent for non-Orthodox nonparticipants with intermarried parents.) However, the estimates for nonparticipants are based on small cell sizes and participation is only marginally significant (p ≤ .1).

The effect of Taglit is still large for participants from inmarried families. Compared to a nonparticipant from an inmarried non-Orthodox family, the odds of a participant being married to a Jew are expected to increase by 128 percent (predicted probability of 76 percent for participants compared to 56 percent for nonparticipants).Footnote 12

Conversion

Among respondents who married partners who were not raised Jewish, 21 percent of the participant spouses were Jewish at the time of the survey compared to 4.6 percent of the spouses of nonparticipants (F 1, 351 = 10.25, p ≤ .01). It is not clear whether the partner had a Jewish parent but was not raised as a Jew or whether he or she had no familial tie to Jewish life and actually converted to Judaism. Nevertheless, this observation may indicate a substantially higher rate of conversion to Judaism among spouses of participants.

Relationship Status

The higher rate of inmarriage among participants—as well as the apparently higher rate of conversion among participant spouses—may be related to another finding: Younger participants were less likely to be married than their nonparticipant comparisons. Above age 30,Footnote 13 Taglit participants and nonparticipants looked very much alike with respect to their relationship status (F 5.9, 7215.8 = .61, p > .1), with 48 percent being married. Below age 30, however, there were significant differences in relationship status (F 5.9, 7218.3 = 3.15, p ≤ .01). Taglit participants were less likely to be married (25 percent for participants compared to 47 percent for nonparticipants) and more likely to be dating (35 percent for participants compared to 23 percent for nonparticipants). One possible explanation for this phenomenon is that Taglit participants are more likely to want to marry a Jewish person and consequently spend a longer time searching for a suitable partner. Participants were not, however, significantly more likely to date Jews (results not shown).

Raising Jewish Children

Of all Taglit applicants surveyed, 20 percent had at least one child, with 49 percent of married applicants having children compared to one percent of unmarried applicants. Due to the small number of cases (n = 265), it is not yet possible to analyze how Taglit may have influenced parental decisions regarding the religious education and socialization of children.

Although it is too early to analyze the decisions of alumni parents, the survey of the 2001–2004 applicants also asked respondents without children to indicate how important it was to them to raise children as Jews. Taglit participants were significantly more likely to view raising their children as Jews as very important (see Table 3).Footnote 14 The odds of a participant indicating that raising Jewish children is very important were 121 percent greater than for a nonparticipant. This effect is greater than the short-term impact of Taglit for these cohorts (Saxe et al. 2004, 2006b). The effect size was even greater for intermarried Taglit participants (p ≤ .1), for whom the odds of finding it very important to raise Jewish children were 190 percent greater than for intermarried nonparticipants.

Table 3 Coefficients of ordinal logit regression of importance of raising Jewish children on selected variables
Full size table

Subjective Meaning of Taglit

The findings about Taglit’s impact on marriage and child-raising on 2001–2004 trip participants are buttressed by comments respondents made at the end of this survey. Respondents were asked, “During the years since your trip, can you think of any decisions that you made that were influenced by your experience on Birthright Israel (for example, decisions about jobs, relationships, religious observance, how you spend your free time, etc.)?” Over 90 percent of those who were asked this question gave some form of open-response answer, providing over 800 different responses. Seventeen percent of participants who responded indicated that Taglit influenced their decisions regarding dating and marriage and their desire to raise children as Jews. In the minds of these respondents, marrying another Jew and raising children as Jews are intimately linked, as evidenced by comments such as the following:

Simply put, going on the Birthright Israel trip made me decide once and for all that I would marry a Jewish man, raise my children Jewish, and really hold on to my Jewish heritage. My fiancé and I have always talked about raising our future children Jewish, but after going on the trip, it really made me want it even more.

Notably, even those who said they had or would be willing to marry a non-Jew stressed the importance of raising a Jewish family. One participant respondent, for example, was “unwilling to date anyone that was unwilling to raise a family Jewish,” implying that the religion of the spouse was unimportant as long as he or she made a commitment to being part of a Jewish family. Another respondent said that “in terms of my relationship, it has been clearly communicated that if I am to marry this person, my kids will be Jewish [and that] is all that is important.” A third respondent stated: “We wanted to raise my kids Jewish and carry on the Jewish tradition. It was difficult marrying a non-Jew, but it made me want to keep the Jewish tradition within my home.” Finally, almost a dozen respondents also mentioned that their current spouse or fiancé/e was someone they met on the trip, suggesting a more direct impact of Taglit on the marriage choices of participants.

Comparative Perspective

How does the size of the Taglit effect compare to those of other forms of Jewish education? Due to the relatively limited educational and background data about 2001–2004 applicants, this question cannot be answered directly. Instead, using more detailed data from the survey of applicants during 2008, we focus on attitudes toward marrying a Jew and the importance of raising Jewish children.

Importance of Marrying a Jew

While a positive attitude regarding inmarriage may not lead to marriage with a Jew, there is considerable reason to believe the two are connected. Taglit’s impact on attitudes toward marrying Jews has been among its most consistent effects in short-term (three months post-trip) evaluations. Medium-term evaluations (conducted one to three years post-trip) demonstrated that the differences between participants and nonparticipants grew over time (Saxe et al. 2004, 2006b). These data on the importance of marrying a Jew from three months to three years post-trip and on spousal choice five to seven years post-trip suggest there is a carry-over from attitudes to behavior.

Separate models are used for the impact of other modes of Jewish education and the impact of Taglit, with a third model adding the interaction term found in the analysis of intermarriage for winter 2001–2004 applicants (see Table 4). Participation in Taglit is associated with a significant increase in the importance placed upon marrying a Jew, and the impact is larger on people raised in intermarried families than on those raised in inmarried families.

Table 4 Coefficients of ordinal logit regression of importance of marrying a Jew on selected variables
Full size table

In terms of other forms of Jewish education, Hebrew school (or other multiday supplementary school) and day school exert a positive effect on attitudes to marrying a Jew, with day school having about 2.5 times the effectiveness of Hebrew school on a per annum basis (see Table 5).Footnote 15 A season at a Jewish overnight summer camp had about as great an effect as a year at Hebrew school. Participation in high school youth groups—measured as never, rarely, occasionally, or often—was also associated with more positive attitudes regarding marrying a Jew. Sunday school, however, had a slight negative effect, with increased length of exposure being associated with slightly lower importance being placed on marriage to a Jew, similar to patterns described by Cohen (1995).

Table 5 Coefficients of ordinal logit regression of importance of raising a Jewish child on selected variables
Full size table

These data indicate that in order to equal the impact of Taglit on importance of marrying a Jew, one would need to attend a Jewish day school for 4.8 years, a Hebrew school for 12.0 years, or a Jewish overnight summer camp for 12.8 years (well beyond the number of years most Jewish summer camps offer sessions).Footnote 16 The impact of Taglit is approximately equal to moving from never attending a Jewish youth group during high school to attending one often. No amount of Sunday school would be equal to the impact of Taglit.

Noneducational factors were also associated with attitudes toward marrying a Jew. Having intermarried parents was associated with placing much lower levels of importance on having a Jewish spouse. Having more Jewish friends in high school had a particularly large effect, with each step on our friendship scale being associated with a 50 percent increase in the odds of deciding it was “very important” to marry a Jew. People raised in observant families placed higher levels of importance on marrying a Jew, as measured using a Mokken scale (Guttman 1950; Loevinger 1948) of family ritual practice during high school ranging from zero to four (based on celebrating Hanukkah, attending a Passover seder, regularly lighting Shabbat candles, and keeping kosher at home). Similarly, people raised by parents involved in multiple Jewish organizations were more likely to think marrying a Jew was important, as measured using a Mokken scale ranging from zero to four (based on belonging to a synagogue, Jewish community center, Jewish federation, or other Jewish organization). Being raised as a secular Jew was associated with lower importance placed on inmarriage than for Reform Jews, while people raised “Just Jewish” did not differ substantially from Reform Jews. Conservative Jews placed higher levels of importance on marrying a Jew than did Reform Jews. Individuals raised Orthodox placed the highest importance on inmarriage.

Although these factors are not as close in nature to Taglit-Birthright Israel as the educational variables, their effects can be compared. Participation in Taglit had slightly greater impact than being raised Conservative (compared to Reform), but not as large as being raised Orthodox (compared to Reform). On the scale of family level of ritual practice, the impact of Taglit is equal to an increase of observance of close to two additional rituals. Similarly, on a scale of the proportion of Jewish friends in high school, Taglit participation is approximately equal to an increase of one level (e.g., from about half to mostly Jewish). In terms of the number of organizations one’s parents belonged to, Taglit participation is approximately equal to involvement in three additional organizations. The impact of Taglit is not, however, as great as having two Jewish parents; it is approximately 40 percent as large of an impact.

Importance of Raising Jewish Children

Participation in Taglit is associated with increased importance placed upon raising Jewish children.

Jewish day school exerts a smaller impact on the importance of raising Jewish children than was the case for the importance placed on marrying a Jew. Essentially, it has the same impact per annum as Hebrew school. Sunday school is associated with a slight increase in the importance placed upon raising Jewish children. The effect of Jewish summer camp on a per year basis lies between that of Sunday school and Hebrew and day school. Jewish youth groups are associated with increases in the importance placed upon raising Jewish children.

The effect of Taglit corresponds to 11.2 years of Hebrew school, 10.5 years of day school, 26.0 years of Jewish summer camp, and 45.0 years of Sunday school, the latter two clearly lying outside the maximum possible length of enrollment. The impact of attendance at Jewish youth group in high school, measured on a scale of never, occasionally, sometimes, and often, is equivalent to going often as compared to never.

Among noneducational factors associated with importance of raising Jewish children, family ritual practice is associated with increased importance; so too is having a higher proportion of Jewish friends in high school and having parents involved in Jewish organizational life. People raised as secular Jews or “Just Jewish” placed less importance on raising Jewish children than did people raised Reform. Controlling for other factors, however, those raised Conservative or Orthodox were not substantially more likely than those raised Reform to feel raising Jewish children was important. As was the case with importance placed on marrying a Jew, people raised in intermarried families were less likely to feel raising children Jewish was important. Women placed more importance on raising Jewish children than did men. The impact of Taglit was equivalent to performing an additional two ritual practices, moving up nearly three levels of the proportion of Jewish friends in high school (none, a few, half, most, or all), and having parents involved in an additional 3.1 more types of Jewish organizations. The impact of Taglit was greater than being raised Orthodox or Conservative (compared to Reform) and was almost as great as having inmarried parents (compared to intermarried parents).

Discussion

The analyses of surveys of Taglit applicants and participants presented here provide strong support for the view that decisions about intermarriage and the religious identity of children are, in part, a function of Jewish educational experience. Jewish youth groups, summer camp, Hebrew school or other multiday supplementary education, and day school were all associated with increased importance being placed on marrying Jews and raising Jewish children. These findings comport with those of Phillips and Fishman (2006) and others (Cohen 1995, 2006; Fishman and Goldstein 1993; Medding et al. 1992; Phillips 1997). In light of early concerns (e.g., Liebler 1999) that Taglit would not represent a strong enough educational intervention to make a difference, the magnitude of the effect on intermarriage is surprising.Footnote 17 That the odds of non-Orthodox participants in Taglit being married to a Jew are much greater than for nonparticipants suggests that Taglit has a far greater effect than previously believed. The finding is buttressed by the narrative responses of many survey respondents.

Taglit’s impact extends beyond marriage to views on child-raising, with the odds of participants considering raising Jewish children as “very important” being more than 120 percent greater than for nonparticipants. Interestingly, Taglit’s influence on participants’ views regarding children extends beyond those who married Jews. Intermarried participants were nearly three times as likely as intermarried nonparticipants to think raising children as Jews was “very important.” Although Taglit’s impact on actual behaviors as opposed to attitudes with respect to raising children remains a question for future research, the present data suggest that both inmarried and intermarried alumni are highly motivated to raise children as Jews.

One surprising finding is that participants were somewhat less likely than nonparticipants to be married at the time of the survey. It is possible that participants are delaying marriage more than nonparticipants, in which case the full extent of Taglit’s impact on the marriage patterns of alumni remains unknown. Examining this hypothesis will have to await future research, but it is possible that the effect reflects Taglit participants’ stronger motivation to marry Jewishly.

This disparity between participants and nonparticipants in relation to marriage could be an indication of fundamental differences between the two groups. As discussed in the “Methods” section, however, there were few meaningful differences in the Jewish backgrounds of participants and nonparticipants. Rather, as noted, logistical factors were key in determining who became a participant versus a nonparticipant. Although there is no evidence of pre-existing differences in the populations, it is possible that more engaged individuals would have been more willing to respond to the survey. The effect of this potential bias, however, would have been more pronounced among nonparticipants (who would presumably be less interested in responding to a survey focused on Jewish issues) and, insofar as the most Jewishly engaged nonparticipants would be the most likely to respond, would actually strengthen our conclusions. In any case, weights were applied to make both groups statistically equivalent and assure that the samples were parallel to the populations.

Another potential question about the study is whether the population of Taglit applicants is representative of American Jewish young adults. As described earlier, it is likely that the average level of Jewish education among Taglit applicants is higher than for the Jewish population at large. Nevertheless, the sample represented all of the varieties of American Jewish life, from those who had no formal Jewish education at all to those with day school backgrounds. As with most surveys, the point estimates (that is the percentages who engage in a particular behavior) may not perfectly match the population (cf. Kadushin et al. 2005); nevertheless, the relationships—including the effect of Taglit on marriage decisions—are robust.

More generally, the findings have important implications for the debate over policy responses to intermarriage and suggest that the claims advanced by both the outreach and inreach sides of the debate are incomplete. The study casts doubt on the central claim of advocates of inreach; specifically, that there is little that can be done to convince intermarried couples to choose to raise their children as Jews. Taglit’s strong impact on the importance intermarried participants attach to raising their children as Jews should encourage advocates of inreach to reconsider the possibility of engaging those who choose a mixed marriage. In parallel, the study also casts doubt on a central claim of outreach advocates. The present findings provide strong evidence that the rate of intermarriage is not fixed and unchangeable. Rather, the likelihood of intermarriage is contingent upon Jewish education and background, including even—or especially—educational interventions that occur after children leave their parents’ homes. Perhaps most importantly, the study suggests that there is no need to decide between inreach and outreach. The educational interventions that reduce the likelihood of intermarriage—including but not limited to Taglit—also increase the likelihood that intermarried Jews will view raising Jewish children as very important.Footnote 18

The present research cannot shed light on a key issue that will continue to divide the inreach and outreach advocates; specifically, whether rabbis and Jewish educators should advocate for endogamy. In the case of Taglit, the program’s effects appear to be due to more general features of the program, such as its impact on participants’ overall Jewish identities and/or social networks. Direct promotion of inmarriage is not part of Taglit’s curriculum. Although some Taglit educators may encourage endogamy, the program’s effects seem too large to be the result of such ad hoc efforts. Accordingly, the question of whether rabbis and educators should advocate for endogamy remains unanswered.

Implications

Three conclusions emerge from this research. First, while Taglit is strongly associated with measurable decreases in the probability of intermarriage and increases in the importance placed on marrying a Jew and raising Jewish children—and the strong comparison group design rules out most alternative explanations—the program does not stand alone. Other forms of Jewish engagement by our respondents—supplementary school, day school, Jewish summer camps, and youth groups—had positive effects on attitudes toward marrying Jews and raising Jewish children as well. The impact of Taglit in conjunction with these programs is greater than its impact alone.

Second, in the case of intermarriage, Taglit had an outsized impact on participants who themselves had intermarried parents. Rather than being lost to Jewish life, this group appears to be particularly susceptible to informal Jewish education, as provided by Taglit. Finally, the potential demographic effects of this result are considerable. It is not simply that Taglit has a powerful effect on participants in theory. Rather, because of the scope of the program (by 2009, engaging nearly 200,000 North American young adults), it has the potential to alter the demographic trajectory of U.S. Jewry.

The strength of the findings about Taglit’s impact notwithstanding, the present study is a snapshot of a dynamic situation. The marital decisions of those currently unmarried may alter the present conclusions. Also, in terms of the impact of Taglit on Jewish life, the picture will not be complete until we can observe how those whom we studied raise children, maintain a connection to Israel, and become involved with Jewish institutions. It is also critical to assess Taglit’s evolution. If Taglit becomes a normative part of socialization for Diaspora young adult Jews, current alumni will be the parents of a new generation of children who will be socialized differently than their parents; presumably, the children’s Jewish identity and relationship to Israel will have new and different forms. Its impact aside, Taglit has provided an extraordinary socio-educational laboratory for exploring the forces that shape and govern some of the most critical decisions in a young adult’s life.