Denominational Variations Across Jewish Communities

Transcription

1 Denominational Variations Across Jewish Communities Harriet Hartman Department of Sociology, Rowan University Ira Sheskin Department of Geography Jewish Demography Project, Sue and Leonard Miller Center for Contemporary Judaic Studies University of Miami April, 2013 Paper prepared for presentation at The Association for the Study of Religion, Economics and Culture (ASREC) Annual Conference 1

2 Denominational Variations across American Jewish Communities Abstract This paper explores variations in Jewish denominational identification and some correlates of these variations in 55 American Jewish communities that have completed local Jewish demographic studies since It also explores the Jewish identity of denominational identification across 22 American Jewish communities that have completed local Jewish demographic studies from for which more detailed data are available in the Decade 2000 data set. Using the 55 communities, five clusters of communities are created via k- means cluster analysis using the community denominational profiles. The relationship of these clusters to demographic and Jewish behavior indicators is explored. Using the Decade 2000 data set with its nearly 20,000 interviews of Jews in 22 American Jewish communities, we examine the extent to which individual Jewish identification varies by the denominational composition of the community in which an individual resides, finding that considerable variation exists in Jewish identity measures depending on the type of denominational profile in the individual s community. That is, Orthodox Jews, for example, behave differently in a community with a significant Orthodox population than in a community with few Orthodox, but many Reform Jews. 2

3 Introduction In previous research, we have studied the considerable variation among American Jewish communities and analyzed the relationship between community context and expressions of an individual s Jewish identity (Hartman & Sheskin, 2011, 2012, 2013). In the current paper, we turn our attention to variations in the denominational composition (Orthodox, Conservative, Reconstructionist, Reform, and Just Jewish) of American Jewish communities, and consider their relationship to individual Jewish identity. In particular, we look at the variation in Jewish identity within denominations. In all cases, note that Jewish identification is a self-definition and is not necessarily based on synagogue membership, ideology, or religious practice (Klaff 2006). In fact, discrepancies between Jewish identification and practice are sometimes evident. For example, in Detroit (Sheskin 2007, 7-9), 20% of respondents who identify as Orthodox and 29% who identify as Conservative report that they do not keep kosher. As a second example, 14% of respondents who identify as Reform report that they never attend synagogue services (p. 6-82). Previous research has established considerable differences between the Orthodox and non-orthodox in terms of their behavior and lifestyle (e.g., age at first marriage, divorce rates, number of children, types of schools their children attend, food consumption, level of religious practice, and living near a synagogue) (Hartman 2007; Hartman & Hartman 2009; Miller et al. 2012, Ukeles et al. 2006, Wertheimer 2005). But such broad strokes likely hide important variations within each group, sometimes variations that may overlap with the comparison group. Because the Orthodox are more often raised Orthodox (81% of Orthodox were raised Orthodox, 3

4 compared to 65% of Conservative being raised Conservative, 57% of Reform being raised Reform, and 55% of Just Jewish being raised Just Jewish (Ament 2005, 12)) and live in Orthodox communities with few non-orthodox Jews or non-jews, their social and cultural norms are expected to be more homogeneous. In her fictitious Kaaterskill Falls, Allegra Goodman (1999) depicts the Orthodox veneer camouflaging a good deal of variation beneath its label. This finding is confirmed most profoundly by Miller, Ukeles, & Cohen (2012) in their study of New York Jews in which they distinguish between Hasidic Jews, Yeshivish Jews, and Modern Orthodox Jews. Variation is even more expected among the Reform, whose members are often expatriates from other Jewish denominations. According to NJPS , 25% of Reform Jews were raised Orthodox, 9%, Conservative, and 9% Just Jewish; for the Conservative, 21% were raised Orthodox, 7% Just Jewish, and 6% Reform; for the Just Jewish, 55% were raised Just Jewish, 18% Orthodox, 16% Reform, and 10% Conservative. Some of the variation within denominational groups could be studied by analyzing the National Jewish Population Survey. However, this analysis was limited by the relatively small number of Orthodox in the sample, as well as the indicators included in that survey. The small sample of Orthodox, as well as the small number of communities represented with adequate sample sizes, also limited the extent to which variation among communities could be examined. And yet it is of interest whether Orthodox are spread over a variety of communities, or only where a certain threshold number (Warf 2006, 287) are located, and how the variation in 4

5 numbers might affect the Orthodox lifestyle. Similar questions can be raised for the other denominational groups, although the expectations of their similarity are a priori lower. Still, we may ask whether some regularity exists in the patterns of variation according to attributes of the community and/or attributes of the (Jewish) population in the community. The current paper focuses on four research questions: 1) How are the denominational groups distributed across communities? 2) How does the denominational composition in communities vary across the United States? 3) How homogenous are each of the denominational groups with respect to their expressions of Jewish identity? 4) Are the individual s expressions of Jewish identity related to the denominational composition of their community of residence? The Two Data Sets Employed in this Study The Community Level Data Set The North American Jewish Data Bank ( archives about 200 local Jewish community studies. The first data set includes community wide measures in 55 of these American Jewish communities that have completed local Jewish community studies from These studies were based on more than 55,000 telephone interviews (using random digit dialing (RDD), distinctive Jewish name (DJN), and list sampling) at a combined cost of well over $10 million. About 80% of the American Jewish population lives in these 55 Jewish communities. Of the 5

6 25 largest American Jewish communities, only two (Rockland County, New York and Orange County, California) have not completed a study. These studies covered diverse topics designed to assist local Jewish Federations in their planning and coordinating role for the Jewish community. The vast majority of these 55,000 interviews took minutes each. Importantly, note that the observations in the Community Level data set are for the communities themselves, not for the respondents to the surveys. Table 1 shows the 55 communities used in this study and the community level data on Jewish identification for each community. These data and hundreds of other variables are available in Comparisons of Jewish Communities: A Compendium of Tables and Bar Charts by Sheskin (2012) at The Decade 2000 Data Set The second data set (Decade 2000) combines the data from 22 Jewish community studies conducted by Ira M. Sheskin as the principal investigator since the completion of NJPS This limitation had a number of significant advantages. First, the questionnaire used in each of these local Jewish community studies was basically the same, with minimal variation from community to community in almost all basic measures of Jewish identity. The survey research literature indicates that even small changes in question wording or in the sequence in which questions are asked in a survey can have a significant impact upon survey results (Bradburn, Sudman, & Wansink 2004), so this was an important plus. Second, and of major import, Sheskin had already compiled all 22 studies into a single meta-data file, having performed the preliminary comparisons of the 6

7 questionnaires and eliminating (for the most part) variation by standardizing response categories. It should be noted that this preparation is extremely timeconsuming and is mentioned as a major drawback for doing this kind of integrative data analysis (Curran et al. 2008; Cooper & Patall 2009). Third, Sheskin used the same basic methodology for determining the survey sample (usually a combination of RDD and DJN techniques) for each study. Note, however, the variations in percentage of the sample determined by RDD and DJN methods (Table 2). Fourth, the same procedure was used to select a respondent from the household to interview (any cooperative adult, Jewish or not, who answered the telephone in a Jewish household). In each study a respondent was pursued intensively until a high cooperation rate was achieved. Fifth, all 22 community studies used the same definition of a Jewish person: A Jewish person is defined as any person who currently considers himself/herself Jewish (or who is identified as such by the respondent) or who was born Jewish or raised Jewish and has not formally converted to another religion and does not regularly attend religious services of another religion (irrespective of formal conversion). A Jewish household is defined as any household containing a Jewish person. Nevertheless, variability exists among the studies. Some questions had to be eliminated from our analysis because they were not asked in all communities or had been altered significantly from study to study. In terms of the samples, while most of the samples were derived from a combination of RDD and DJN techniques, they varied in the percentage of the sample that each sampling technique provided, and in Jacksonville, in addition to RDD and DJN samples, a sample was drawn from the 7

8 Jewish Federation mailing list. Further, the studies varied in their cooperation rates, ranging from 49%-97% for the screener (which determined eligibility for the survey) and from 64%-99% for the survey itself. Further, the studies were conducted over ten years, which may affect the results. Note that the Decade 2000 Data Set includes 19, minute interviews, and is a random sample of 547,000 Jewish households in the 22 communities. Indicators The typical question asked in these studies to determine denominational identification is: Do you consider yourself Orthodox, Conservative, Reform, Reconstructionist, or Just Jewish? For the purposes of this paper, we have combined Reconstructionist and Reform. In addition, respondents were permitted to provide their own responses. Traditional was combined with Conservative; Jewish Renewal, Jewish Humanist, agnostic, and atheist were combined with Just Jewish. For the most part, other than Reconstructionist and Traditional, the percentage volunteering other responses was minimal (see the footnotes to Table 1). Although ethnic or religious identity is usually considered as part of an individual s self-concept in terms of group belonging and the value and emotional significance attached to it (Rebhun 2004), local Jewish community studies do not usually query an individuals self-concept, but rather ask about various behavioral expressions of that identity. Therefore, our indicators of Jewish identity primarily represent behavioral practices or inclinations and expressions of Jewish identity in various ways, rather than Jewish identity itself. Since many studies of Jewish 8

9 identity conflate the self-concept and religious/ethnic practices or their value, using such indicators to indicate Jewish identity is not unusual. To determine how Jewish identity was expressed by the 17 relevant variables common to all 22 communities, we performed a principal components factor analysis with Varimax rotation, resulting in four factors (Table 3). Note that factor loadings (shown in parentheses below) of 0.45 or higher were considered when interpreting each factor. The four factors combined account for 57% of the variance in the original 17 variables. 1 Factor 1 The Communal Ritual Factor reflects the religious identity of American Jews and includes the more common religious practices observed by many American Jews, often in communal or extended family groups: lighting Chanukah candles (.828), participating in a Passover Seder (.782), Mezuzah on the front door (.615) (a public display), attending synagogue services (.523), and synagogue membership (.496). For the most part, the variables loading highly on Factor 1 are those which Alexander (1987, 124) referred to as ceremony, and affirming membership in the social and cosmological order. Factor 2 The Broad Ethnic Factor reflects the ethnic identity of American Jews vis-à-vis the broader Jewish community: visits to Israel (.683), Jewish organization membership (.657), donation to the local Jewish Federation in the past year (.587), 1 Note that the names of some of the factors have been modified, for simplicity and clarity, from earlier publications (Hartman & Sheskin 2011, 2012, 2013) using these same indicators. 9

10 emotional attachment to Israel (.583), and donations to Jewish charities other than the local Jewish Federation in the past year (.572). Factor 3 The Local Ethnic Factor reflects the ethnic identity of American Jews in terms of integration into the local community through non-religious or nonsynagogue Jewish institutions: familiarity with the local Jewish Federation (.780), familiarity with the local Jewish Family Service (.772), being at least somewhat familiar with at least one local Jewish agency (.754), and participating in or attending any program at, or sponsored by, the local JCC in the past year (.489). Factor 4 The Private Ritual Factor reflects the religious practices related to personal rituals that reveal stricter, daily, and personal commitment to ritual, many performed in the home: keep kosher outside the home (.879), keep kosher inside the home (.865), and lighting Friday night candles (.584). Findings Denominational Identification of American Jews (55 communities) According to the latest National Jewish Population Survey (NJPS ) (Kotler- Berkowtiz et al. 2003), the denominational self-identification of American Jews is: 8% of households are Orthodox, 25% Conservative, 2% Reconstructionist, 35% Reform, and 30% Just Jewish. Note again that this is self-identification, and may or may not reflect actual synagogue membership (Klaff 2006). Deconstructing this national average, however, reveals significant geographic variation across communities in the United States. Comparing the 55 communities in which local Jewish community studies have been conducted from reveals a range of 1-21% Orthodox households across communities; 14-39% 10

11 Conservative; 22-60% Reform, and 14-48% Just Jewish. Note that these percentages reflect households rather than individuals. 2 The communities with the highest percentage Orthodox are Baltimore (21%) and New York (20%), with significantly more Orthodox than any others (the next highest percentage Orthodox is 12% in Bergen, New Jersey. The majority of communities have only 5% or fewer Orthodox, with four communities having only 1% (Howard County, Atlantic County, NJ; Martin-St. Luce; York, PA). The median percentage is 4%. Note that even within the same state (e.g., Bergen and Atlantic County, New Jersey) communities may differ significantly in percentage Orthodox. The percentage Conservative varies from 14% in Portland, ME to 39% in Tidewater, VA. The median percentage Conservative is 28%. The percentage Reform varies from 22% in Harrisburg, PA, to 60% in St. Louis, MO. The median percentage is 37%. Communities differ significantly in terms of the percentage who do not identify with any of the main denominational groups (the Just Jewish group), from 14% in Palm Springs, CA to 48% in Portland, ME. The median percentage Just Jewish is 18%. Changes in Denominational Identification Over Time For 20 communities in the Compendium (Sheskin 2012), we have data on denominational self-identification for two points in time, with the range between studies varying from 6 years (in West Palm Beach) to 20 years (in Washington DC). 3 2 Note that because of higher average household sizes, Orthodox Jews are a higher percentage of all Jews than they are of all Jewish households. 11

12 The percentage Orthodox seems to be fairly stable over time in most of the 20 communities. A notable increase is seen in Detroit (from 7-11% over a 16 year period) and Atlanta (from 7-10% over a 10 year period), balanced by notable decreases in Atlantic County (from 6-1% over a 19 year period) and in Milwaukee (from 7-3% over a 13 year period). This stability is reflected in the relatively stable percentage Orthodox measured in the three National Jewish Population Surveys (1971, 1990, and ) (Massarik & Chenkin 1973; Kosmin 1991; Kotler- Berkowitz 2003): 9% of American Jewish households were Orthodox in 1971, 6% in 1990, and 8% in In contrast, the percentage Conservative has decreased notably in 10 of the 20 communities compared over time, and has not increased notably in any of the communities. In 6 communities, the percentage Reform has increased notably, often with very large changes (in Atlanta, from 34% to 45%; while in 5 other communities large decreases are noted (e.g., Milwaukee, from 52% to 39%). 3 Note that in a few communities, such as Milwaukee, in the earlier study the question asked was: Do you consider yourself Orthodox, Conservative, Reconstructionist, Reform, or something else? while in most of the studies the question was: Do you consider yourself Orthodox, Conservative, Reconstructionist, Reform, or Just Jewish? The former question yields a higher percentage of Reform and a lower percentage of Just Jewish. 12

13 The percentage Just Jewish has increased notably over time in 9 communities (e.g., from 17% to 29% in Atlantic County), while decreasing notably in just 2 communities (from 33% to 18% in Atlanta and from 43% to 35% in Seattle). Thus, considerable flux may be noted in the denominational composition of communities, with the exception of percentage Orthodox. Cluster Analysis of the Denominational Composition of Communities Communities vary not only in terms of the percentage for each of the denominations, but also in terms of the combined profile of their denominational composition. We used a k-means cluster analysis to group the communities shown in Table 1 on the basis of their values for all four denominational groups. Four, five and six cluster solutions were examined and a five-cluster solution was adopted. Table 4 and Figure 1 present the average percentage of each denominational group for each cluster, and list the communities in each cluster. The assignment of the 55 communities to the five clusters explains about 97% of the sum of squares in the analysis of variance produced by the cluster analysis. We labeled each cluster with a name that distinguishes its denominational profile from the other clusters: The clusters were particularly defined by the percentage Orthodox, Reform and/or Just Jewish in the community. Cluster 1 (High Orthodox) was characterized as having the highest percentage Orthodox of all of the communities with approximately equal percentages of Reform, Conservative, and Just Jewish. Cluster 2 (Equal, Except Orthodox) was characterized by having about equal Conservative, Reform, and Just Jewish percentages. Cluster 3 (High 13

14 Reform) had the highest Reform percentages. Cluster 4 (High Reform/Low Just Jewish) had high Reform and low Just Jewish percentages. Cluster 5 (High Just Jewish) had the highest Just Jewish percentages. How the Clusters Differ In this section, we used the information on the 55 communities in the Community Level Data Set, which reported on the level of practice or behavior in the community (as opposed to looking at the individual data in the Decade 2000 data set). Demographic and Geographic Differences Among Clusters (Table 5) We considered the following nine demographic/geographic characteristics of communities as possible sources of variation among the five clusters: (1) the size of the Jewish population (expecting that a larger Jewish population might be able to support the infrastructure conducive to an Orthodox life style, for example); (2) the density of the Jewish population (as opposed to dispersion among the non-jews in the area); (3) the year of the study (to see if changes in the denominational structure from are extant); (4) the percentage Jewish households in the community; (5) the level of secular education (percentage of adults age 25 and over with a four year college degree or higher); (6) the percentage of persons age 65 and over in the Jewish population (expecting higher percentages of Just Jewish among more educated populations, perhaps); (7) the percentage of persons age 0-17 in the Jewish population; (8) the median household income (expecting communities with greater rates of affiliation to be wealthier); and (9) the region of the country (Census 14

15 Division Northeast, Midwest, South, West). Analyses of variance were used to examine variables 1-8 and both chi-square and z-tests were used to examine variable 9. The examination of the relationship between these nine variables and the Jewish identity clusters was exploratory in the sense that the literature, while offering abundant examples of the relationship between Jewish identity and the demographics of individuals, offers little guidance on the community level. Table 5 shows that the only significant difference between the clusters on these nine demographic/geographic variables was the mean size of the Jewish population, with the largest Jewish populations supporting communities with the highest percentage of Orthodox (as expected). And even in this case, if Cluster 1 (High Orthodox) is removed from the ANOVA, no significant difference is seen among Clusters 2-5. Jewish Behavior Differences Among Clusters (Table 6) Given the known differences at the individual level among the four denominations, we expected to find significant differences in Jewish behavior indicators among the clusters. We considered the following nine Jewish behavior characteristics of communities as possible sources of the variation between the clusters: (1) keep a kosher home; (2) always/usually light Hanukkah candles; (3) the couples intermarriage rate; (4) attend synagogue services once a month or more; (5) always/usually light candles Friday night; (6) participated in local Jewish Community Center (JCC) activities or programs in the past year; (7) donated to a Jewish charity in the past year; (8) someone in the household belongs to a Jewish 15

16 organization other than a synagogue or JCC; and (9) being very familiar with the Jewish federation. Unlike the demographic and geographic characteristics, four of the nine Jewish behavior relationships (Variables 1-4) are significant at the.05 level and another three (Variables 5-7) are significant at the.10 level. There was significant variation in the religious identity variables (kosher, Chanukah, intermarriage, attendance at services, and Friday night candles) exhibited among the clusters, and less so in the ethnic identity variables (JCC participation, Jewish charity, Jewish organizational member, and Federation familiarity). Not surprisingly, communities in Cluster 1 (High Orthodox) had higher percentages who keep kosher at home, attend synagogue services, and light Friday night candles and a much lower couples intermarriage rate. As will be seen below, however, this is not only because of the Orthodox in the community, but is an effect that spreads to the Conservative as well. Communities in Clusters 3 and 4 (High Reform and High Reform/Low Just Jewish) have high levels of Chanukah observance and are relatively high on the ethnic identity variables. Interestingly, synagogue attendance in the High Reform communities does not vary significantly from the High Orthodox communities. Communities in Cluster 5 (High Just Jewish) have the lowest percent who keep kosher, light Friday night candles, and light Chanukah candles, and the highest couples intermarriage rate. Attendance at services, while the lowest in Cluster 5, is not significantly different from Cluster 2 (Equal, Except Orthodox). The communities 16

17 in the High Just Jewish Cluster (Cluster 5) are also the lowest on three of the four ethnic identity factors. Variation in Individual Jewish Identity within Denominations Across Clusters In previous work, we have developed indicators reflecting four different types of expression of an individual s Jewish identity, using variables common to the 22 communities in the Decade 2000 data set (Table 3). The denominations vary in terms of their median factor scores for each of the four factors, most particularly in their scores on the Personal Ritual Factor, where indeed the Orthodox have much higher median scores than the other groups (Figures 2-5). However, the median scores on the other factors are not that far apart, and no one denomination has the highest median scores on every factor. Note that the factor scores are standardized, so that the median factor score is 0; stronger expressions of each factor of Jewish identity are greater than 0, while weaker expressions are less than 0. Notable in the current context in particular, however, is the variation within denominations in terms of their factor scores on each indicator. In Figures 2-5, we show box plots of the factor scores for each denomination on each factor. Box plots are a convenient tool for comparing distributions. The thick line for each denomination is the median value and the box extends from the 25 th to the 75 th percentile. The whiskers at the ends of the lines extend to the highest and lowest observations within 1.5 box lengths. High outliers are shown with circles and low outliers are shown with asterisks (Tukey 1977). While we see in Figure 2 that the Orthodox have much higher scores on the Private Ritual Factor than any of the other 17

18 denominational groups, some (albeit a minority) of the Orthodox scores on this factor are as low as in the other denominational groups; and some of the Conservative scores are well within the range of the more common Orthodox scores (Figure 5). The distribution of values for the Reform/Reconstructionist and the Just Jewish groups are quite similar. Examining the factor scores on the Communal Ritual Factor (Figure 2) considerable overlap exists in the range of factor scores from one denominational group to another, and even in terms of the most common scores for each denominational group between the 25 th and 75 th percentile. Similar overlaps can be seen for the Broad Ethnic and Local Ethnic Factors (Figures 3 and 4). The overall conclusion is that considerable overlap exists among the denominations. As we show below, some of the variation within denominations is related to the denominational composition of the communities in which they live. Individual Expressions of Jewish Identity within the Jewish Denominational Clusters In this paper, we analyzed the variation in factor scores for each of four factors (Communal Ritual, Broad Ethnic, Local Ethnic, and Private Ritual), within four denominations (Orthodox, Conservative, Reform/Reconstructionist, Just Jewish), across five community clusters (High Orthodox, Equal Except Orthodox, High Reform, High Reform/Low Just Jewish, and High Just Jewish). For the following analysis, we could use only the 22 communities in the Decade 2000 data set. Only four of the clusters identified from the 55 communities were represented in the 22-18

19 community data set: High Orthodox (Cluster 1), High Reform (Cluster 3), High Just Jewish (Cluster 5), and Equal, except Orthodox (Cluster 2). None of the 22 community surveys were conducted in Cluster 4 communities (High Reform/Low Just Jewish). Note that the sample sizes (N s) in Table 7, for example refer to the number of Orthodox individuals in the Decade 2000 data set who live in communities in clusters of a given type. For example, 129 Orthodox individuals in the sample live in communities included in the High Orthodox Cluster (Cluster 3); while the majority (584 of 885 cases) of Orthodox live in communities in Cluster 4 (Equal, Except Orthodox). Table 7 examines respondents who self-identify as Orthodox. Significant variation is seen in mean factor scores on three of the four Jewish identity factors (the exception being the Communal Ritual Factor). Although the Orthodox have positive mean factor scores on the Private Ritual Factor for respondents in all four clusters, scores are much higher among Orthodox who live in communities in the High Orthodox Cluster (3.0982); and much lower among Orthodox who live communities in the High Just Jewish Cluster ( ). With regard to the ethnic factors, the highest Orthodox mean score on the Broad Ethnic Factor is also for respondents in the High Orthodox Cluster (.6728), while the lowest mean score is for respondents in the High Reform Cluster (.1930). As to the Local Ethnic Factor, the Orthodox have their highest mean scores for respondents in the High Reform Cluster (.5402) and their lowest mean score for respondents in communities in the High Just Jewish Cluster (-.0013). 19

20 Table 8 examines respondents who self-identify as Conservative. The mean factor scores vary significantly across the clusters for all four Jewish identity factors. The highest mean score for the Communal Ritual Factor is found for respondents in the High Orthodox Cluster (.5286) and in the Equal, Except Orthodox Cluster (.5232). The lowest mean score is found for respondents in the High Reform Cluster (.3534). The highest scores on the Private Ritual factor are for respondents in the High Orthodox Cluster (.6060) and the lowest scores are for respondents in the High Just Jewish Cluster (-.0069), similar to the pattern for Orthodox (although the Orthodox still have higher mean scores on the Private Ritual Factor (3.0982) compared to Conservatives for respondents in communities in the High Orthodox Cluster (.6060)). In terms of the Broad Ethnic Factor, Conservatives have significantly higher scores for respondents in the High Reform Cluster (.4369). For the Local Ethnic Factor, scores are highest for respondents in communities in the High Reform Cluster (.8925). The lowest scores, however, are for respondents in communities in the High Orthodox (.2117) and High Just Jewish (.2157) Clusters (the latter, similar to the Orthodox). Table 9 examines respondents who self-identify as Reform/Reconstructionist. The mean factor scores vary significantly across the clusters for all four Jewish identity factors. Communal Ritual Factor scores are much higher for respondents in communities in the High Orthodox Cluster (.3859); perhaps because some Reform Jews have parents or other family members who are Orthodox, and they share the holidays and at least occasional synagogue attendance with them. Communal Ritual 20

21 Factor scores are lowest for respondents in communities in the High Just Jewish Cluster (.1618). Private Ritual Factor scores are lower for the Reform/Reconstructionist (-.331) than for the Orthodox (2.339) and Conservative (.1826), but, interestingly enough, are lowest for respondents in communities in the High Orthodox Cluster (-.4082). Perhaps this is, for them, a way of distinguishing themselves from the more visible presence of Orthodox in communities of this Cluster. Scores on the Broad Ethnic Factor do not vary significantly. For the Local Ethnic Factor, scores are highest for respondents in communities in the High Reform Cluster (.9012) (as they were for the Orthodox (.5402) and Conservative (.8925)), and lowest for respondents in communities in the High Just Jewish Cluster (.0631). Table 10 examines respondents who self-identify as Just Jewish. The mean factor scores vary significantly across the clusters for all four Jewish identity factors, not surprisingly given the somewhat eclectic nature of the Just Jewish group (some are young and have not committed to a particular denomination; some cannot afford to belong to a congregation so do not identify with any particular denomination; others have left denominational groups, but may not have left all their practices or observances behind). Scores on the Communal Ritual Factor are considerably lower than for the Orthodox, Conservative and Reform/Reconstructionist groups; however, they are lowest for respondents in communities in the High Just Jewish Cluster (-.8083) (similar to the Conservative and Reform/Reconstructionist groups) and highest for respondents in communities in the High Orthodox Cluster (-.3229) (similar to the Conservative and Reform). 21

22 Variation on the Private Ritual Factor is much smaller. Note that their scores are not as low as the scores of the Reform/Reconstructionist for any of the clusters, again reflecting the eclectic nature of this group. For the Broad Ethnic Factor, scores are highest for respondents in communities in the High Reform Cluster (-.1690) and lowest for respondents in communities in the Equal, Except Orthodox Cluster (-.3163). For the Local Ethnic Factor, scores are highest for respondents in communities in the High Reform Cluster (.4621) and lowest for respondents in communities in the High Just Jewish Cluster (-.2524) (as it is for the Orthodox and the Reform/Reconstructionist). In sum, the most important finding in Tables 7-10 is that the behavior of individuals who identify with each of the denominational groups differs by the type of community cluster in which they reside. For example, for the Communal Ritual Factor, Orthodox Jews who reside in communities in the High Just Jewish Cluster (.4752) behave differently than Orthodox Jews who reside in communities in the High Reform Cluster (.3354). Overall, community type would seem to be impacting behavior within denominations. Several other generalities are of interest in Tables The Orthodox and the Conservative have their highest scores on the Private Ritual Factor in the High Orthodox Cluster ( and.6060, respectively), but the Reform/ Reconstructionist have their lowest scores (-.4082) in this Cluster. The Just Jewish have their highest mean score on the Broad Ethnic Factor in the High Reform Cluster (-.1690), but the Orthodox have their weakest mean score on that factor in 22

23 this Cluster (.1930). All four denominational groups have their highest mean score on the Local Ethnic Factor in the High Reform Cluster. Summary and Conclusions Using two data sets, one with information on the community level on 55 Jewish communities, and one with information on 19,800 individual Jews living in 22 communities, we have explored the subject of denominational variation and temporal change across US Jewish communities. We show that not only do the percentages for each denomination vary across communities, but the denominational profiles of communities fall into five clusters. Each of the clusters is characterized by particularly high or particularly low percentages of some of the denominational groups (Orthodox, Conservative, Reform/Reconstructionist and Just Jewish). We identified both demographic characteristics and expressions of Jewish behavior that vary significantly across these five clusters. Thus, the cluster characterized by the highest percentage of Orthodox has communities with significantly larger Jewish populations than any of the other clusters, while the cluster characterized by the highest percentage of Just Jewish has communities with the smallest Jewish population size among the other clusters. In terms of religious behavior, the High Orthodox Cluster is characterized by the highest percentages of ritual performance (keep kosher, light Chanukah and Friday night candles, attend synagogue services) and a significantly lower rate of intermarriage; while the Just Jewish Cluster exhibits the opposite tendencies. 23

24 In previous research on the variation among Jewish communities, we found few characteristics of the Jewish community or of the broader religious context--that could clearly be related to an individual s Jewish identity. In this paper, in contrast, by examining the Jewish identity of individuals living in these clusters, using the 22 community data set, Decade 2000, we were able to show that expressions of Jewish identity both religious and ethnic--of individuals self-identifying as Orthodox, Conservative, Reform/Reconstructionist, and Just Jewish vary according to the cluster to which their community belongs. For example, how strictly Orthodox Jews observe the rituals that contribute to our Personal Ritual Factor is significantly related to the type of community cluster in which they live. In fact, a high percentage of Orthodox in the community raises the level of personal ritual practice for Orthodox and Conservative; but it actually lowers the level of ritual adherence for Reform (who perhaps are defining themselves as different from the Orthodox in those communities, or who find a greater need to carry on traditional Judaism in other clusters that have lower percentages of Orthodox). In contrast, a high percentage of Just Jewish in the community lowers the level of private ritual practice, as well as ethnic identity, for individuals in almost all of the denominational groups. How strongly individuals are identified with their local Jewish community also varies by the type of community cluster in which they live, for each of the denominations. A high percentage of Reform in the community raises the level of local ethnic identity among all of the groups affiliated with a denomination, suggesting that perhaps the activities in which the Reform invest are impacting all denominations in the local community. 24

25 Further research should aim at deconstructing the manner in which the denominational profile affects individuals expressions of Jewish identity. Is it the infrastructure and types of activities in which the various denominations invest? Is it simply the presence of more individuals with a similar orientation to a Jewish lifestyle? By including individuals from more communities in the mega-data set of 22 communities, we may be able to learn more about this variation and the manner in which it impacts individuals. Of particular interest would be to include individuals living in all 5 clusters (only 4 were represented in the current Decade 2000 data set). Finally, we would be remiss if we did not advocate for greater uniformity in the questions asked in local Jewish community studies, so that future research can examine in even greater depth the relationship between individual behavior and orientations to Jewish identity on the one hand and the community context in which Jews live, on the other. 25

26 References Alexander, Bobby Ceremony. In The encyclopedia of religion 3, ed. M. Eliade, New York: Macmillan. Ament, Jonathan American religious denominations, United Jewish Communities Series on the National Jewish Population Survey 2000, Report Number 10, online at (accessed 4 March 2013). Cooper, Harris and Erika A. Patall The relative benefits of meta-analysis using individual participant data or aggregated data. Psychological Methods 14: Curran, Patrick, Andrea M. Hussong, Li Cai, Wenjing Huang, Laurie Chassin, Kenneth J. Sher, and Robert A. Zucker Pooling data from multiple longitudinal studies: The role of item response theory in integrative data analysis. Developmental Psychology 44(2): Goodman, Allegra Kaaterskill Falls. New York: Bantam Dell. Hartman, Harriet The Intersection of Gender and Religion in the Demography of Today s American Jewish Families. Seminar on Creating and Maintaining Jewish Families, Brandeis University. Hartman, Moshe and Harriet Hartman Jewish Identity, Denomination, and Denominational Mobility. Social Identities 22:2 (Winter). Hartman, Harriet and Moshe Hartman Gender and American Jews: Patterns in work, education, and family in contemporary life. Waltham, MA: University Press of New England/Brandeis University Press. Hartman, Harriet and Ira Sheskin The influence of community context and individual characteristics on Jewish identity: A 21-community study. CT: North American Jewish Data Bank. Hartman, Harriet and Ira Sheskin The Relationship of Jewish Community Contexts and Jewish Identity: A 22-Community Study. Contemporary Jewry 32(3): Hartman, Harriet and Ira Sheskin, Forthcoming, The (Dis)similarity of a Minority Religion to its Broader Religious Context: The Case of American Jews, Review of Religious Research. Klaff, Vivian Defining American Jewry from religious and ethnic perspectives: The transitions to greater heterogeneity. Sociology of Religion 67(4):

27 Kosmin, Barry A., et al Highlights of the CJF 1990 National Jewish population survey. New York: Council of Jewish Federations. Kotler-Berkowitz, Laurence, Steven M. Cohen, Jonathon Ament, Vivian Klaff, Frank Mott, and Danyelle Peckerman Strength, challenge and diversity in the American Jewish population. New York: United Jewish Communities. Massarik, Fred and Alvin Chenkin United States National Jewish Population Study, In American Jewish Year Book, 73, eds. David Singer and Larry Grossman, New York: American Jewish Committee. Miller, Ron. Jacob B Ukeles, and Steven M. Cohen Jewish Community Study of New York: Comprehensive Report. UJA-Federation of New York. (accessed 6 April 2013). Sheskin, Ira The Jewish community study of Detroit. Bloomfield Hills, MI: The Jewish Federation of Metropolitan Detroit. Sheskin, Ira M Comparisons of Jewish communities: A compendium of tables and bar charts. Storrs, CT: Mandell Berman Institute, North American Jewish Data Bank and the Jewish Federations of North America. (accessed 6 April 2013). Tukey, John W Exploratory data analysis. Reading, MA: Addison-Wesley. Ukeles, Jacob B., Ron Miller, and Pearl Beck Young Jewish adults in the United States today. New York: American Jewish Committee. Warf, Barney Encyclopedia of human geography. Thousand Oaks, CA: Sage Publications. Wertheimer, Jack All quiet on the religious front? Jewish unity, denominationalism, and postdenominationalism in the United States. New York: American Jewish Committee. (accessed 2 July 2012). 27

28 Table 1 Jewish Identification Community Comparisons Base: Jewish Respondents Community Year Orthodox Conservative Reconstructionist Reform Just Jewish Portland (ME) % Las Vegas % Tucson % San Francisco % New York % St. Paul % Sarasota % Jacksonville % St. Petersburg % New Haven % Minneapolis % Rhode Island % Seattle % 19 NA Westport % Orlando % Washington % Hartford % Broward % Milwaukee % Wilmington % San Diego % Charlotte % Harrisburg %

29 Table 1 Jewish Identification Community Comparisons Base: Jewish Respondents Community Year Orthodox Conservative Reconstructionist Reform Just Jewish Denver % Lehigh Valley % Miami % San Antonio % Bergen % Richmond % Middlesex % W Palm Beach % Atlantic County % Columbus % Rochester % Phoenix % Tidewater % Monmouth % 37 NA S Palm Beach % Martin-St. Lucie % Howard County % York % Buffalo % Cincinnati % Detroit % Los Angeles % Baltimore %

30 Community Year Orthodox Table 1 Jewish Identification Community Comparisons Base: Jewish Respondents Conservative Reconstructionist Reform Just Jewish Essex-Morris % 27 NA Philadelphia % Atlanta % Boston % Pittsburgh % Cleveland % St. Louis % Chicago % Palm Springs % NJPS % Includes 1% of respondents who identify as Jewish Renewal. 2 6% of respondents identify as Traditional. 3 Includes 3% of respondents who identify as Jewish Renewal. 4 Includes 3% of respondents who identify as Jewish Humanistic and 1% as Jewish Renewal. 5 5% of respondents identify as Traditional. 6 3% identify with another denomination. 7 8% of respondents identify as Traditional. 8 7% of respondents identify as Traditional. 9 NJPS 2000 data are for the more Jewishly-connected sample. Note: Respondents who identify as Sephardic are included in Orthodox. 30

31 Community (Region) Atlantic County (MA) Table 2 Attributes of Community Studies in the Decade 2000 Data Set Percent of Sample Year of Field Work Sampl e Size Number of Jewish Household s Samplin g Fraction 1 RDD DJN Cooperation Rate for RDD Sample Screene r Surve y , % 32% 68% 90% 96% Bergen (MA) ,003 28, % 100% 0% 90% 84% Detroit (MW) ,274 30, % 32% 68% 64% 67% Hartford (NE) , % 28% 72% 95% 95% Jacksonville (S) , % 35% 38% 94% 98% Las Vegas (P) ,197 42, % 33% 67% 49% 64% Lehigh Valley (MA) , % 40% 60% 89% 96% Miami (S) ,808 54, % 100% 0% 90% 86% Middlesex (MA) ,076 24, % 44% 56% 88% 90% Minneapolis (MW) , % 28% 72% 89% 94% New Haven (CT) , % 36% 64% 83% 94% Portland (NE) , % 36% 64% 85% 95% Rhode Island (NE) , % 37% 63% 93% 91% San Antonio (SC) , % 43% 57% 87% 92% St. Paul (MW) , % 41% 59% 89% 94% Sarasota (S) , % 31% 69% 95% 93% South Palm Beach (S) ,511 73, % 100% 0% 87% 92% Tidewater (S) , % 29% 71% 97% 99% Tucson (MW) , % 37% 63% 95% 93% Washington DC (MA) West Palm Beach (S) , , % 33% 67% 80% 91% ,534 69, % 100% 0% 87% 92% Westport (NE) , % 32% 68% 94% 80% 31

32 Total 18, , % 55% 44% 85% 88% 1 The percentage of households in the community completing the survey. 2 27% of surveys were completed with households on the Federation mailing list. (MA) = Middle Atlantic, (MW) = Midwest, (NE) = New England, (S) = South, (P) = Pacific, (SC) = Southern Crossroads, (MW) = Mountain West Table 3 Jewish Identity Factors: Loadings from Principal Components Analysis with Varimax Rotation (Decade 2000 Data Set) Jewish Identity Factors Mean (Standar d Deviation Jewish Identity Indicator ) Light Chanukah candles * 1.67 (1.07) Participate in a Passover Seder * 1.63 (1.00) Religious Identity Communal Ritual Factor (Ceremony ) Ethnic Identity Broad Ethnic Factor Mezuzah on front door of home 1.26(.44).615 Attend synagogue services ** (1.79) Synagogue member.38 (.48).496 Visit to Israel.53 (.50).683 Jewish Organization member.31 (.46).657 Donated to local Jewish federation in the past year 1.48 (.50).587 Emotional attachment to Israel*** Donated to a Jewish charity other than Jewish federation in the past year Familiar with the local Jewish Federation **** Familiar with Jewish Family Service **** At least somewhat familiar with at least one local Jewish agency (.95).46 (.50).572 Local Ethnic Factor (.77) (.76).80 (.40).754 Religious Identity Private Ritual Factor (Ritual) 32

33 Participated in or attended a program at, or sponsored by, the local Jewish Community Center in the past year.31 (.46).489 Keep kosher outside the home.07 (.26).879 Keep kosher in the home.14 (.35).865 Light Friday night candles* 3.06 (1.12).584 % of variance explained 15.2% 14.4% 14.1% 13.3% *1=always, 2=usually, 3=sometimes, 4=never. **1=several times per week, 2=weekly, 3=a few times per month, 4=about once per month, 5=a few times per year, 6=High Holidays only, 7=never except special occasions, 8=never. ***1=extremely, 2=very, 3=somewhat, 4=not at all emotionally attached. ****1=very familiar, 2=somewhat familiar, 3=not at all familiar. Notes: 1) All variables, except as noted with asterisks, are yes/no responses (1=Yes; 0=No). 2) Attended synagogue services, Emotional attachment to Israel, Familiar with local Jewish federation, Familiar with Jewish Family Service, At least somewhat familiar with at least on local Jewish agency, and Keep kosher outside home are respondent only questions. All other questions are anyone in the household. 3) Loadings of.45 or less are not reported. Cluster 1 High Orthodox (n=3) Table 4 Cluster Analysis of Jewish Identification (Community Level Data Set) Cluster 2 Equal, Except Orthodox (n=20) Cluster 3 High Reform (n=14) Cluster 4 High Reform/Low Just Jewish (n=6) Cluster 5 High Just Jewish (n=12) Orthodox 18% 4% 6% 4% 3% Conservative 27% 33% 29% 25% 21% Reform/ Reconstructionis t 26% 31% 43% 52% 39% Just Jewish 29% 32% 23% 19% 38% Total 100% 100% 100% 100% 100% 33

34 Bergen Baltimore New York Lehigh Valley Hartford Orlando New Haven Richmond Middlesex Minneapolis Rhode Island Miami Washington St. Paul Wilmington Broward Tidewater Harrisburg South Palm Beach Monmouth West Palm Beach Jacksonville Atlantic County Los Angeles Buffalo Cincinnati Philadelphia Howard County Pittsburgh Detroit Atlanta Phoenix Rochester Boston San Antonio Columbus Palm Springs Essex-Morris York Cleveland Martin- Lucie Chicago St. Louis Note: Communities are listed in order from most typical of the cluster to least typical. St. Sarasota St. Petersburg Westport Milwaukee Seattle San Francisco San Diego Charlotte Tucson Denver Portland (ME) Las Vegas Cluster 1 High Orthodox (n=3) Table 5 Demographic Differences Among the Clusters (Table shows means for each cluster) (Community Level Data Set) Cluster 2 Equal, Except Orthodox (n=20) Cluster 3 High Reform (n=14) Cluster 4 High Reform/Low Just Jewish (n=6) Cluster 5 High Just Jewish (n=12) Size of Jewish population 567,700 58,037 99,844 88,233 51,475 Density of the Jewish population* Year of study % Jewish households in the community % with 4 year college degree % age Significance F=4.846 Alpha=.002 F=1.391 Alpha-.250 F=1.130 Alpha=.353 F=1.043 Alpha =.395 F=1.041 Alpha=.396 F Alpha=

35 % age Median household income $96,000 $92,900 93,929 86,166 93,583 F=0.441 Alpha=.778 F=0.185 Alpha=.945 * Percentage of Jewish households living in the top three zip codes for Jewish households. Note: The variable Census Division was crosstabulated with the variable Cluster and no significant relationship was found. 35

36 Cluster 1 High Orthodox (n=3) Table 6 Religiosity Differences Among the Clusters (Table shows means for each cluster) (Community Level Data Set) Cluster 2 Equal, Except Orthodox (n=20) Cluster 3 High Reform (n=14) Cluster 4 High Reform/Low Just Jewish (n=6) Cluster 5 High Just Jewish (n=12) Significance Religious Identity Variables Keep kosher in the home Always/usually light Chanukah candles Couple s intermarriage rate* Attend synagogue services once per month or more Always/Usually light candles Friday night Ethnic Identity Variables Participate in the JCC in the past year Donate to a Jewish charity in the past year Someone in the household is a member of a Jewish organization Very familiar with the Jewish Federation F=12.28 Alpha=.000 F=5.29 Alpha=.001 F=4.728 alpha=.003 F=3.34 Alpha=.018 F=2.261 Alpha=.-076 F=2.20 Alpha=.088 F=2.144 alpha=.091 F=1.558 Alpha=.204 F=1.370 Alpha=0.266 * The percentage of all married couples in the Jewish community in which one spouse is not currently Jewish. 36

37 High Reform (Cluster 1) (N = 96) Table 7 Mean Factor Scores for Orthodox Jews in Four Jewish Identity Clusters Communal Ritual Factor M:.3354 SD: Religious Identity Private Ritual Factor* M: SD: Broad Ethnic Factor* M:.1930 SD: Ethnic Identity Local Ethnic Factor* M:.5402 SD: High Orthodox (Cluster 3) (N=129) M:.4062 SD: M: SD: M:.6728 SD: M:.1517 SD: Equal, Except Orthodox (Cluster 4) (N=584) M:.4025 SD: M: SD: M:.2543 SD: M:.1637 SD: High Just Jewish (Cluster 5) (N=76) M:.4752 SD: M: SD: M:.3159 SD: M: SD: Total (N=885) M:.4020 SD: M: SD: M:.3140 SD: M:.1877 SD: * ANOVA significant at p<.01 M=Mean SD=Standard Deviation 37

38 Table 8 Factor Scores for Conservative Jews in Four Jewish Identity Clusters High Reform (Cluster 1) (N =455 ) High Orthodox (Cluster 3) (N=278) Communal Ritual Factor* M:.3534 SD: M:.5286 SD: Religious Identity Private Ritual Factor* M:.2648 SD: M:.6060 SD: Broad Ethnic Factor* M:.4369 SD: M:.2706 SD: Ethnic Identity Local Ethnic Factor* M:.8925 SD: M:.2117 SD: Equal, Except Orthodox (Cluster 4) (N=4,265) High Just Jewish (Cluster 5) (N=787) M:.5232 SD: M:.3992 SD: M:.1812 SD: M: SD: M:.2333 SD: M:.2380 SD: M:.3499 SD: M:.2157 SD: Total (N=5,785) M:.4932 SD: M:.1826 SD: M:.2518 SD: M:.3677 SD: * ANOVA significant at p<.01 M=Mean SD=Standard Deviation 38

39 Table 9 Factor Scores for Reform/Reconstructionist Jews in Four Jewish Identity Clusters High Reform (Cluster 1) (N =688) High Orthodox (Cluster 3) (N=266) Communal Ritual Factor* M:.2383 SD: M:.3859 SD: Religious Identity Private Ritual Factor* M: SD: M: SD: Broad Ethnic Factor M: SD: M: SD: Ethnic Identity Local Ethnic Factor* M:.9012 SD: M:.1306 SD: Equal, Except Orthodox (Cluster 4) (N=4,053) High Just Jewish (Cluster 5) (N=1,294) M:.2958 SD: M:.1618 SD: M: SD: M: SD: M: SD: M: SD: M:.1894 SD: M:.0631 SD: Total (N=6,301) M:.2658 SD: M: SD: M: SD: M:.2387 SD: * ANOVA significant at p<.01 M=Mean SD=Standard Deviation 39

40 High Reform (Cluster 1) (N =377) Table 10 Factor Scores for the Just Jewish in Four Jewish Identity Clusters Religious Identity Communal Ritual Factor* M: SD: Private Ritual Factor M: SD: Broad Ethnic Factor* M: SD: Ethnic Identity Local Ethnic Factor* M:.4621 SD:.92394k High Orthodox (Cluster 3) (N=278) M: SD: M: SD: M: SD: M: SD: Equal, Except Orthodox (Cluster 4) (N=3,684) High Just Jewish (Cluster 5) (N=1,347) M: SD: M: SD: M: SD: M: SD: M: SD: M: SD: M: SD: M: SD: Total (N=5,686) M: SD: M: SD: M: SD: M: SD: * ANOVA significant at p<.01 M=Mean SD=Standard Deviation 40

41 Figure 1 Orthodox Conservative Reform Just Jewish Profiles of Jewish Identity Clusters 41

42 Figure 2 Boxplot for Communal Ritual Factor (1) 42

43 Figure 3 Boxplot for Broad Ethnic Factor (II) 43

44 Figure 4 Boxplot for Local Ethnic Factor (III) 44

45 Figure 5 Boxplot for Private Ritual Factor (IV) 45