Correlates of experiences and perceptions of anti-Semitism among Jews in the United States.

Social Science Research 47 (2014) 44–60

Contents lists available at ScienceDirect

Social Science Research journal homepage: www.elsevier.com/locate/ssresearch

Correlates of experiences and perceptions of anti-Semitism among Jews in the United States Uzi Rebhun ⇑ Division of Jewish Demography & Statistics, The A. Harman Institute of Contemporary Jewry, The Hebrew University of Jerusalem, Jerusalem 91905, Israel

a r t i c l e

i n f o

Article history: Received 2 July 2012 Revised 23 June 2013 Accepted 17 March 2014 Available online 1 April 2014 Keywords: Anti-Semitism Experiences Multivariate analysis NJPS-2000/2001 Perceptions USA

a b s t r a c t This paper investigates American Jews’ personal experiencing of anti-Semitism and perception of its extent. Analysis of NJPS-2000/2001 indicates that lower age, less education, and American nativity increase experiencing of anti-Semitism. Religious identification and attachment to Israel are positively associated with anti-Semitic experience; friendship with other Jews has the opposite effect. Contextual factors are not significant for the experiencing of anti-Semitism but living in a state that leans toward the Democratic Party has a downward effect. Contrary to experience, younger age and higher education are negatively associated with the perception of a high incidence of anti-Semitism. Being a woman, American born, and living in states with high concentrations of Jews positively affect Jews’ perception of anti-Semitism. A paramount determinant of the perception of anti-Semitism is the individual’s belief that he or she has experienced it. The results are discussed in reference to three working hypotheses of integration, group identification, and environment. Ó 2014 Elsevier Inc. All rights reserved.

1. Introduction Anti-Semitism – hostility toward Jews who are prejudged as radically ‘‘other’’ (Chanes, 1999; Halpern, 1981) – is something we should care about because it attests to intergroup relations and the extent of tolerance toward minorities in society. This study examines and seeks to expand our current understanding of anti-Semitism as the targeted population interprets it. The study looks into the origins of experiences of anti-Semitism and the correlates of perceptions of the extent of anti-Semitism among American Jews. The investigation is based on the National Jewish Population Survey (NJPS) from 2000/2001, which we enhanced by blending area contextual data from official sources. As a major religio-ethnic group in the United States, Jews are exposed to the possibility of experiences and self-perceptions of anti-Semitism. Anti-Semitism views Jews as a collective or personal threat in various respects – social, economic, political, and so on. The perceived threat varies in intensity and may ostensibly trace to fringe groups or major political forces (Tobin, 1988). Anti-Semitism is manifested in both attitudes and behaviors. The spectrum of anti-Semitic incidents is broad, ranging from facial expressions or other subtle communication via demonstrations and hate propaganda to severe crimes (Chanes, 1999; Tobin, 1988). More recently, some observers have postulated that anti-Semitism also includes anti-Israel manifestations – a linkage, especially when presented in the United States, where most Americans support the State of Israel (even if questioning aspects of its policies towards the Palestinians), that is contested (Kaplan and Small, 2006; Wistrich, 2010). Anti-Semitism may be expressed in informal contacts between Jews and non-Jews, in the workplace, in residential ⇑ Fax: +972 2 588 1243. E-mail address: [email protected] http://dx.doi.org/10.1016/j.ssresearch.2014.03.007 0049-089X/Ó 2014 Elsevier Inc. All rights reserved.

U. Rebhun / Social Science Research 47 (2014) 44–60

45

neighborhoods, in applying for municipal or governmental services, and the like. It may be manifested by individuals or by larger bodies or movements. Its effect is amplified when propagated by persons in high public positions and when it reaches a larger audience (Tobin, 1988). Notably, in the United States there are racist organizations and anti-religious groups that act against minorities in general, Jews being but one of their targeted populations. The present study assesses Jewish observation of anti-Semitism as a two-stage process with experience having an important influence on perceptions. The analytical model incorporates socio-demographic characteristics and indicators of Jewish identification. The model also introduces social and economic context variables that are assumed to influence anti-Semitic outcomes. A comprehensive analysis of this kind should yield insights into the yet-unsolved issue of the role of individual characteristics and macro-environmental conditions in influencing or informing experiences and perceptions, and should help understand the dissonance between the successful integration of this religio-ethnic minority into the American social and cultural mainstream (Pyle, 2006) and the same minority’s fears of mistreatment due to its group affiliation (Tobin, 1988). Hence, it contemplates the socio-psychological dynamics of anti-Semitism through the eye of the beholder.1 2. Experiences and perceptions of anti-Semitism: a literature review Earlier studies on personal experiencing and perceptions of anti-Semitism among American Jews concerned themselves mainly with describing the levels of these indicators and were limited to local communities. Seldom did they look into factors that might influence these experiences and perceptions at issue. Their findings suggest that most Jews (50–80%) have experienced some anti-Semitism in their lifetimes but that most among them did so to a small extent only (Tobin, 1988). Data confined to a fixed period show a much lower incidence, of course. In local community studies that were conducted in the 1980s the proportion of respondents who reported having experienced anti-Semitism in the preceding twelve months ranged from a low of 17% in Essex and Morris counties in New Jersey to 28% in Washington, D.C. (Tobin, 1988); in local studies in the 1990s, the proportions ranged from 11% in South Palm Beach, Florida, to 31% in Orlando, Florida (Sheskin, 2001). Since local studies are conducted independently of each other, it is difficult to assess whether the differences reflect periodic effects associated with the time of data collection or contextual effects of the specific locality. Common among Jews in different communities is the high proportion of young people among those who report having experiencing anti-Semitism – a proportion that declines inversely with age. First- and second-generation American Jews who spent their formative years in the first half of the twentieth century retain memories of prejudice and discrimination in acceptance to universities, workplaces, residential neighborhoods, and social clubs, aimed at Jews from all social strata, as well as tensions with non-Jews that sporadically escalated into violence (Diner, 2004; Sarna, 2004; Sorin, 1997). Subsequent generations of Jews were socialized while young after World War II, when where negative feelings about Jews in America declined dramatically (Shapiro, 1992), or even later, since the 1960s, following the civil rights movement and the growing ideological emphasis on multiculturalism (Wistrich, 2010). Thus, they no longer needed to struggle; their successful social integration in America is a fact. Due to their intensive involvement with non-Jews in both formal and informal circles and their exposure to the general world, however, they ‘‘are more likely [than their predecessor] to come into contact with antiSemitic behavior and attitudes where they exist’’ (Tobin, 1988: 20). Empirical evidences also suggest a strong relationship between Jewish identification and self-reported experiencing of anti-Semitism (Dubow et al., 2000). Among Jews in early adolescence (middle-school students), the higher they score on measures of Jewish identity the more they express ethnic-related stress, including remarks about other children making anti-Semitic comments about them. 1 For historical and social background, it is noteworthy that even though anti-Semitism in America dates to the seventeenth century and attained special pinnacles during the Civil War (Sarna, 2004), it was never translated into the kind of official political action against the Jews à la Europe (Diner, 2004). Furthermore, by and large, it did not seriously affect much of American Jewry in those years; it was channeled toward Jews in the military and attacks by evangelists and missionaries in the media and popular culture (Gurock, 1998). A conspicuous change took place in the last quarter of the nineteenth century as more and more Jews from Eastern Europe immigrated to America and climbed the socioeconomic ladder (Gurock, 1998). At this time, prejudices and social discrimination against Jews were also part of an attempt to mitigate the threat that the recent immigrants posed to America’s white Protestant complexion. Illegalities by Jews that evolved into public scandals, mainly concerning illicit sales of alcohol and conspiracies with gamblers to ‘‘throw’’ sporting events, contributed to this. In those years and up to World War II, social discrimination against Jews was manifested in various ways, e.g., student quotas at leading universities and limits Jewish membership in clubs, residence in certain prestige neighborhoods, and staying in hotels. Physical violence against the background of anti-Semitic tension also occurred at times (Sarna, 2004). A dramatic change took place after World War II, American anti-Semitism beginning a major decline. As religion in America gathered strength, Americans’ interest in Judaism took an upward turn as well. It was then that Judaism became widely recognized as America’s ‘‘third faith,’’ alongside Protestantism and Catholicism (Herberg, 1955). These developments placed anti-Semites on the defensive and powered a perceptible decrease in organized anti-Semitic activity. Anti-Semitism now became mostly a fringe social phenomenon (Diner, 2004). Furthermore, by means of federal and state legislation coupled with pressure from returning war veterans and the highlighting of Jews in government and the media, antiJewish discrimination in employment, housing, and daily life decreased significantly. Practically speaking, by the early 1960s, almost all quotas for Jews at colleges were eliminated, all entrance barriers to the liberal professions were downed, and the use of restrictive measures in residential and recreation venues ceased (Diner, 2004). The salient decline in American anti-Semitism has also been reflected in changes in the traditional images of Jews in business as assertive, deceitful, and tricky. Concern among Americans that Jews have too much power has also waned markedly (Quinley and Glock, 1979). Although anti-Semitism has not disappeared and no few Americans continue to adhere to anti-Jewish images and prejudices, overtly or covertly (Smith, 1991), most do not think anything should be done about the Jews’ power or money. Anti-Semitism in the U.S. is not a strong social and political factor today. Concurrently, positive attitudes toward specific characteristics of Jews have become more prevalent (Tobin, 1988), non-Jews are more willing than before to marry Jews, and the number of Jews in the U.S. Senate and House of Representatives has grown (Shapiro, 1992).

46


Other studies focus on Jewish perceptions of anti-Semitism. Despite substantial variations among Jews in socio-demographic characteristics, religious identification, and community of residence, the vast majority agrees with the claim that there is some or much anti-Semitism in America (Breshtel and Graubard, 1992; Chanes, 1999; Tobin, 1988). When Jews say that anti-Semitism exists in America, they do not necessarily refer to the economic or political realm. Rather, they often pinpoint attacks against Jewish institutions or Israel-related activities or base their judgments on hearsay (Chanes, 1999). In fact, most Jews who define anti-Semitism as a problem do not refer to actual prejudice against Jews but to fear of antiSemitism. It is also possible that contemporary Jewish perceptions of anti-Semitism are vestiges of periods when Jews faced real obstacles to social and economic advancement. Today, too, issues on the general American social agenda such as the Christian religious Right, the notion of America as a Christian nation, and criticism of the separation of church and state, may somewhat threaten the security of the Jews (Glock and Stark, 1966). Anti-Semitic incidents in Europe rattle the selfconfidence of Jews in America (Chanes, 1999). Still, respondents in Tobin’s study (1988) were united in claiming that during times of economic crisis anti-Semitism inevitably increases. According to Kim (1995), who utilized data from both the 1970/1971 and 1990 NJPS, community size does not play a significant role in the extent of fear of anti-Semitism. Kim suggests that American Jews’ perception of anti-Semitism may be influenced primarily by national factors and much less by local conditions, and that Jews identified as Orthodox or Conservative have stronger perceptions of anti-Semitism than their Reform or non-denominational counterparts do. On the basis of data from 2001 to 2005, Cohen (2010) suggests that perceptions of the severity of anti-Semitism are positively associated with strong Jewish identity, low income stratification, membership in older generations, local anti-Semitic incidents, and the use of anti-Semitic terms in the news media. Likewise, women are more likely than men, and Democrats are more likely than Republicans, to see anti-Semitism as a serious problem. The importance of group identity in concern about anti-Semitism is also characteristic of the Jewish clergy (Djupe and Sokhey, 2003). Orthodox rabbis publicly address anti-Semitism more often and are more associated with the wish to pledge more resources to combating it than their counterparts in the more progressive Jewish denominations. Djupe and Sokhey’s interpretation is that Orthodox rabbis place more stock than others in Jewish unity and maintaining the boundaries between Jews and non-Jews and between the religious and the secular world; therefore, they view antiSemitism as a more significant problem. However, Djupe and Sokhey found no meaningful effect of the proportion of Jews in the total population of the counties they investigated on differences in rabbis’ perceptions of and actions regarding anti-Semitism. DellaPergola et al. (2009) apply facet theory to assess the empirical structure of dozens of Jewish identificational variables. To this end, they distinguished among Jews who self-define as Orthodox, other Jews who self-define by religion or ethnicity and belong to some Jewish organization, still other Jews who self-define by religion or ethnicity and do not belong to any Jewish organization, and individuals who profess a Jewish connection by criteria other than religion or ethnicity. The analysis shows substantial inter-group variations in the experiencing of anti-Semitism – 31% of the Orthodox reporting an experience in the past year, 24% in the two sub-groups defined by membership in organizations, and only 14% of the nebulously Jewish-connected. Perceptions of the extent of anti-Semitism are more similar among all four subgroups. Both the experiencing and the perception of anti-Semitism correlate strongly with enhanced Jewish activities, prayer, and dating Jews in high school.2 These findings allude to the existence of three major blocs of variables that may be significant in the experiencing and perception of anti-Semitism: individual socio-demographic variables, group identity, and socioeconomic and political macro-characteristics. The discussion that follows elaborates on each of these blocs, places them in an appropriate theoretical context, and develops working hypotheses for the sample obtained. Notably, the data at issue overcome a major limitation that exists in Cohen’s data: unlike Cohen, who had data on only one construct of anti-Semitism – perception – sample used in this study also contains information on the self-experiencing of anti-Semitism. The self-perception theory, as well as the abundance of empirical studies, argues decisively for the significance and inseparability of ties between experience and perception (James, 1890; Laird, 2007; Warr and Knapper, 1968). Likewise, the data used here represent more accurately the entirety of American Jewry, i.e., not only Jews by religion as in Cohen’s study but also others who profess no religion but consider themselves Jewish. Further, Cohen merged five attributes of Jewish identification into a single index. In contrast, and consistent with previous studies that illuminated the multifacetedness of Jewish identification (e.g., DellaPergola et al., 2009), the indicators in this study are designed to reflect different but complementary domains of group commitment among American Jews. This allows an evaluation of the extent to which specific Jewish contexts are associated with the experiencing of anti-Semitism and the crystallization of self-perception of anti-Jewish prejudice and hostility. Thus, the investigation here transcends the existing corpus of literature, which is mainly descriptive (e.g., Chanes, 1999; Dubow et al., 2000; Tobin, 1988), and focuses on a single dimension of the experiencing, or perception, of anti-Semitism (Cohen, 2010; Tobin, 1988).

2 Notably, there are also studies, though very few, on the experiencing and perception of anti-Semitism among Jews in countries other than the United States. This literature focuses on major Jewish communities including in Europe (Staetsky and Boyd, 2012), Mexico (Comite Central Israelita de Mexico, 2005), and Israel (Levy, 1996). These studies are merely descriptive, setting forth the levels of experiencing and perception of anti-Semitism; they make no attempt to link the findings to individual or contextual factors hence, multivariate analysis. Of course, the historical background of each of these countries (e.g., the medieval bonds and the process of ‘‘emancipation’’ in Europe [Sorin, 1997]), the religious composition of the inhabitants of these countries (of Catholics, Protestants, or Muslims), and their economic and political situation are very different from those of the United States.


47

3. Theoretical considerations and hypotheses Perceptions and assessments of prejudice and its manifestations, including behaviors, against an individual’s group may vary among people and across space. Here are developed three working hypotheses on differences in American Jews’ experiencing and perception of the extent of anti-Semitism that are likely to be associated with individual characteristics and area-contextual factors. These three non-mutually-exclusive hypotheses may be defined as the ‘‘integration hypothesis’’, the ‘‘group identification hypothesis’’, and the ‘‘environmental hypothesis’’. The integration hypothesis postulates that members of minority groups perceive local society according to the extent of their structural assimilation. The more they enter and are integrated into the host society’s socioeconomic core, the less alienation one expects them to experience and the more they tend to deny the existence of prejudice against them. Assuming equal-status participation, such close contacts are expected to mitigate preexisting ethnic prejudice and intergroup tension and enhance positive feelings toward the other (Allport, 1954; Hewstone and Brown, 1986). Differences in the experiencing and perceptions of anti-Semitism commensurate with the integration of Jews into American life are reflected in such structural characteristics as education, nature of employment, and nativity status. Below this is called the ‘‘descending’’ version of the integration hypothesis. The effect of minority–majority relations, however, is not unequivocal (Allport, 1954). Structural integration does not necessarily attest to true acquaintance, or interpersonal attraction, between people (Brown and Hewstone, 1995). Daily contacts in schools, on the job, or in the residential neighborhood may be too casual and non-integrative to dispel hostility. Furthermore, according to the ‘‘law of frequency’’, high visibility of members of an out-group amplifies the traditional perceptions and stereotypes by which this out-group is known and strengthens adverse mental associations toward its members. Social convergence may be seen by the dominant group as a threat (Hewstone and Brown, 1986). Arguably, then, Jews who successfully integrated into American society expanded their interaction with non-Jews and, hence, are not protected from but are more exposed to anti-Semitic incidents. Moreover, Jews of high socioeconomic status harbor strong values of social justice and reject any type of racial or ethnic prejudice. As such, they are more sensitive than lower-status Jews to non-Jews’ words and actions (Tobin, 1988). Below this is termed the ‘‘ascending’’ version of the integration hypothesis. The group-identification hypothesis is highly current with the cognitive and motivational processes associated with the social-identity theory (Tajfel and Turner, 1986). This theory departs from the assumption that individuals in intergroup situations ‘‘interact [not] as individuals . . . but as members of their groups standing in certain defined relationships to members of other groups’’ (p. 10). When a person has strong feelings about a distinct group, s/he will view the social environment by means of social categorization that perceptually distinguishes h/her from other groups (Turner, 1991). Consciousness of group membership strengthens emotional attachment and particular behaviors, hence social identification. Since group membership is authentic and figures importantly in a person’s self-esteem, a person will attempt to enhance the status of h/her group by emphasizing social differences vis-à-vis other groups including perceptions and attitudes toward ‘‘them’’. Thus, it is expected that Jewish identification will be positively associated both with the experiencing of anti-Semitism and with the perception of the extent of anti-Semitism in the United States. However, American Jews are spread along a continuum of strengths of group commitment in both the private and the public spheres, in areas including ritual observance, institutional involvement, attachment to Israel, and membership in informal Jewish networks. It is postulated in particular that members of Jewish institutions that organize Jewish communal activities are frequently involved in events that take place in the general public space and, thus, will be more strongly exposed than others to anti-Semitic incidents. We also suggest that this is true for Jews who strongly identify with Israel; these Jews are likely to participate intensively in proIsrael rallies and may find themselves attacked (physically or verbally) or interpreting such attacks as anti-Semitic. Jewish social ties are also expected to be strongly and positively associated with the experiencing and perception of anti-Semitism because they attest to frequent gathering of Jews as well as exchange of information. Consistent with the focus on intergroup relations, one may depart from the aforementioned theory of social identity, which concerns itself with relative social comparison, and shift the emphasis to the importance of the real value of the outcome of group conflict (Sherif, 1967; Tajfel and Turner, 1979). This strategy suggests that the rejection of an out-group, e.g., a prejudice, derives from differences in interests in, and competition for, resources and tangible issues (Kelley and Thibaut, 1978). By saying this, we postulate that attitudes and behaviors are often cumulatively affected over time by incentives from the surrounding environment and by social conditions (Harvey, 1969). In stratified societies, such a conflict reflects social, economic, and political power as well as social prestige. The attitude toward ‘‘the other’’ is largely functional in nature, originating in the other’s being a member of an under – or over-privileged group (Oberschall, 1973). Accordingly, the environmental hypothesis postulates that personal experiences and perceptions among members of a minority group regarding hatred directed against them by the general population are unevenly distributed across the country. This view assumes that events in the nearby environment are more tangible and have a stronger effect than more distant events, especially if associated with personal safety. Since the environmental hypothesis attaches paramount importance to contextual characteristics of area of residence, anti-Semitism is likely to be expressed mainly in the general public arena in demonstrations or political speeches and much less in face-to-face interactions with Jews. Thus, while contextual variables are expected to have an effect of similar direction on the experiencing of anti-Semitism and the perception of its extent, the significance and power of the effect is often likely to be stronger for perception than for experience.

48


A major macro-factor that elicits accusations against Jews in various places and times is the worsening of economic conditions (Anti-Defamation League, 2008; Tobin, 1988).3 Such reactions reflect the view that Jews hold key economic positions and wield excessive power. A paramount economic factor in this respect is the unemployment rate; it can be speculated that the higher the unemployment rate is in a given area it more strongly provokes expressions against and stereotyping of Jews, manifested in experiences of anti-Semitism and perceptions among Jews that anti-Semitism is rife. Studies have also shown racial and religious variations in prejudice toward Jews. Especially high rates of anti-Jewish sentiments emanate from African Americans (Chanes, 1999). We expect to find that large concentrations of African Americans encourage different types of anti-Jewish manifestations and, accordingly, more experiencing and stronger perceptions of anti-Semitism among local Jews. Although liberal-minded people are characterized by strong tolerance of minorities, states in which many people have such political and social orientations may be characterized by strong anti-Jewish attitudes. Partly, this reflects criticism of Israel in regard to the Middle East conflict. More generally, as Lipset and Schneider (1979) already showed back in the 1970s, anti-Semitism has gained strength among young liberals in the intergenerational exchange. Malhotra and Margalit (2009) also revealed that Democrats were especially prone to blaming Jews for the financial crisis of 2008–2009. Thus, Jews who live in pro-Democratic states will report stronger self-experiencing of anti-Semitism and, especially, stronger perceptions of the extent of anti-Semitism in America, than those in Republican-leaning states. Finally, the larger the proportion a minority has in the local population, the higher its implied visibility in major social, economic, and political positions. A large proportion of members of a minority group is perceived as a competition or threat by the majority population. This perspective of the social context may aggravate prejudice against and discrimination toward the minority (LeVine and Campbell, 1972; Welch et al., 2001). Likewise, a large proportion of Jews in a given population may be better able to combat anti-Semitism collectively and in a way that will receive wide public resonance, hence enhancing awareness among Jews of the seriousness of the problem. Overall, the purpose of this study is to explore the correlates of socio-demographic characteristics, group identification, and environmental factors with Jews’ experience and perception of anti-Semitisim in the United States, all of which along the three working hypotheses described above. My data allow me to examine these effects directly. In Section 6, I link these hypotheses with the empirical findings and discuss several broad social and cultural implications. 4. Data, model specification, and sample characteristics 4.1. Data The data for this study are drawn from the 2000/2001 NJPS (Kotler-Berkowitz et al., 2003). The survey was based on a random sample of telephone numbers accrued by a random digit dialed procedure in all fifty states and the District of Columbia. A series of screening questions was introduced to verify any current or past connection with Judaism on the part of household members. The complete sample constituted 5148 respondents.4 My analysis focuses on respondents who defined themselves as Jewish, including those who indicated Jewish as their current religion (Jews by ‘‘religion’’) and those who reported no religion but who considered themselves Jewish (‘‘ethnic’’ Jews). I focus on one adult (aged eighteen and over) from each household, rather than multiple adults, in order to eliminate the potential bias of interdependence in experiences and attitudes. The application of these criteria yielded a sample of 4147 respondents. The data were weighted to account for their differential selection probabilities. Contextual measures at the state level are culled from official publications of the United States Bureau of the Census. 4.2. Model specification The analysis addresses two dependent variables separately. The first is the personal experiencing of anti-Semitism in the past year. Respondents were asked: ‘‘Have you personally experienced any anti-Semitism in the past year?’’ A distinction was made between those who did not experience anti-Semitism (scored as 0) and those who reported having experienced anti-Semitism (scored as 1). The second dependent variable is the perception of the extent of anti-Semitism in the United

3 During the 1973 OPEC oil embargo, for example, when Americans faced long lines at gasoline stations and had to pay steeply escalating prices for petroleum products, various bumper stickers blamed the Jews for the resulting economic hardship (e.g., ‘‘When you run out of oil, burn Jews’’) (Tobin, 1988: 103). In the mid-1980s, in the midst of the farm crisis, extremist groups distributed pamphlets ‘‘filled with anti-Semitic dogma’’ blaming Jewish bankers and a Jewish conspiracy for the entire situation (Tobin, 1988: 198–199). More recently, polls conducted for the ADL (2008) showed a substantial surge in anti-Semitic electronic posts and discussions alleging that Jews and Zionists were at fault for the economic crisis then under way (ADL, 2008). 4 NJPS-2000/2001 was not accepted without controversy (DellaPergola, 2005; Kadushin et al., 2005; Perlmann, 2007). This main challenges concerned design aspects such as low response rate, screening for Jews, and weighting scheme. Yet both senior independent referees (Schulman, 2003) and major critics of the management of the survey (Kadushin et al., 2005) postulated that the methodological limitations, while possibly ruling out an appropriate estimation of the size of the American Jewish population, ‘‘[do not] render NJPS useless; on the contrary, the survey provides valuable data for analyzing interrelations between variables [. . .] which factors are associated with which outcomes [. . .] [and that] sophisticated analysis can illuminate the processes of Jewish life in ways that head counts and cross-tabulations do not’’ (Kadushin et al., 2005: 22).


49

States today. The wording of this measure was, ‘‘How much anti-Semitism do you think there is in the United States today?’’ with four possible answers: not at all (0), a little (1), a moderate amount (2), and a great deal (3).5 The explanatory variables were clustered into three major blocs: Bloc A, composed of demographic and socioeconomic characteristics, is meant to test the validity of the integration hypothesis; Bloc B, relating to Jewish identification, assesses the robustness of the group identification hypothesis; and Bloc C, comprising state-context variables, provides a basis for examination the environmental hypothesis.

4.2.1. Demographic and socioeconomic characteristics The socio-demographic control characteristics used in this analysis are age, gender, marital status, education, nature of employment, nativity status, and political belief. Age is represented by the cohorts 18–24, 25–44, 45–64, and 65 and over (the omitted category). Gender is set to 1 if the respondent is female; males are the omitted category. Marital status distinguishes between married persons and persons currently not married (the omitted category, comprising singles and persons who are separated, divorced, or widowed). Schooling is decomposed into four dummy variables: up to high-school graduation as the reference category, some college, baccalaureate degree, and advanced academic education. Nature of employment distinguishes among respondents who do not work (the reference category), respondents who work for others, and the self-employed. Nativity status is classified between the native-born, coded 1, and those born abroad as the reference category. Also added as a socio-demographic control variable is the respondent’s own political orientation: liberal (slightly liberal, liberal, and extremely liberal combined), moderate and, as the reference category, conservative (slightly conservative, conservative and extremely conservative combined).

4.2.2. Jewish identification Four measures of Jewish identification were established: a ‘‘ritual-religious index’’, a ‘‘communal-ethnic index’’, ‘‘informal social cohesion’’, and ‘‘Israeli identification’’. This multifaceted approach coincides with the broadly described identification of Jewishness in the United States as being concurrently religious and ethnic (Woocher, 1986).6 The ritual-religious index is a composite of three key Jewish religious practices: lighting of Sabbath candles, fasting on the Day of Atonement (Yom Kippur), and attending the Seder (the Passover ritual meal). Since these practices reflect different intensities of religious life of weekly or once-annually observance, they were weighted differently in the index. Lighting Sabbath candles in the home was weighted 3 for always, 2 for usually, 1 for sometimes, and 0 for never. Fasting on last Yom Kippur received a weight of 1 if yes and 0 if no; and attending the Passover Seder was weighted 1 if the respondent reported attending in the most recent year and 0 if s/he did not. This gives the index a range of 0–5. The communal-ethnic index is composed of four variables: synagogue/temple membership, Jewish organizational membership, volunteer work for Jewish organizations, and Jewish philanthropy (Jewish charitable donations). Respondents received 1 point for participation in each of the above and 0 for non-participation. Accordingly, the index ranges from 0 to 4.7 Informal social cohesion is measured by a single variable: the proportion of closest friends who are Jewish. This friendship pattern was ranked by whether none of the respondent’s closest friends is Jewish (0), some are Jewish (1), about half are Jewish (2), most are Jewish (3), and all are Jewish (4). I used three indicators to construct an index of Israeliness. They are the level of emotional attachment to Israel, familiarity with the social and political situation in Israel, and visits to Israel. The first two indicators sorted among those reporting none/little (0), somewhat (1), or very (2), and visits to Israel among those who never visited the country (0), visited once (1) or visited twice or more (2). The index adds the values for each indicator producing a range of 0–6. 5 Notably, this study, like nearly all research studies, has some shortcomings. It assesses anti-Semitism through only two questions, both of which worded in a somewhat limited way. NJPS did not develop them in a way that would inquire into important aspects of the anti-Semitic experience, such as type (physical or verbal), severity, identity of the perpetrator, and location; as well as aspects of perception and, especially, how respondents define anti-Semitism and whence they get their information about it. The approach of the survey – to leave the interpretations of the questions on anti-Semitism to the respondents’ judgment – is largely consistent with previous NJPS, the annual surveys of the American Jewish Committee, and Jewish communal surveys. More detailed indexing could have allowed us to understand whether respondents meant similar or different things when assessing non-Jews’ behaviors and attitudes toward them personally and toward Jews more generally and, thus, to examine different types of experiences and different bases for the perception of anti-Semitism. The aforementioned missing elements are even more salient in light of studies directed at the American population at large, such as the General Social Survey, the American National Election Studies, and public opinion pools of Princeton Survey Research Associates, or Gallup, which include specific items on attitudes toward and interaction with Jews. Likewise, the analysis is constrained to events occurring in past year, whereas those occurring over a much longer time period may also shape perceived ant-Semitism. The question on perceived anti-Semitism may have been improved by using a more optimal constructed measure. Even with these caveats born in mind, the present investigation is a first attempt toward a comprehensive analysis of the associations between individual and area-contextual characteristics and anti-Semitism through Jewish eyes, and of the relations between the experiencing and perception of antiSemitism. Thus, it contributes to a topic that empirical social science has somewhat overlooked in recent decades. 6 Each of the indices of Jewish identification is composed of items that are strongly interrelated. Accordingly, they create clusters, or facets, that are distinguishable from each other. Each such cluster corresponds with a specific content of Jewish identification e.g., rituals, community, and Israel (DellaPergola et al., 2009; Rebhun, 2004). 7 I realize that some components of the religious-ritual index and the communal-ethnic index include both individual and household characteristics. However, as Goldstein and Goldstein (1996) suggested, it is justified to combine them because the household provides the religioethnic context and the means for individuals’ behavior.

50


Table 1 Definitions and summary statistics for analysis variables, NJPS-2000/2001 (N = 4147). Variablea

Definition

Mean (SD)b

A. Demographic and socioeconomic characteristics Age 18–24 Age 25–44 Age 45–64 Gender Marital status Some college Baccalaureate degree M.A. degree or more Self-employed Employee Nativity status Political orientation: Moderate Political orientation: Liberal

=1 =1 =1 =1 =1 =1 =1 =1 =1 =1 =1 =1 =1

0.103 0.296 0.348 0.527 0.595 0.224 0.290 0.292 0.139 0.578 0.833 0.196 0.519

B. Jewish identification Ritual observance Communal affiliation Informal social cohesion Israeli identification

Range Range Range Range

C. State context variables Unemployment rate Political orientation Racial composition

Average percentage 1995–2000 % votes for Democrat candidate in 2000 Presidential elections % of African Americans

5.308 (0.876) 57.536 (7.271) 28.002 (9.867)

D. Group size Jewish population Total population

Percentage of Jews Natural log of group size

4.2851 (2.838) 16.3454 (0.766)

for 18–24 years old for 25–44 years old for 45–64 years old for female for married persons for some college for B.A. diploma for M.A. or higherc for self-employed for employee Born in the U.S. if moderate if liberal 0–5 0–4 0–4 0–6

4.301 1.637 1.415 2.838

(1.660) (1.424) (0.760) (1.706)

a Reference categories are as follow: for age – 65 years and over; for gender – male; for marital status – not-married including single, divorce, widow; for education – through high school graduation; for employment status – not employed; for nativity status – foreign-born; political orientation – conservative. b For continuous measures, numbers in parenthesis are standard deviation. c Including professional diploma.

4.2.3. State context variables Three measures to evaluate the effect of contextual conditions on anti-Semitism were employed: unemployment rate, racial composition, and political orientation. To reduce the possible effect of intertemporal fluctuations in rates of unemployment, the average of the mean total unemployment rate for each state in the five-year interval of 1995–2000 was used. Racial composition is the proportion of African Americans in the population of a given state. State political orientation is evaluated by the percentage of votes cast for the Democratic nominee Al Gore in the 2000 presidential elections. All contextual measures are introduced as continuous variables and are attached to individuals’ records according to their state of residence at the time of the survey. I also included two variables of population size as independent variables. The first is the percentage of Jews in the total population of a given state. Theories of group threat focus on relative, rather than absolute, group size. Data on the number of Jews, which were used as the numerator for calculating this measure, are derived from annual reports on the size and distribution of American Jews by states (Schwartz and Scheckner, 2001). Another independent variable in this category is the total state population. Americans are concentrated in a rather small number of states; approximately half live in only eight states, each having a population of ten million or more. To control for the effects of such a skewed numerical distribution, which may also bias the estimates of the other state-contextual coefficients, the absolute estimate of the total population for each state was inverted to its natural log. All respondents in a given state have the same value for each of the group size’s variables above.8 After the determinants of the experiencing of anti-Semitism were examined, this variable was introduced as an independent variable for the assessment of variations in the perception of the extent of anti-Semitism in the country.9 Summary statistics of the dependent variables and the explanatory variables are shown in Table 1.

8 It should be noted that the attachment of similar values of context measures to all individuals residing in a given state may lead these individuals to share other characteristics that are not included in the survey data, resulting in correlated regression disturbances. Under such conditions, there is a risk that context measures standard errors will be estimated at artificially low levels. To adjust for this bias we utilize Stata’s cluster option, which moderates the assumption of independency of observations within the corresponding variables and requires only that the observations be independent across clusters such as states (Gurak and Kritz, 2000). 9 The highest proportions of missing data pertain to political belief (8.0%) and the ‘‘ritual-religious index’’ (5.9%). The proportion of missing data for the other independent variables varies from nil (gender) to 2.9% (nature of employment). The missing data were untreated and excluded from the analysis.

51

U. Rebhun / Social Science Research 47 (2014) 44–60 Table 2 Percentage of respondents who personally experienced anti-Semitism, the distribution of the perception of the extent of anti-Semitism in the U.S., and the experience (self)–perception (other) relationships: NJPS 2000/2001 (N = 4044). Extent of anti-Semitism (other)

Total A great deal Moderate amount Little Not at all % of Total

Personally experienced of anti-Semitism (self) Yes

No

Total

100.0 53.4 41.2 5.1 0.3

100.0 26.4 50.1 21.9 1.7

100.0 32.9 48.2 17.6 1.3

23.5

76.5

100.0

Spearman’s correlation: .245. Pearson chi-square: 295.1; significance (two-sided) < .001.

100 Perception (P)

90 80

Percentage

70 60 50 Differential P-E

40 30 20

Experience (E)

10 0

Age Fig. 1. Experience of anti-Semitism and perception of extent of anti-Semitism, by age: American Jews in 2000/2001.

5. Results 5.1. Levels of experiencing and perception of anti-Semitism Table 2 shows the percentage distribution of each of the variables of experiencing anti-Semitism (bottom bolded line) and perception of the extent of anti-Semitism (right-end bolded column) and the relationships between these two variables (inner cells). Approximately one of every four adult Jews (23.5%) personally experienced anti-Semitism during the reference year. A slightly higher proportion, one-third (32.9%), believes that there is much anti-Semitism in the United States and another half (48.2%) judges the extent of American anti-Semitism as moderate. Hence, eight of every ten Jews assess the amount of anti-Semitism in the United States as considerable. These two dimensions are interrelated. More than twice as many Jews who experienced a negative act or expression associated with their Jewishness, as against those who did not experience of anti-Semitism, believe that there is much anti-Semitism in the United States (53.4% and 26.4%, respectively). Still, some three-fourths of the latter agree that the extent of anti-Semitism is moderate or even strong. The Spearman’s coefficient suggests a mild to moderate relationship between experienced anti-Semitism and perceived anti-Semitism of r = .245. The experiencing of anti-Semitism fluctuates by age (Fig. 1). Within the pattern of frequent ups and downs, there are several clear differences among broad age intervals. Notable among them is the relatively low level of experiencing anti-Semitism among those aged 25–30. The rate of experiencing anti-Semitism again falls below 30% among people aged 50–65 and declines thereafter to under 20%.10 The perception of the extent of anti-Semitism in the United States as moderate or large also fluctuates by age. However, it is difficult to identify any systematic variations by individual age or between young and elderly respondents. Accordingly, the differential between personally experiencing anti-Semitism and perceiving antiSemitism as a rather significant phenomenon is especially large among elderly Jews.

10 The dip for ages 25–30 is not due to sampling variation. The number of respondents in this age group is 324, identical to the number of respondents in the younger group (18–24), and only slightly smaller than the 31–36 age group (N = 349). Further, the mean for ages 25–30 is .2255 as against .3274 for ages 18–24 and .3053 for ages 31–36 (with a t-test of 2.98 for the comparison of the mean of ages 18–24 and 25–30, and a t-test of 2.38 for the comparison of the mean of ages 25–30 and 31–36 both of which are statistically significant at an alpha level of .05).

52


5.2. Logit analysis of the experiencing of anti-Semitism Table 3 presents the results of binary logistic regression models that predict the occurrence of an anti-Semitic experience during the reference year as against no such occurrence. The first model assesses how individual demographic and socioeconomic characteristics shape the experiencing of anti-Semitism. Model 2 incorporates Jewish identification factors. Model 3 considers the additional influence of contextual socioeconomic and political conditions. Finally, Model 4 retains all the independent variables including population size (Jewish and total). A major demographic determinant of the experiencing of anti-Semitism is age (Model 1). Younger Jews are more likely than elderly Jews to have experienced anti-Semitism. Every age group between 18 and 64 was about three times more likely to experience anti-Semitism than the omitted category of 65+, with the odds ratios slightly diminishing commensurate with the rise in age. No significant differences between Jewish women and Jewish men were found in the experiencing of antiSemitism. Being married decreased the experiencing of anti-Semitism by approximately 20% relative to the unmarried. All of the educational categories are well above the ‘‘high school education and less’’ category in their experiencing of anti-Semitism, although only the ‘‘some college’’ category shows a statistically significant association. It is possible that these latter respondents are in the midst of their studies in universities, which are lively environments for different types of social and cultural demonstrations and political protests, particularly in recent years concerning the Middle East conflict. Others with some college education are dropouts who may be socially and economically unstable and may trace their failure to anti-Jewish attitudes in the surrounding society. Neither of the other socioeconomic characteristics – academic diploma and employment status – is an important determinant in the experiencing of anti-Semitism in the past year. American-born Jews were more exposed to anti-Semitic incidents than the foreign-born were. Since many of the latter are recent immigrants from such countries as the former Soviet Union (FSU), Iran, and Israel, they are likely to be residentially and socially segregated, a situation that mitigates interaction with the general population, including opportunities to experience antiSemitism. Respondents’ own political orientation does not affect their experience with anti-Semitism. Only a few of the above variables have a statistically significant effect; all are positively associated with the experiencing of anti-Semitism. Insofar as these results support one of the two versions of the integration hypothesis, the version supported is the ascending one, according to which intensive contacts between members of a minority group, i.e., Jews, and the majority population aggravate prejudice and discrimination on the part of the latter, which may eventually evolve into manifestations of anti-Semitism. After controlling for demographic and socioeconomic variables, the model estimated on the basis of patterns of Jewish identification shows that tangible current behaviors and attitudes contribute to variations in the experiencing of anti-Semitism (Model 2). Both ritual observance and Israeli identification had statistically significant positive effects, eliciting odds ratios of 1.101 and 1.145, respectively. Friendship with other Jews deters the experiencing of anti-Semitism (an odds ratio of 0.847), possibly attesting to limited contacts with non-Jews, hence fewer opportunities to encounter remarks associated with group belonging. By the same token, this suggests that having more non-Jewish friends is positively associated with the experiencing of anti-Semitism. The religious nature of informal social networks is an indicator not only of group identity but also of the extent of integration into the general society. Hence, this finding corresponds to our aforementioned inference from the socioeconomic variables that integration aggravates the experiencing of anti-Semitism. I had expected affiliation with the organized Jewish community to amplify the experiencing of anti-Semitism. Here the findings disproved the hypothesis. It is possible that membership and communal involvement take place mainly in the Jewish square and involve little interaction with the general non-Jewish population, at least in regard to issues that may evoke anti-Semitic sentiments. Nevertheless, the group-identity hypothesis is supported by the finding that more identificational variables had a significant and positive relation with the experiencing of anti-Semitism. Socioeconomic and political state-context characteristics are not very meaningful in reporting an anti-Semitic experience (Model 3).11 An exception is the political orientation of the local population: the more pro-Democratic it is, the less anti-Semitism Jews experience (an odds ratio of .983). This may reflect a general atmosphere of weak prejudice against minorities in general and Jews in particular, as well as strong acceptance of cultural pluralism. It seems a valid assumption that since Jews are concentrated in a limited number of states and since the overwhelming majority of American Jews are Democrats, their voting patterns may delineate both the substantial concentration of Jews in a given state and, accordingly, the lesser likelihood of their experiencing anti-Semitism. Indeed, the incorporation of the group-size variables into Model 4 accounted for the effect of the political orientation of the state’s population, which is no longer statistically significant.12 Neither of the group-size variable, Jews or total population, has a statistically significant effect. These observations largely refute the foregoing assumptions in the environmental hypothesis about the direction and significance of the relations between state-context characteristics and the experiencing of anti-Semitism in the past year. The more variables we introduced in the model, the greater its explanatory power became. An improvement in explaining variation in experiencing of anti-Semitism was observed in the passage from Model 1 (of demographic and socioeconomic

11 It should be noted that contextual variables often have a somewhat weaker effect than survey variables in survey-based models and thus should be assessed with some caution. For an excellent review of methodological considerations in efforts to explain individual-level dependent variables through combinations of individual- and group-level independent variables, see: Blalock, 1984. 12 A preliminary analysis, not shown here, in which I introduced the three socioeconomic and political contextual factors and the variable of Jewish population (without that of the total population size), already resulted in the cancellation of the statistically significant effect of the state’s political orientation.

53


Table 3 Logistic regression (odds ratios) of personal experience of anti-Semitism in past year on demographic and socioeconomic characteristics, Jewish identification, state context characteristics, and group size: NJPS-2000/2001.a Independent variables

Integration Hypothesis Model 1

Demographic and socioeconomic characteristics Age 18–24 3.316*** (.170) Age 25–44 3.026*** (.143) Age 45–64 2.934*** (.137) Female 0.872 (.079) Married 0.814* (.083) Some college 1.530*** (.125) Baccalaureate degree 1.047 (.126) M.A. degree or more 1.203 (.129) Employee 1.049 (.119) Self-employed 0.962 (.148) Native-born 1.506*** (.120) Political orientation: Moderate 1.135 (.115) Political orientation: Liberal 1.061 (.093)

Group identification Hypothesis Model 2

Environmental Hypothesis Model 3

Total Sample Model 4

3.178*** (.180) 3.162*** (.155) 3.051*** (.146) 0.865 (.083) 0.720*** (.090) 1.506** (.131) 0.948 (.134) 1.074 (.137) 1.071 (.126) 0.970 (.156) 1.697*** (.130) 1.131 (.121) 1.123 (.098)

3.174*** (.940) 3.008*** (.601) 2.947*** (.556) 0.867 (.104) 0.689*** (.068) 1.472** (.238) 0.951 (.110) 1.089 (.160) 1.142 (.168) 1.000 (.189) 1.662*** (.265) 1.131 (.174) 1.096 (.128)

3.211*** (.959) 3.033*** (.613) 2.981*** (.568) 0.862 (.102) 0.693*** (.069) 1.483** (.237) 0.958 (.112) 1.097 (.160) 1.127 (.167) 0.983 (.190) 1.679*** (.271) 1.117 (.177) 1.082 (.129)

1.101** (.033) 1.043 (.038) 0.847** (.063) 1.145*** (.028)

1.093** (.064) 1.050 (.054) 0.855* (.060) 1.148*** (.035)

1.095** (.065) 1.051 (.054) 0.857* (.058) 1.146*** (.035)

1.021 (.118) 1.001 (.006) 0.983** (.011)

0.982 (.132) 0.998 (.007) 0.991 (.019) 0.971 (.032) 1.142 (.119) 0.096 (3593)

Jewish identification Ritual observance

–

Communal affiliation

–

Informal social cohesion

–

Israeli identification

–

State context characteristics Unemployment

–

–

Proportion of African–American

–

–

Vote for democrate candidate

–

–

Group size Jewish population

–

–

–

Total population

–

–

–

Nagelkerke R2 (N)

0.069 (3968)

0.092 (3693)

0.094 (3593)

*

p < .05. p < .01. *** p < .001. a Number in parentheses are standard errors. **

characteristics) to Model 2 (which incorporated Jewish identification). The contribution of state-context characteristics (in Model 3) was slim. Taken together, the twenty-two independent variables in the full model were moderately effective in explaining variation in the experiencing of anti-Semitism, with an R2 of 9.6%. 5.3. Ordinal logistic regression of perception of anti-Semitism Next, perceptions of the extent of anti-Semitism were examined (Table 4). These perceptions are assumed to be determined by demographic and socioeconomic characteristics, Jewish identification, and state-context characteristics (including

54


group concentration), as well as actual anti-Semitic experiences. Given the character of the dependent variable, with four possible values that are inconsistently spaced (not at all, a little, moderate, a great deal), the data were applied by means of an ordered-logit regression. While younger Jews reported more anti-Semitic experiences than elderly Jews did (in the past year), they are negatively associated with judging the level of anti-Semitism in the United States to be high (Model 1). The coefficients are statistically significant but their size declines inversely with age. Jewish women are more sensitive than their male counterparts to prejudice and activities against their group, and are more likely to perceive the level of anti-Semitism in the United States as high. Being married is negatively associated with the perception of much anti-Semitism. Likewise, as education rises the probability of judging anti-Semitism to be rife declines. Employment status is not significantly associated with the perceived extent of anti-Semitism. American-born Jews tend more than foreign-born Jews to perceive a great deal of anti-Semitism in their country. This may reflect the areas of origin of many Jewish immigrants to the United States in recent decades – especially the FSU and some Muslim countries – which had or have formal anti-Jewish policies. Respondents’ political orientation is not significantly associated with their self-perception of the extent of anti-Semitism. Unlike the findings on the experiencing of anti-Semitism, the relations between structural characteristics and the perception of the extent of anti-Semitism coincide with the first version of the integration hypothesis: insofar as socioeconomic characteristics are significant determinants of the perception of anti-Semitism, they often have a descending effect. The findings are also consistent with the assumption that whereas young and middle-aged Jews experience more anti-Semitism than elderly Jews do, the younger they are the less they perceive anti-Semitism as rife. Young and middle-aged Jews were raised in post-World War II America; they are confident of their acceptance by the surrounding environment, evidenced inter alia by their academic achievements and their assessment of themselves as an integral part of the social mainstream. Being Jewish has not been an obstacle to their unprecedented success in America. Among the identificational factors (Model 2), only ritual observance and Israel identification display statistically significant relations with perceptions of anti-Semitism – positive and negative, respectively, with the size of both coefficients being rather small. In fact, in later models (4–6) the positive relation between ritual observance and perception of the extent of anti-Semitism turned insignificant. Thus, as far as the perception of anti-Semitism is concerned, the findings do not support the group-identity hypothesis. Consistent with their more limited experience of anti-Semitism, Jews who live in states that have large proportions of Democrats (Model 4) also report perceiving less anti-Semitism in the United States. Other contextual factors have no significant effect. High concentration of Jews appears to affect perceptions of anti-Semitism positively (Model 4). High proportions of Jews strengthen the communal infrastructure and contacts among Jews, encouraging the exchange of information about anti-Jewish incidents. Contrary to the environmental hypothesis, Jews who inhabit a state that has a strong liberal orientation perceive less anti-Semitism in America. Consistent with expectation is the positive Jewish population-perception of anti-Semitic association. The paramount determinant of the perception of anti-Semitism, however, is the individual’s having experienced it (Model 5). For each unit increase in the experiencing of anti-Semitism (from 0 to 1), a 1.5 increase in the order log odds of the perception of the extent of anti-Semitism is expected, all other variables in the model held constant. Moreover, while the upgrading of earlier models as far as Model 4 did not improve explanatory efficacy meaningfully, the insertion of experiencing anti-Semitism doubled the explanatory power from 9.9% in Model 4 to 19.7% in Model 5. Several interaction terms of anti-Semitic experience by Jewish identification variables were specified and each was entered into the model separately. Only the interaction between experiencing anti-Semitism and the Israel identification index attained statistical significance (Model 6). Thus, while Jews who identify strongly with Israel are less likely to consider antiSemitism widespread in the United States, their perception turns around if they personally have an anti-Semitic experience.13

13 The testing model used here, according to which experience effects perception, relies heavily on the theoretical socio-psychological literature (e.g., James, 1890; Heider, 1958; Warr and Knapper, 1968). Likewise, previous research on anti-Semitism argues that a ‘‘predisposition to manufacture anti-Semitic experiences because of the belief that a great deal of anti-Semitism is prevalent has not been demonstrated’’ (Tobin, 1988: 108). To further justify the order of this analysis, i.e., to control for the possibility that the two variables of interest – experience and perception – simultaneously determine each other (i.e., endogeneity), a two-stage estimation was employed, with the exogenous (instrumental) variable being the respondents’ stance on the claim that Jews have a special responsibility to take care of Jews in need. Among many individual and contextual variables that were tested, this was the one that best met the condition of not being correlated with the perception of anti-Semitism (r = .015, p = .344) while being correlated with the experiencing of anti-Semitism (r = .085, p = .000). The model estimated below with the predicted value for the endogenous variable, which replaced the original endogenous variable, shows a positive and statistically significant effect of experiencing anti-Semitism (PREDICTEDEXPERIENCE) on the perceived extent of anti-Semitism (PEREXTANTI). More generally, many of the independent variables perform similarly across both equations of the original variable of experience and the predicted variable. Hence, there is reason to feel comfortable with testing model chosen.PEREXTANTIi = 1.077AGE18-24i*** 0.419AGE25-44i*** 0.236AGE45-64i* + 0.514FEMALEi*** 0.205MARRIEDi ** 0.125SOMECOLLEGEi-0.445BAi *** 0.761MAORMOREi *** 0.091EMPLOYEEi + 0.004SELFEMPLOYEDi + 0.801 *** + 0.107MODERATEi + 0.017LIBERALi + 0.043RITUALSi 0.034COMMUNALAFFILIATIONi + 0.022SOCIALCOHESIONi 0.046ISRAELIIDi + 0.022UNEMPLOYMENTi 0.001AFRICANAMERICANi 0.018VOTEDEMOCRATEi** + 0.058JEWISHPOPi** + 0.045TOTALPOPi + 0.348PREDICTEDEXPERIENCEi*.

55


Table 4 Ordered-logit coefficients from the regression of perception of the extent of anti-Semitisim on demographic and socioeconomic characteristics, Jewish identification, state context characteristics, group size and personal experience of anti-Semitism: NJPS-2000/2001.a Independent variables

Integration hypothesis Model 1

Demographic and socioeconomic characteristics Age 18–24 1.006*** (.138) Age 25–44 0.343*** (.107) Age 45–64 0.170 (.101) Female 0.520*** (.066) Married 0.174* (.070) Some college 0.094 (.103) Baccalaureate degree 0.465*** (.101) M.A. degree or more 0.742*** (.105) Employee 0.140 (.096) Self-employed 0.004 (.120) Native-born 0.906*** (.094) Political orientation: Moderate 0.040 (.098) Political orientation: Liberal 0.019 (.082)

*

Total sample and exp. anti-Semitism with interaction terms Model 6

1.083*** (.226) 0.424*** (.140) 0.238* (.101) 0.512*** (.085) 0.214** (.070) 0.127 (.145) 0.456*** (.167) 0.767*** (.180) 0.106 (.165) 0.011 (.152) 0.810*** (.114) 0.068 (.157) 0.040 (.106)

1.077*** (.227) 0.419*** (.140) 0.229* (.105) 0.513*** (.086) 0.220** (.071) 0.116 (.148) 0.445*** (.171) 0.756*** (.181) 0.092 (.162) 0.022 (.151) 0.827*** (.111) 0.074 (.154) 0.007 (.108)

1.421*** (.271) 0.708*** (.155) 0.496*** (.139) 0.594*** (.092) 0.149* (.083) 0.214* (.148) 0.453*** (.171) 0.827*** (.180) 0.120 (.171) 0.054 (.177) 0.750*** (.123) 0.025 (.164) 0.036 (.112)

1.429*** (.271) 0.710*** (.154) 0.500*** (.139) 0.589*** (.093) 0.159* (.083) 0.218* (.149) 0.454*** (.174) 0.827*** (.182) 0.139 (.171) 0.042 (.177) 0.742*** (.125) 0.026 (.164) 0.033 (.113)

0.061* (.027) 0.019 (.031) 0.036 (.052) 0.045* (.023)

0.058* (.027) 0.029 (.030) 0.050 (.047) 0.036 (.046)

0.052 (.028) 0.022 (.029) 0.032 (.044) 0.038 (.047)

0.022 (.035) 0.033 (.022) 0.089 (.046) 0.079*** (.050)

0.023 (.035) 0.030 (.022) 0.087 (.047) 0.111*** (.056)

0.045 (.091) 0.002 (.007) 0.005 (.007)

0.026 (.081) 0.001 (.006) 0.019** (.007)

0.036 (.086) 0.002 (.007) 0.018* (.010)

0.039 (.086) 0.001 (.006) 0.017* (.010)

0.060*** (.019) 0.035 (.057)

0.069*** (.021) 0.009 (.057)

0.068*** (.021) 0.008 (.057)

Environmental hypothesis

Total sample

Model 3

1.109*** (.145) 0.409*** (.116) 0.227* (.107) 0.482*** (.069) 0.212** (.074) 0.113 (.107) 0.460*** (.106) 0.751*** (.110) 0.118 (.100) 0.020 (.125) 0.835*** (.101) 0.077 (.101) 0.013 (.085)

Jewish identification Ritual observance

–

Communal affiliation

–

Informal social cohesion

–

Israeli identification

–

State context characteristics Unemployment

–

–

Proportion of African–American

–

–

Vote for democrate candidate

–

–

Group size Jewish population

–

–

–

Total population

–

–

–

Experience in anti-Semitisim Experienced anti-Semitism in past year

–

–

–

–

1.525*** (.128)

1.148*** (.228)

Interaction: experienced anti-Semitism* Israeli identification

–

–

–

–

–

Nagelkerke R2 (N)

0.095 (3629)

0.095 (3431)

0.095 (3332)

0.099 (3332)

0.197 (3325)

0.128** (.060) 0.199 (3325)

p < .05. p < .01. *** p < .001. a Number in parentheses are standard errors. **

Model 4

Total sample and exp. anti-Semitism Model 5

Group identification hypothesis Model 2

56


6. Discussion and conclusions 6.1. Discussion This article probes the correlates of anti-Semitism in the United States as viewed by Jews. It expands on earlier investigations by examining anti-Semitism as perceived by those targeted by it, an approach postulating that behavior or experience creates feelings. The findings attest to the importance of analyzing experience and perception simultaneously. Jews’ self-experiencing of anti-Semitism is positively associated with their perception of the extent of anti-Semitism in the United States. Despite the strong relationship, however, one cannot ignore the substantial differences between the level of American anti-Semitism as experienced and the level perceived.14 In this context, the finding in NJPS-2000/2001 that slightly more than one-fifth of adult American Jews reported personally having experienced anti-Semitism in the preceding year closely resembles the proportion of non-Jews whom Jews believe to be actively anti-Semitic (Tobin, 1988: 113). It also coincides with the percentage of Americans who were identified as having anti-Semitic propensities (ADL, 2011). Obviously, not all persons whom Jews consider actively anti-Semitic, or who express anti-Semitic attitudes, act on their beliefs. By the same token, others frequently make remarks with anti-Semitic contents. What is intriguing is that three independent sources, addressing both Jews and nonJews, are in much harmony regarding the proportion of persons who belong to the hard core of American anti-Semitism. The one-fifth rate reported in NJPS, suggesting that nearly 1 million American Jews experienced anti-Semitism over a fixed period of one year, is wholly incongruous with reports from the Anti-Defamation League, which in 1999 documented 1547 anti-Semitic incidents in the United States (ADL, 1999), and with those of the FBI, which documented 1043 anti-Jewish incidents in 2001 (FBI, 2001). The differences between NJPS and the two latter sources trace to different methodologies of data collection and definitions. The NJPS definitions clash with the rigid definitions used by ADL and FBI which document only meaningful overt expressions of anti-Jewish hostility. Similarly, ‘‘a series of apparently related incidents, such as similar anti-Semitic graffiti painted on neighboring Jewish properties in one night, or a mass mailing of anti-Semitic material to many recipients in a particular neighborhood’’ (ADL, 1999) is counted by ADL as one incident even though all Jews who are effected by such harassment, if asked in a survey, are likely to report that they personally experienced anti-Semitism, yielding a high rate of positive responses. Further, the vast majority of Jews who say they have experienced anti-Semitism do not report it to any Jewish organization or federal agency (Tobin, 1988). Both self-experience and reportage by such organizations as the ADL nourishes a sense of the extent of anti-Semitism. Likewise, people who experience anti-Semitism may tell friends or family about it even if they do not report it to any organization increasing the number of Jews who perceive anti-Semitism in the United States. In addition, Jews who did not experience anti-Semitism in recent years may have been targets of anti-Jewish remarks or behaviors at an earlier time, establishing a perception for the long term. Not less significant is that within the generally precipitous but not linear decline in anti-Semitism as documented by the above sources, each year in the 1990s, ADL documented more than one-thousand acute anti-Semitic incidents in the United States – some twenty per week on average. Many were probably reported in Jewish newspapers; some even caught the attention of the American mass media. Even if carried out by a small number of people, such a large volume of significant attacks may have a strong influence on the consciousness of and, in turn, perceptions about the presence of anti-Semitism. While an integrated look at experience and perception is of paramount importance, the explanatory factors affect each of these two dimensions of anti-Semitism differently. A positive relation was found between structural assimilation and the experiencing of anti-Semitism. This direction of the relation supports the ‘‘ascending’’ version of the integration hypothesis. It was also found, however, that social and economic success is negatively associated with the perception of the extent of antiSemitism and accordingly accords with the ‘‘descending’’ version of the same hypothesis. One may speculate that native-born American Jews who climbed to high rungs on the socioeconomic ladder are deeply acclimated in general American society. When such people encounter anti-Semitism, it is manifested by one person in an inter-personal contact, appears under specific circumstances, and is momentary (‘‘situational cues’’; Laird, 2007). Most of the socioeconomically advanced population does not entertain pejorative attitudes toward Jews; as education rises, anti-Semitism gradually evaporates (ADL, 2011). The environment that most Jews inhabit rejects any display of anti-Semitism and, more generally, beliefs and behaviors against people based on group membership (Oliver and Mendelberg, 2000). Highly educated and economically successful Jews who have experienced anti-Semitism are in contact with non-Jewish peers of similar socioeconomic status, are aware of the latter’s reaction to prejudice against them, and can appropriately judge the true severity and significance of anti-Semitism. Of similarly effect is the evidence that strong Jewish identification amplifies the experiencing of anti-Semitism. This strongly coincides with the group identification hypothesis. Be it cause or effect, an important element of prejudice and discrimination against American Jews is contained within particularistic group behaviors. Strong Jewish identification, however, is not channeled to the perception of much anti-Semitism in the United States. This may attest to a uniform level of self-confidence among Jews in their status in America. The role of contextual factors is not as important as that of individual affinities. The relations associated with the environmental hypothesis were either insignificant or in directions opposite of those predicted and, overall, provide weak 14 I know of no study that examines the experiencing and perception of prejudice and hatred among other American minority groups. Accordingly, it is difficult to assess whether the disconnect between these two dimensions is large, about average, or small. This is an important issue that remains open; it deserves attention in studies on specific religious and ethnic populations as well as general surveys such as GSS or those of Pew Forum.


57

support for this hypothesis. True economic and social resources such as jobs, and the presence of other minorities that carry the potential of inter-group competition, are not important determinants of the experiencing or perception of anti-Semitism. A factor not associated with a tangible good, i.e., the Democratic political orientation of a state’s population, attenuates both the experiencing and the perception of anti-Semitism. This contradicts the working hypothesis, drawn from several studies, that shows that states with large Democratic propensities have strong anti-Jewish attitudes and that Democrats, more than Republicans, tend to blame Jews for economic crises. Thus, my findings coincide with the argument that liberals view minority groups more positively than conservatives do (e.g., Huddy et al., 2005), and are more supportive of civil rights and equal opportunities in society (Link and Oldendick, 1996). In the past, African Americans’ negative attitudes toward Jews stemmed primarily from tensions caused by the ‘‘middleman minority’’, i.e., the consumer-retailer relationship that typified the two groups’ economic interaction (Martire and Clark, 1982). Over time, large numbers of Jews have shifted from self-to salaried employment, substantially diminishing their non-equal economic relationships with African Americans. These trends have helped to attenuate prejudice and negative stereotypes among African Americans toward Jews. Ostensibly, the Jews’ transition to salaried labor and, in some cases, high-ranking positions in large corporations and public and political institutions may draw arrows of guilt in their direction when economic conditions decline. The fact that upturns in unemployment do not exacerbate Jews’ experiencing and perception of anti-Semitism may attest that early twenty-first century America does not tend to blame the Jews, or other origin groups for that matter, for economic downturns. A high concentration of Jews has somewhat contrasting effects on both dimensions of anti-Semitism: a large proportion of Jews in a specific state attenuates, though not significantly, the experiencing of anti-Semitism but strengthens the perception that anti-Semitism is widespread in the United States. The latter finding fits the working hypothesis of this study and coincides with the notion of the threat that a large minority poses to the majority’s standing. The decline in experiencing of anti-Semitism, however, largely refutes the threat theory. This study does not adopt the proposition as floated in Germany that an increase in the percentage of members of an ethnic minority affords the majority greater opportunities for intergroup contact and thus mitigates majority’s prejudice (Wagner et al., 2006). As shown above, social integration is positively associated with the experiencing of anti-Semitism. Rather, it is argued that the presence of a large percentage of people who share a group identity reflects ethnic segregation and not integration. A critical mass of co-ethnics facilitates the establishment of various communal structures (Weissbach, 2005); this, in turn, enhances intra-group contacts, strengthens group cohesion, and provides the necessary power to combat prejudice. This group cohesion is likely to be utilized to diminish hatred in the local sphere and hence to attenuate the experiencing of anti-Semitism while leaving the perception of the extent of anti-Semitism, which concerns itself with the national domain, unaffected. According to the testing model proposed and empirically validated above, the experiencing of anti-Semitism affects the perception of the extent of anti-Semitism. Nevertheless, there is some relation in the opposite direction as well, i.e., perceptions may sensitize people to incidents. If a person feels that anti-Semitism is a big problem, s/he may be more inclined to construe a snub, a remark, or a facial expression as an anti-Semitic incident, whereas a person who does not consider antiSemitism in the community at large much of a problem may disagree. More generally, various dimensions including feelings and actual behaviors are subject to complex interactions and inter-effects, meaning that a given behavior may be the consequence rather than the cause of emotion or feelings experienced (DellaPergola, 2011). 6.2. Conclusions Two years ago, Cohen (2010) contributed his article ‘‘Perceptions of Anti-Semitism among American Jews, 2000–2005, A Survey Analysis’’ to the field of studies of anti-Semitism studies in the United States. Unlike the social-science literature on anti-Semitism to that time, which was descriptive in nature, Cohen attempted to understand the determinants of the severity of anti-Semitism as perceived by Jews. He traced changes in these determinants over time using a set of annual surveys that repeated similar questions on anti-Semitism and key socio-demographic and identificational attributes. By incorporating data from independent sources into his inquiry, he was able to assess the effects of individual variables and of contextual factors, including major interaction terms. This in-depth analysis should be viewed as a seminal probe into the topic of antiSemitism, triggering a discourse wider than the parochial boundaries of American Jewish studies. This newly opened area now calls for the investigation of complementary dimensions, the incorporation of additional determinants, follow-up of their dynamics, and the accumulation of findings and cogent disagreements, all of which contributing to the development of the topic into an important subfield in social-science research. The present investigation is a development in this direction. The similarity between the findings above and Cohen’s yields robust paradigms of how Jews perceive anti-Semitism. Although the question in this study on self-perception of anti-Semitism was worded somewhat differently, both studies reveal that the overwhelming majority of Jews perceive some or much anti-Semitism in the United States and/or that anti-Semitism is a somewhat serious or serious problem there.15 Both studies conclude that older Jews are more likely than younger Jews to describe the severity of anti-Semitism as strong, a response that 15 Of course, the data can be framed in a milder way, according to which the vast majority (67%) report perceiving moderate to no anti-Semitism. However, on a four-level scale of not at all, little, moderate amount, and a great deal, some 80% checked the two highest categories. My main purpose in reporting this is to underscore the dissonance between the Jews’ successful social integration and full acceptance in the American core society and their widely held feeling that there is much anti-Semitism in the country.

58


presumably reflects the socialization of the former group in earlier periods of American life, before or shortly after World War II. Likewise, both studies find that educational and economic attainment dissuades Jews from feeling that the surrounding environment hates and discriminates against them. Given intergenerational replacement and taking into account the American Jews’ ongoing social advancement, it stands to reason that these Jews’ perception of the extent of anti-Semitism will diminish in the future. The two studies do appear to disagree somewhat about the association of Jewish identification with the perception of anti-Semitism. While the findings above do not attest to the importance of identification for Jews’ perception of the extent of anti-Semitism and indicate that strong identification may sometimes actually have a mitigating effect on such a perception, Cohen documented a positive association between Jewish identification and the likelihood of viewing anti-Semitism with gravitas. However, Cohen is cautious. Employing a refined two-stage analysis – which indeed should be treated carefully due to some dissatisfaction with the exogamic variable – he shows that Jewish identification does not have a statistically significant effect on perception of anti-Semitism. Although this study measured Jewish identification differently, its findings are reasonably harmonious with Cohen’s two-stage estimation. Thus, research seems to be on the path to cementing the notion that Jewish group identification does not affect perceptions of the extent or severity of anti-Semitism. It is also suggested that the actual experiencing of anti-Semitic incidents – either indirectly, through the media or interpersonal communication, or directly, as a victim – fosters feelings of the severity of this phenomenon. The magnitude of this effect, however, differs widely depending on the physical proximity of the individual Jew to the location of the anti-Semitic incident occurred. An offense or injury to the individual h/herself has an especially strong effect on the extent of anti-Semitism that s/he perceives. Further studies on this topic may make it possible to develop a scale of an individual Jew’s distance from an anti-Semitic incident and the intensity of h/her perception of the extent of anti-Semitism. In the context of the United States, the physical distance may be the entire country or the individual Jew’s state, city, town, or neighborhood of residence. Of course, if the incident occurs far away but involves a family member or an acquaintance, this familiarity may be more important than physical proximity per se. In this regard, both Cohen and the present study found that strongly identified Jews are more sensitive to anti-Semitic incidents than others. It was not possible from Cohen’s identificational metric to determine whether different components of Jewish identification react similarly to incidents in the crystallization of his respondents’ perception of the severity of antiSemitism. The present investigation, however, found that a single group factor – identification with Israel – is significantly sensitive to the experiencing of anti-Semitism. Overall, prejudice against Jews – anti-Semitism – should be analyzed via a two-dimensional matrix. The first dimension expresses the mode of prejudice; the second dimension reflects the locus, or area, that shapes the view of prejudice in the eyes of the targeted population. The mode of prejudice is comprised of experience/incident and perception; the locus distinguishes among individual socio-demographic characteristics, group identity, and macro-contextual factors. This creates six cells, each showing the presence or absence of a moderating or escalating effect on the two dimensions of anti-Semitism. While this testing model needs to be replicated and reassessed to strengthen its empirical validity, we should also try to expand it. For the mode of prejudice, given the paramount role of the self-experiencing of anti-Semitism, attempts should be made to incorporate the evaluation of cognitive processes that may elicit interpretations, beliefs, and attitudes toward other people or groups. Research should seek a better understanding of what Jews mean by anti-Semitism and what kinds of antiSemitic experiences they undergo. It should also inquire into the tools they use to judge the extent of anti-Semitism and, perhaps, create a uniform definition of anti-Semitism that can be used as a guide in collecting data on personal or environmental incidents, including their ranking by severity. Such in-depth inquiries into the descriptions of experiences can take this field of research on anti-Semitism to new horizons with important implications also on policy. In the locus of areas, more variables should be included that attest to the extent of integration, e.g., the nature and intensity of relationships between Jews and non-Jews. In the area of identity, a more precise measure of the practical meaning of activity related to institutional membership and attachment to homeland is needed. As for contextual factors, it would be helpful if data were available on smaller geographic units at the SMSA or county level. Further at the macro-level, future research may try to incorporate exogenous measures of anti-Semitism in the general population as contextual variables. As discussed in Footnote 4, several studies, such as the GSS or Gallup, have included questions on intergroup relations, some specifically aimed at attitudes toward and interaction with Jews. This kind of investigation, however, would entail aggregation over multiple surveys, i.e., over time, to obtain enough sample cases. It may also prove difficult if the foregoing proposal, to organize the contextual data according to small geographic units, is adopted. If it is done with new surveys, it will certainly be expensive. Finally, new research may yield wider socio-psychological insights if it adopts a comparative perspective by including other religious and ethnic minorities and assessing their experiencing and perception of a climate inimical to them. The findings above may reflect sensitivities associated with the unique history of the Jews. A broader investigation of the sort proposed would shed more general light on Jewish exceptionalism versus minority status in the open and pluralistic America of the early twenty-first century. References Allport, Gordon W., 1954. The Nature of Prejudice. The Beacon Press, Boston, MA. Anti-Defamation League, 1999. 1999 Audit of Anti-Semitic Incidents. New York, NY. .


59

Anti-Defamation League, 2008. Anti-Semitism and the Economic Crisis. New York, NY. . Anti-Defamation League, 2011. A Survey of American Attitudes toward Jews in America. . Blalock, Hubert M., 1984. Contextual effects models: theoretical and methodological issues. Annu. Rev. Sociol. 10, 353–372. Breshtel, Sara, Graubard, Allan, 1992. Saving Remnants; Feeling Jewish in America. Free Press, New York. Brown, Rupert, Hewstone, Miles, 1995. Contact hypothesis. In: Manstead, Antony S.R., Hewstone, M. (Eds.), The Blackwell Encyclopedia of Social Psychology. Blackwell Reference, Oxford, pp. 123–125. Chanes, Jerome A., 1999. Antisemitism and Jewish security in contemporary America: why can’t Jews take yes for an answer? In: Farber, R. Rosenberg, Waxman, C.I. (Eds.), Jews in America: A Contemporary Reader. Brandeis University Press/University Press of New England, Hanover and London, pp. 124–150. Cohen, Jeffrey E., 2010. Perceptions of Antisemitism among American Jews, 2000–2005: a survey analysis. Polit. Psychol. 31, 85–107. Comite Central Israelita de Mexico, 2005. Estudio Sociodemografico, Economico y Cultural de la Comunidad Judia de Mexico. DellaPergola, Sergio, 2005. Was it the demography?: A reassessment of U.S. Jewish population estimates, 1945–2001. Contemp. Jewry 25, 85–131. DellaPergola, Sergio, 2011. Jewish Demographic Policies: Population Trends and Options in Israel and in the Diaspora. The Jewish People Policy Institute, Jerusalem. DellaPergola, Sergio, Levy, Shlomit, Rebhun, Uzi, Sagi, Dalia, 2009. Patterns of Jewish identification in the United States, 2001. In: Elizur, D., Yaniv, E. (Eds.), Theory Construction and Multivariate Analysis: Applications of Facet Approach. FTA Publications, Israel, pp. 305–318. Diner, Hasia R., 2004. The Jews of the United States, 1654 to 2000. University of California Press, Berkeley. Djupe, Paul A., Sokhey, Anand E., 2003. The mobilization of elite opinion: rabbi perceptions of and responses to anti-Semitism. J. Sci. Study Relig. 42, 443– 453. Dubow, Erick F., Pargament, Kenneth I., Boxer, Paul, Tarakeshwar, Nalini, 2000. Initial investigation of Jewish early adolescents’ ethnic identity, stress, and coping. J. Early Adolesc. 20, 418–441. FBI, 2001. Uniform Crime Reporting. Washington, DC. . Glock, Charles Y., Stark, Rodney, 1966. Christian Beliefs and Anti-Semitism. Harper & Row, New York. Goldstein, Sidney, Goldstein, Alice, 1996. Jews on the Move: Implications for Jewish Identity. State University of New York Press, Albany. Gurak, Douglas T., Kritz, Mary M., 2000. The interstate migration of U.S. immigrants: individual and contextual determinants. Soc. Forces 78, 1017–1039. Gurock, Jeffrey S. (Ed.), 1998. Anti-Semitism in America. Routledge, New York, NY (two volumes). Halpern, Ben, 1981. What is anti-Semitism? Mod. Judaism 1, 251–262. Harvey, David, 1969. Conceptual and measurement problems in the cognitive behavioral approach to location theory. In: Cox, K.R., Golledge, R.G. (Eds.), Behavioral Problems in Geography: A Symposium. Northwestern University Press, Evanstone, IL, pp. 16–28. Heider, Fritz, 1958. The Psychology of Interpersonal Relations. Wiley, New York. Herberg, Will, 1955. Protestant-Catholic-Jew. Doubleday, New York. Hewstone, Miles, Brown, Rupert, 1986. Contact is not enough: an intergroup perspective on the ‘contact hypothesis’. In: Hewstone, M., Brown, R. (Eds.), Contact and Conflict in Intergroup Encounters. Basil Blackwell, New York, pp. 1–44. Huddy, Leonie, Feldman, Stanely, Taber, Charles, Lahav, Gallya, 2005. Threat, anxiety, and support of antiterrorism policies. Am. J. Polit. Sci. 49, 593–608. James, William, 1890. The Principles of Psychology, I–II. Holt, New York. Kadushin, Charles, Phillips, Benjamin T., Leonard, Saxe, 2005. National Jewish population survey 2000–2001: a guide for the perplexed. Contemp. Jewry 25, 1–32. Kaplan, Edward H., Small, Charles A., 2006. Anti-Israel sentiments predicts anti-Semitism in Europe. J. Conflict Resolut. 50 (4), 548–561. Kelley, Harold H., Thibaut, John W., 1978. Interpersonal Relations. Wiley, New York. Kim, Israel, 1995. Perceived anti-Semitism and minority sub-community size. J. Jewish Communal Serv. 71 (4), 293–296. Kotler-Berkowitz, Laurence, Cohen, Steven M., Ament, Jonathan, Klaff, Vivian, Mott, Frank, Peckerman-Neuman, Danyelle, 2003. The National Jewish Population Survey 2000–2001: Strength, Challenge and Diversity in the American Jewish Population. United Jewish Communities, New York. Laird, James D., 2007. Feelings: The Perception of Self. Oxford University Press, Oxford. LeVine, Robert A., Campbell, Donald T., 1972. Ethnocentrism: Theories of Conflict, Ethnic Attitudes, and Group Behavior. Wiley, New York. Levy, Shlomit, 1996. Israeli Perceptions of Antisemitism. Analysis of Current Trends in Antisemitism (acta), No. 10. The Vidal Sasson International Center for the Study of Antisemitism, The Hebrew University of Jerusalem, Jerusalem. Link, Michael W., Oldendick, Robert W., 1996. Social construction and white attitudes toward equal opportunity and multiculturalism. J. Polit. 58, 149–168. Lipset, Seymour M., Schneider, William, 1979. Antisemitism and Israel: A Report on American Public Opinion. Unpublished Manuscript. Department of Sociology, Stanford University, Stanford, CA. Malhotra, Neil, Margalit, Yotam, 2009. Anti-Semitism and the Economic Crisis. Boston Review. May/June. . Martire, Gregory, Clark, Ruth, 1982. Antisemitism in the United States: A Study of Prejudice in the 1980s. Praeger, New York, NY. Oberschall, Anthony, 1973. Social Conflict and Social Movements. Prentice-Hall, Englewood Cliffs, NJ. Oliver, J. Eric, Mendelberg, Tali, 2000. Reconsidering the environmental determinants of white racial attitudes. Am. J. Polit. Sci. 44 (3), 574–589. Perlmann, Joel, 2007. Two National Surveys of American Jews, 2000–2001: A Comparison of the NJPS and AJIS. Working Paper No. 501. The Levy Economics Institute of Bard College, Annandale-on-Hudson. Pyle, Ralph E., 2006. Trends in religious stratification: have religious group socioeconomic distinctions declined in recent decades? Sociol. Relig. 67, 61–79. Quinley, Harold E., Glock, Charles Y., 1979. Anti-Semitism in America. The Free Press, New York. Rebhun, Uzi, 2004. Jewish identity in America: structural analyses of attitudes and behaviours. Rev. Relig. Res. 46 (1), 43–63. Sarna, Jonathan D., 2004. American Judaism: A history. Yale University Press, New Haven and London. Schulman, Mark A., 2003. National Jewish Population Survey 2000–2001: Study Review Memo. United Jewish Communities, New York. Schwartz, Jim, Scheckner, Jeffrey, 2001. Jewish Population in the United States, 2000. American Jewish Year Book, vol. 101, pp. 253–280. Shapiro, Edward S., 1992. A Time for Healing: American Jewry Since World War II. The Johns Hopkins University Press, Baltimore and London. Sherif, Muzafer, 1967. Social Interaction: Process and Products: Selected Essays. Aldine Pub. Co, Chicago. Sheskin, Ira M., 2001. How Jewish Communities Differ: Variations in the Findings of Local Jewish Population Studies. North American Jewish Data Bank, New York. Smith, Tom W., 1991. What do Americans Think about Jews? The American Jewish Committee, New York. Sorin, Gerald, 1997. Tradition Transformed: The Jewish Experience in America. The Johns Hopkins University Press, Baltimore and London. Staetsky, Laura, Boyd, Jonathan, 2012. Jewish Perceptions and Experiences of Antisemitism: Key Findings and Recommendations from the Survey. Institute for Jewish Policy Research, London. Tajfel, Henri, Turner, John C., 1979. An integrative theory of intergroup conflict. In: Austin, W.G., Worchel, S. (Eds.), The Social Psychology of Intergroup Relations. Brooks-Cole, Monterey, CA, pp. 33–47. Tajfel, Henri, Turner, John, 1986. The social identity theory of intergroup behavior. In: Worchel, S., Austin, W.G. (Eds.), Psychology of Intergroup Relations, second edition. Nelson-Hall Publishers, Chicago, pp. 7–24. Tobin, Gary A., 1988. Jewish Perceptions of Antisemitism. Plenum Press, New York and London. Turner, John C., 1991. Social Influence. Open University Press, Milton Keynes, Brooks-Cole, Monterey, CA.

60


Wagner, Ulrich, Christ, Oliver, Pettigrew, Thomas F., Stellmacher, Jost, Wolf, Carina, 2006. Prejudice and minority proportion: contact instead of threat effects. Soc. Psychol. Q. 69, 380–390. Warr, Peter B., Knapper, Christopher, 1968. The Perception of People and Events. John Wiley & Sons, London. Weissbach, L. Shai, 2005. Jewish Life in Small-Town America: A History. Yale University Press, New Haven, CT. Welch, Susan, Lee, Sidelman, Timothy, Bledsoe, Michael, Combs, 2001. Race and Place: Race Relations in an American City. Cambridge University Press, Cambridge, UK. Wistrich, Robert, 2010. A Lethal Obsession: Anti-Semitism from Antiquity to the Global Jihad. Random House, New York. Woocher, Jonathan S., 1986. Sacred Survival: The Civil Religion of American Jews. Indiana University Press, Bloomington, IN.

Adverse Childhood Experiences Among Immigrants to the United States.

Perceptions of drinking problems among undergraduate students in the United States and Scotland.

Race and Gender Differences in Correlates of Death Anxiety Among Elderly in the United States.

Rumble: Prevalence and Correlates of Group Fighting among Adolescents in the United States.

Prevalence and correlates of psychiatric disorders among former juvenile detainees in the United States.

Perceptions of the hospital ethical environment among hospital social workers in the United States.

Female college students' experiences with and perceptions of over-the-counter emergency contraception in the United States.

An examination of sociodemographic correlates of ecstasy use among high school seniors in the United States.

Perinatal experiences of Somali couples in the United States.

Prevalence and correlates of cryptococcal antigen positivity among AIDS patients--United States, 1986-2012.

Prevalence, correlates, and outcomes of cryptococcal antigen positivity among patients with AIDS, United States, 1986-2012.

Trends and correlates of substance use disorders among probationers and parolees in the United States 2002-2014.

Demographic and socioeconomic correlates of powder cocaine and crack use among high school seniors in the United States.

The prevalence of dysphagia among adults in the United States.

Prevalence of hypercalcemia among cancer patients in the United States.

Use of inhaled corticosteroids among Hispanics in the United States.

Characteristics and health perceptions of complementary and alternative medicine users in the United States.

Perceptions and psychosocial correlates of bullying among Lumbee Indian youth.

Correlates of new psychoactive substance use among a self-selected sample of nightclub attendees in the United States.

Voice problems and depression among adults in the United States.

Radiation safety knowledge and perceptions among residents: a potential improvement opportunity for graduate medical education in the United States.

Socio-demographic correlates and trends in low-calorie sweetener use among adults in the United States from 1999 to 2008.

Postdeployment reintegration experiences of female soldiers from national guard and reserve units in the United States.

Neonatal nurses' perceptions of pain management: survey of the United States and China.