¹ We use the term ‘race/ethnicity’ in this article because African-Americans are usually referred to as a racial group and Latinos as an ethnic group. The racial/ethnic groups are defined on the basis of self-identification. Respondents were asked first whether they considered themselves to be Hispanic or Latino. Then all respondents – regardless of their answer to the first question – were asked their race. The small number who identified themselves both as Latino (or Hispanic or Chicane) and as African-American (or black) were asked which they considered themselves mostly, Hispanic or black.

There is no generally accepted nomenclature for the groups to which we refer, and what are the appropriate designations is often a politically volatile question. We use the terms ‘African-Americans’ or ‘blacks’ for one of the minority groups and ‘Latinos’ for the other. We use the term ‘Anglo-Whites’ to denote those who described themselves as white, but not as Latino or Hispanic. The locution is admittedly awkward. Since ‘white’ is usually juxtaposed to ‘black’ or ‘African-American’ and ‘Anglo’ to ‘Latino’ or ‘Hispanic’, however, the conglomerate term for the majority group seems appropriate.

The distinction between African-Americans and Latinos with respect to the date of immigration is not a hard and fast one. Many Latinos – particularly Mexican-Americans – have been in the United States for many generations.

² That African-Americans and Anglo-Whites differ in their party identification, candidate preferences and policy preferences is well documented in the literature. Black Americans are more liberal than Anglo-Whites on matters of economics as well as race. Although often similar in their circumstance of disadvantage, Latinos and African-Americans do not always agree on policy preferences. See, Falcon, Angelo, ‘Black and Latino Politics in New York City: Race and Ethnicity in a Changing Urban Context’, in Garcia, F. Chris, ed., Latinos and the Political System (Notre Dame, Ind.: University of Notre Dame Press, 1988), pp. 178–9.Google Scholar

When the politically salient differences between active and inactive citizens are less apparent than they are in the case of the groups under consideration here, the situation is more complicated. It is well known that political participants are drawn proportionately from the ranks of those of higher socio-economic status. This does not mean, however, that there are corresponding differences in policy preferences. Indeed, Raymond E. Wolfinger and Steven J. Rosenstone have shown that unmistakable demographic differences between voters and non-voters are not matched by differences in attitudes on policy issues as expressed in responses to standard National Election Studies questions: Who Votes? (New Haven, Conn.: Yale University Press, 1980), chap. 6.Google Scholar

Nevertheless, in an analysis that includes consideration of many other participatory acts beyond voting and examination of citizens' actual needs with respect to government policy as well as their political attitudes, we have argued that it does matter whether or not particular groups of citizens are active. See Verba, Sidney, Schlozman, Kay Lehman, Brady, Henry and Nie, Norman H., ‘Citizen Activity: Who Participates? What Do They Say?’ American Political Science Review, 87 (1993), 303–18.CrossRefGoogle Scholar

³ For a full explication of the resource model including an explication of its statistical properties and an analysis of its causal status, see the Verba, Sidney, Brady, Henry, Schlozman, Kay Lehman and Nie, Norman H., ‘Beyond SES: A Resource Model of Political Participation’, unpublished paper.Google Scholar

⁴ See, among others, Olsen, Marvin, ‘Social and Political Participation of Blacks’, American Sociological Review, 35 (1970), 682–97CrossRefGoogle Scholar; Verba, Sidney and Nie, Norman, Participation in America: Political Democracy and Social Equality (New York: Harper and Row, 1972), chap. 10Google Scholar; Danigelis, Nicholas L., ‘Black Political Participation in the United States: Some Recent Evidence’, American Sociological Review, 43 (1978), 756–71CrossRefGoogle Scholar; Welch, Susan, Comer, John and Steinman, Michael, ‘Ethnic Differences in Political and Social Participation: A Comparison of Some Anglo and Mexican-Americans’, Pacific Sociological Review, 18 (1975), 361–82CrossRefGoogle Scholar; Shingles, Richard D., ‘Black Consciousness and Political Participation: The Missing Link’, American Political Science Review, 75 (1981), 76–91CrossRefGoogle Scholar; Guterbock, Thomas M. and London, Bruce, ‘Race, Political Orientation, and Participation: An Empirical Test of Four Competing Theories’, American Sociological Review, 48 (1983), 439–53CrossRefGoogle Scholar; Uhlaner, Carole J., Cain, Bruce E. and Kiewiet, D. Roderick, ‘Political Participation of Ethnic Minorities in the 1980s’, California Institute of Technology, Social Science Working Paper 647, 06 1987Google Scholar; Bobo, Lawrence and Gilliam, Franklin D. Jr, ‘Race, Socio-Political Participation, and Black Empowerment’, American Political Science Review, 84 (1990), 379–92.CrossRefGoogle Scholar

⁵ Our data set provides access to other ethnic classifications though, unfortunately, with inadequate numbers for analysis even in our large sample. We would like to have been able to include Asian-Americans in the analysis. However, we were unable to do so because our screener included only 157 Asian-Americans and the follow-up even fewer.

Most of the data analyses and displays in this article use data from the follow-up survey which contains most of our analytic variables. For some analyses we had to rely on the larger sample for adequate cases. When we use screener data it is indicated. Otherwise, the data are from the follow-up survey. For a table with the effective number of cases for various sub-divisions of our sample, see Appendix A.

⁶ See Olsen, , ‘Social and Political Participation’Google Scholar, and Verba, and Nie, , Participation in AmericaGoogle Scholar. Bobo, and Gilliam, , ‘Race, Socio-Political Participation, and Black Empowerment’Google Scholar and Ellison and Gay argue that, while African-Americans participated at even higher rates than would have been predicted on the basis of their socio-economic characteristics during the late 1960s, there has been convergence since then (Ellison, Christopher G. and Gay, David A., ‘Black Political Participation Revisited: A Test of Compensatory, Ethnic Community, and Public Arena Models’, Social Science Quarterly, 70 (1989), 101–19).Google Scholar

⁷ The scale has a standard deviation of 1.63. The following activities are included in the scale (with proportions engaging in the acts in parentheses): voting in the 1988 presidential election (70 per cent); working in a campaign in the 1988 election cycle (8 per cent); making a campaign contribution in the 1988 election cycle (24 per cent); contacting a government official within the past year (34 per cent); attending a protest, march or demonstration within the past two years (6 per cent); being a member of, or a donor to, a voluntary organization that takes stands in politics (48 per cent); working informally with others in the community to deal with some community issue or problem within the past year (25 per cent); serving in a voluntary capacity on a local governing board or attending meetings of such a board on a regular basis within the past two years (4 per cent). The scale is a simple additive measure of the number of acts in which the respondent engaged. As with other studies, the percentage reporting voting is somewhat inflated.

⁸ Although it could be argued that, from the perspective of democratic theory, this should be a study of citizens only, we did make a deliberate choice to interview non-citizens. Non-citizens are affected by American laws, and many are permanent residents (legally or illegally). There are a number of philosophical questions as to whether they are appropriately part of the universe for a participation study. Since they may – and do – participate in many ways even though they cannot vote, we decided to include them since they can always be separated in analysis. Garcia and Arce note that ‘ineligibility from voting does not totally remove the Mexican-born from the electoral process’ (Garcia, John A. and Arce, Carlos, ‘Political Orientations and Behaviors of Chicanos: Trying to Make Sense Out of Attitudes and Participation’Google Scholar, in Garcia, , ed., Latinos and the Political System, p. 147).Google Scholar

⁹ Our findings about Latino citizens and non-citizens are consistent with those from the National Latino Immigrant Survey as reported by the National Association of Latino Elected and Appointed Officials in ‘Political Participation among Latino Immigrants’, NLIS Research Noies, 1 (10 1992).Google Scholar

¹⁰ The differences among the Latino groups in terms of political activity are consistent with their different socio-economic profiles, with Cuban Americans more likely to have higher levels of education and higher status occupations. For an overview of social and political data among Latino groups, see de la Garza, Rodolfo O. et al. , Latino Voices: Mexican, Puerto Rican and Cuban Perspectives on American Politics (Boulder, Colo.: Westview Press, 1992)Google Scholar. The number of cases of Latinos from the several groups is, in some cases, small (see Appendix A). The data are similar in our larger screener sample using a somewhat different activity scale since not all participation questions are on the screener survey.

¹¹ Paul R. Abramson and William Claggett indicate that the gap between self-reported and validated turnout is especially wide for blacks. (‘Race-Related Differences in Self-Reported and Validated Turnout’, Journal of Politics, 46 (1984), 719–38Google Scholar). We are uncertain as to the applicability of their findings to self-reports of other kinds of political activity.

It is interesting to note how many of the African-American electoral activists were supporting black candidates. We asked those who had either worked in a campaign or made a contribution about the race of the candidate they supported. Seventy-four per cent of the forty-one African-American campaign activists reported supporting a black candidate. In contrast, only 31 per cent of the twenty-five Latino activists reported supporting a Latino candidate. (The numbers are estimated effective cases – see Appendix A. The sample contains more actual cases.) More than 90 per cent of the Anglo-Whites reported supporting a white candidate. Because African-Americans were somewhat more likely than other respondents to indicate having been involved in a primary rather than a general election, we suspect that many of the black campaigners were supporting Jesse Jackson in his bid for the Democratic presidential nomination.

Because Anglo-Whites are the overwhelming majority of the population, if citizens supported candidates on a random basis we would expect a majority of Anglo-Whites, African-Americans and Latinos to support Anglo-White candidates. Clearly, the process is not random, however: patterns of residential segregation affect the distribution of candidates, and citizens are differentially attracted to candidates who share their race or ethnicity. The bottom line is that all three groups, but especially African-Americans, supported candidates from their own group more often than we would expect on the basis of sheer probability.

¹² Our measure of organizational affiliation and the definition of a political organization bear elaboration. Respondents were given a comprehensive list of twenty kinds of voluntary organizations – unions, professional associations, service organizations, fraternal groups, block clubs, recreational organizations, political issue organizations and the like. For each category we asked whether the respondent either is a member of or, in order to capture organizational affiliations and charitable donations that derive from responses to mass mailings, has made a contribution to such an organization within the past twelve months. In addition, we asked about contributions of either time or money to institutions like hospitals, museums or schools that are not membership groups but utilize voluntary labour and depend upon contributions from citizens. For each category for which the respondent indicated an organizational affiliation, we asked a series of follow-up questions about that organization (or, if more than one, about the one with which the respondent was most involved). Among those questions was whether the organization takes stands on public issues. We consider any organization that takes stands on public issues to be political and support, even passive support, of such an organization to be a form of political activity.

¹³ The theme of protest as the tactic of the powerless is explored in Lipsky, Michael, Protest in City Politics (Chicago: Rand McNally, 1970), especially chaps 1, 6, and 7.Google Scholar

¹⁴ The numbers are estimated effective cases – see Appendix A. The sample contains more actual cases.

¹⁵ Among Latinos, citizens are more likely than non-citizens to contact public officials. Consistent with the notion that protest is an outsider activity, however, Latino non-citizens are slightly more likely than citizens to protest.

¹⁶ Studies of political participation often put the three components of socio-economic status – education, income and occupation – together in a single scale. With respect to voting, Wolfinger and Rosenstone (Who Votes?, chap. 2) demonstrate that it is education rather than occupation or income that drives electoral turnout. When it comes to various other participatory acts, however, occupation and income play a complicated role. See Verba, , Brady, , Schlozman, and Nie, , ‘Beyond SES: A Resource Model of Participation’.Google Scholar

¹⁷ In this context, it is interesting to note the differences among the three groups across the generations. Considering only those respondents who could report on the educational attainment of both their parents, 20 per cent of the Anglo-White respondents, as opposed to 28 per cent of the African-Americans and 40 per cent of the Latinos, indicated that neither of their parents had gone beyond eighth grade. In contrast, 24 per cent of the Anglo-Whites, 16 per cent of the African-Americans, and 13 per cent of the Latinos indicated that at least one of their parents had graduated from college.

¹⁸ The question about family income referred to the ‘total 1989 income before taxes of all members of [the] family living at home… includ[ing] salaries, wages, pensions, dividends, interest, and all other income’. This is clearly only an approximation of the money available to an individual for use in making political contributions. A fuller account would require information about such matters as household expenses and debts, as well as control over household finances. Elsewhere we shall consider the relative importance of personal wages as opposed to family income.

¹⁹ The items about allocation of time asked respondents about a typical day. We would have been able to generate more precise data if we had asked respondents to keep time budgets. However, this would have been too complicated and costly in a survey designed to cover a wider range of concerns. In fact, the results based on our approximations accord very well with the results contained in the literature on time use.

Time has some properties that distinguish it from money as a resource. For example, it is finite in its upper limit and more equally distributed across individuals. In addition, it cannot be banked for future use. For a discussion of time as a resource for politics and relevant citations, see Verba, , Brady, , Schlozman, and Nie, , ‘Beyond SES: A Resource Model of Participation’.Google Scholar

²⁰ There are differences between our Latino respondents and the remainder of the sample in terms of time use. Latinos spend more time on housework and slightly more time on working and studying. In sum, Anglo-Whites, African-Americans and Latinos respectively spent the following hours per day on average in each way: working, 9.4, 9.5, 9.6; household responsibilities, 3.9, 4.4, 5.2; studying, 0.4,0.6,0.7; sleeping, 6.8,6.4,6.7.

The data on housework are consistent with the family circumstances of the Latino respondents who are somewhat more likely to have children at home (65 per cent in comparison with 45 per cent for Anglo-Whites and 56 per cent for African-Americans).

²¹ The discussion in this paragraph is based on the analysis in Verba, , Brady, , Schlozman, and Nie, , ‘Beyond SES: A Resource Model of Participation’.Google Scholar

²² The regressions shown in Appendix B, in which the dependent variables are income and free time respectively, demonstrate that income is related to the stratification variables that distinguish in various ways the advantaged from the disadvantaged. These variables include education and job level (as measured by how many people the respondent supervises), race/ethnicity and gender. Free time is a function of such life circumstances as having a job, a working spouse, or small children at home. Note that for income, the level of the job is significant but not whether one is working or not; for free time what counts is whether one is working but the level of the job does not seem to affect the amount of time one has outside of work.

²³ Because we translated the English interview only into Spanish, we have clearly undersampled those whose English is poor but who speak a language other than Spanish.

²⁴ A majority of Latinos operate in a bilingual environment. Sixty per cent of those whose first language at home was Spanish now use at least some English at home, divided among those who use a mixture of languages (40 per cent) and those who use English exclusively (20 per cent). In addition, a quarter of the Latinos whose first home language was English now use at least some Spanish at home: most of them (22 per cent) speak a mixture of English and Spanish, and a few (3 per cent) only Spanish.

There are interesting differences with respect to language among the Latino nationality groups. The Mexican-Americans were considerably more likely to have been born in the United States than members of the other Latino groups and are somewhat more proficient in English. It is, in fact, the Cuban-Americans – who are in terms of their level of income and education the most advantaged Latino group – who are most likely to have been born abroad and least likely to be comfortable in English. Although they are more likely to be citizens of the United States than are the Mexican-Americans and Latinos who hail from other parts of the hemisphere, the Cuban-Americans are least likely to speak only English at home and least likely to have been interviewed in English. Indeed, 59 per cent of the Cuban-Americans speak only Spanish at home, in comparison with no more than a quarter of the other Latino groups.

²⁵ Discussing ‘Religion as a Political Resource’, Kenneth D. Wald mentions the extent to which ‘congregational organizations may serve as leadership training institutes for people who lack other means of exposure to organizational skills’, in Religion and Politics in the United States, 2nd edn (Washington, DC: Congressional Quarterly Press, 1992), p. 35Google Scholar. The studies he cites find (as we do) a strong relationship between attendance at church services and electoral turnout, but not between religious attendance and other forms of political activity. In his study of parish-connected, non-Latino Catholics, David C. Leege finds a relationship between parish activity and political activity and discusses the potential of parish activity for developing the kinds of skills we measure here (‘Catholics and the Civic Order: Parish Participation, Politics, and Civic Participation’, Review of Politics, 50(1988), 704–36.CrossRefGoogle Scholar

In their study of voting turnout, Strate, Parrish, Elder and Ford demonstrate the importance of ‘civic competence’. The components of their measure of civic competence (for example, attentiveness to politics and level of political information) are more directly connected to political activity than the skills we discuss here (Strate, John M., Parrish, Charles J., Elder, Charles D. and Ford, Coit III, ‘Life Span Civic Development and Voting Participation’, American Political Science Review, 83 (1989), 445–63).CrossRefGoogle Scholar

²⁶ For a review of relevant literature, see Knoke, David, ‘Associations and Interest Groups’, Annual Review of Sociology, 12 (1986), 8–9CrossRefGoogle Scholar. See also, Verba, and Nie, , Participation in America, chap. 11Google Scholar; Baumgartner, Frank R. and Walker, Jack L., ‘Survey Research and Membership in Voluntary Associations’, American Journal of Political Science, 32 (1988), 908–28CrossRefGoogle Scholar; and Erickson, Bonnie H. and Nosanchuk, T. A., ‘How an Apolitical Association Politicizes’, Canadian Review of Sociology and Anthropology, 27 (1990), 206–19.CrossRefGoogle Scholar

There are many studies of minority, especially African-American, affiliation with voluntary associations and a long-standing debate within sociology as to whether high levels of African-American organizational activity represent ‘overcompensation’. See, for example, Babchuk, Nicholas and Thompson, Ralph V., ‘The Voluntary Associations of Negroes’, American Sociological Review, 27 (1962), 647–55CrossRefGoogle Scholar; Orum, Anthony M., ‘A Reappraisal of the Social and Political Participation of Negroes’, American Journal of Sociology, 72 (1966), 32–46CrossRefGoogle ScholarPubMed; Williams, J. Allen Jr, Babchuk, Nicholas and Johnson, David R., ‘Voluntary Associations and Minority Status: A Comparative Analysis of Anglo, Black, and Mexican Americans’, American Sociological Review, 38 (1973), 637–46CrossRefGoogle Scholar; and McPherson, J. Miller, ‘Correlates of Social Participation: A Comparison of the Ethnic Community and Compensatory Theories’, Sociological Quarterly, 18 (1977), 197–208.CrossRefGoogle Scholar

²⁷ There is extensive literature on the history of the black churches. For a discussion and bibliographical references, see Wills, David W., ‘Beyond Commonality and Plurality: Persistent Racial Polarity in American Religion and Politics’, in Noll, Mark A., ed., Religion and American Politics (New York: Oxford University Press, 1990), chap. 9Google Scholar. Although some dispute the role of the black churches, arguing that their otherworldliness reduces political engagement, most agree that they are a powerful political force among blacks. See Nelsen, Hart M. and Nelsen, Anne Kusener, Black Church in the Sixties (Lexington, Ky: University of Kentucky Press, 1975), p. 1.Google Scholar

²⁸ Franklin, John Hope, From Slavery to Freedom: A History of Negro Americans, 3rd edn (New York: Vintage Books, 1969), pp. 199, 310, 404–5.Google Scholar

²⁹ Earlier we outlined the way we used a list of twenty organizational categories used to measure organizational ties. As indicated, we consider membership in or contribution to an organization to constitute affiliation. A non-political organization is one that, according to the report of the respondent, does not take stands on any public issues either locally or nationally. Because this question was asked for each of the twenty categories for which organizational affiliation was indicated, we have independent measures of involvement in political and non-political organizations.

³⁰ The group of church members is a composite of two categories. Sixty-two per cent of all respondents indicated that they belong to or are members of a church, synagogue or other religious institution in their own or a nearby community. Those who reported attending religious services more than once a month, but who did not indicate membership in a local church, were asked whether they usually attend services at the same congregation or parish. The overwhelming majority, 84 per cent – or 7 per cent of all respondents – said that they usually go to services at the same church.

³¹ On the low level of organizational involvement among Mexican-Americans, see Garcia, and Arce, , ‘Political Orientations and Behaviors of Chicanos’.Google Scholar

³² In our emphasis upon the cultivation of skills in non-political settings, we do not mean to imply that these skills are not also developed in the course of political activity. Those who take part in politics usually receive on-the-job training for future participation. We emphasize the skills acquired outside politics because of our concern with the consequences of inequalities of extra-political resources for equality of political participation.

³³ The data are presented for all members of each of the three groups and, then, for those who might be in a position to obtain the skill – i.e., those who are working, affiliated with a non-political organization or members of a local church. Considering only those who are involved focuses on the opportunities for skill development available within an institution. Looking at all respondents reflects, in addition, the effect on skill development of differences among groups in institutional involvements. The questions about skills learned in an organization were asked about the individual's ‘main’ organization – based on the one to which he or she gave the most time and money. Thus the skills are for ‘members’ in a non-political organization that they consider their main organization.

³⁴ See Farley, Reynolds, Blacks and Whites: Narrowing the Gap? (Cambridge, Mass.: Harvard University Press, 1984), pp. 46–50.CrossRefGoogle Scholar

³⁵ Considering only those organizations – for example, the United Negro College Fund or Jack and Jill – that have some kind of racial focus, we find an even higher level of skill opportunities for African-Americans. Although the number of cases (17) is small, 75 per cent of the blacks who belong to a race-related organization and who do not have a high school diploma report an opportunity to practise a skill. The data on the role of organizations in connection with the development of skills within the African-American population is consistent with earlier findings of Matthews and Prothro. See Matthews, Donald R. and Prothro, James W., Negroes and the New Southern Politics (New York: Harcourt, Brace and World, 1966).Google Scholar

³⁶ Considering separately some of the specific skill opportunities about which we asked amplifies our understanding of the role that black churches can play in providing skills. Among African-Americans having no high school diploma, 30 per cent reported planning a meeting in church, in contrast to 1 per cent who indicated doing so on the job; 27 per cent reported making a public presentation in church, as opposed to 4 per cent who indicated doing so at work. The exception is writing a letter: only 4 per cent said that they had written a letter in church in comparison with 9 per cent who reported writing a letter on the job.

³⁷ In our sample, 25 per cent of the Latino respondents are Protestant, and 66 per cent are Catholic; 85 per cent of the African-American are Protestant, and 7 per cent are Catholic; 62 per cent of the Anglo-White respondents are Protestant, and 26 per cent are Catholic.

³⁸ The differences between Protestant and Catholic congregations were probably even more pronounced a generation ago. Although Catholic parishes remain, on average, substantially larger than Protestant congregations, among the important consequences of Vatican II has been increased lay participation both in the liturgy and in parish governance through parish councils. The effects of Vatican II have, presumably, been reinforced by enhanced educational levels among Catholic parishioners and the decrease in priestly vocations. See Gremillion, Joseph and Leege, David C., ‘Post-Vatican II Parish Life in the United States: Review and Preview’, Notre Dame Study of Catholic Parish Life, Report No. 15, 1989Google Scholar. Whether the unambiguous distinction between church-going Protestants and Catholics with respect to the number of hours they spend on other activities associated with their churches will disappear in the future we cannot speculate.

³⁹ On our short screener, we asked about neither opportunities to exercise skills nor membership in a local church. Hence, we report the data for regular church attenders (those who attend more than once a month).

⁴⁰ Our findings about the absence of church-based skill opportunities for Latino Catholics have met with scepticism from some readers who consider Latino Catholic churches in America to be important centres of political activity. They have suggested that the explanation for the apparent contradiction between their impression and our data lies in the fact that many Latinos report themselves as Catholics, but are only nominally Catholic and rarely attend church. However, data taken from our screener survey suggest that Latino Catholics are not less likely to attend church than are African-American or Anglo-White Catholics. The percentage of Anglo-Whites, African-Americans and Latinos respectively attending church more than once a month were 53 per cent, 68 per cent and 68 per cent for Protestants; and 59 per cent, 54 per cent and 63 per cent for Catholics.

⁴¹ It has been suggested to us that the distribution between Catholics and Protestants is too crude and that we should distinguish conservative from mainline Protestants. However, this distinction – a useful one for many purposes – does not capture the dimension critical for our concerns. Mainline denominations differ substantially in the extent to which they provide opportunities to learn civic skills in church with, for example, Episcopal churches providing relatively few and Congregational churches providing many more.

Unfortunately, the dictates of space prevent us from engaging in the extended analysis of the specific institutional difference between Catholic and Protestant churches that leads to the difference in skill opportunities. The main factor, however, appears to be, not the characteristics of church members, but the nature of church governance, in particular, the control over local church affairs that exists in many Protestant denominations. Comparing Protestant denominations that differ in terms of their internal governance – the degree to which congregants rather than a denominational hierarchy have control over the congregation – we find that hierarchical denominations provide fewer skill opportunities than do denominations with more congregational self rule. The fact that it is the nature of church governance that affects skill opportunities provides another way in which African-Americans are advantaged by their church involvement. Not only are African-Americans more likely than Anglo-Whites or, especially, Latinos to be Protestant, but they are also more likely to belong to the particular Protestant denominations in which church members have opportunities to exercise civic skills.

Data on this point and further analyses of these religious differences will appear in later publications from this research project.

⁴² We should note that we do not know how many of these contacts were about routine matters pertaining to the workplace or the administration of an organizatipn or church. The requests to vote or take other political action are more unambiguous as acts of political mobilization.

⁴³ The absence of a Latino deficit when it comes to exposure to political stimuli in church may provide a clue to the source of the impression that the Catholic churches attended by Latinos are highly politicized. We should also note we did not ask about the subject matter of the political communications from the pulpit or the political meetings attended in churches. Another possible explanation for the impression that the Latino Catholic churches are politicized is that the political cues arising in them may be distinctive in terms of their substantive content.

⁴⁴ The measure of being asked to be active is whether the respondent was asked to engage in any one of the activities listed in Table 4. The measure of exposure to political messages is whether the respondent was exposed to either of the stimuli listed on Table 5.

⁴⁵ The number of African-American Catholics is relatively small – twenty-one effective cases – making the estimate uncertain.

⁴⁶ We had considered the possibility that the extent of exposure to political stimuli in church would be related to the racial composition of the congregation. In fact, the relationship between the congregational segregation and the likelihood of being exposed to political stimuli is not monotonic. In each case, the mostly black congregations are the most politicizing (see Appendix C). That the all-black churches are somewhat less politicized was surprising to us. The explanation, however, may lie in the fact that the more racially segregated the congregation, the lower the average level of the education of African-American members.

⁴⁷ For a comparative analysis showing the consequences of this particular institutional configuration for political activity, see Verba, Sidney, Nie, Norman H. and Kim, Jae-On, Participation and Political Equality (Cambridge: Cambridge University Press, 1978), chaps 6–8.Google Scholar

⁴⁸ We had considered adding a final column to the table including only those in occupations that are likely to be unionized, but realized that – in an era of unionization of public employees and professionals like nurses – the appropriate pool is no longer restricted to blue-collar workers. There is a range across occupational groupings in the proportion of union members. In our data, 25 per cent of the workers in the machinist/operative category are unionized compared with 12 per cent of the clerical workers. However, 16 per cent of the workers in the professional and technical work category are unionized.

⁴⁹ The measures of civic skills exercised in churches and unions are not completely comparable. All active church members were asked the civic skill questions. However, because we could not ask a detailed battery of items about every one of twenty categories of organizations, we asked the civic skill questions about a single organization only. Therefore, only union activists who are members of a single organization or (if involved in more than one organization) who designated the union as their most important organization were asked the skill questions about their union activity.

Fortunately, we have good reason to surmise that the number of skills reported by such members is not different from the number that would have been reported by those union members who are involved in other organizations beside the union and did not choose the union as their most important organization. We compared the two groups of union members – those for whom we have skill measures and those for whom we do not – in terms of three measures of organizational activity that together are good predictors of the exercise of skills (whether they attended union meetings, were active in the union or had held an official union position). We found no difference in the union activity of the two groups. That those who say a union is their most important organization are not more active in their unions than those for whom some other organization is more important would seem to be an anomaly. However, it is important to recognize that many of the union members for whom we have skill measures are members of no other organization, while union members for whom we do not have civic skill measures are members of some other organization (that is, the one they named as most important to them instead of naming their union). Hence the union members for whom the union is not their most important organization are, on average, more active than those union members citing their union as most important.

⁵⁰ It is interesting that, even though Latinos are less likely than the other two groups to report being active or being on a board in their churches, they are still more likely to report such activity than are Latino union members.

⁵¹ We use ordinary least square (OLS) for this analysis. Elsewhere, we develop a model of how resources affect political participation, consider a number of reasons why OLS might not be the appropriate estimator for this model and develop arguments for a set of instruments that can be used in two-stage least squares estimation of the resource model. This analysis provides a firmer understanding of how people develop skills through institutional involvements. It also demonstrates that the general pattern of results obtained using OLS is similar to that obtained with much more sophisticated estimation methods. Rather than provide a reprise of our efforts to build a suitable model and to obtain acceptable instruments for two-stage least squares estimation, we simply use OLS here. See Verba, , Brady, , Schlozman, and Nie, , ‘Beyond SES: A Resource Model of Political Participation’.Google Scholar

⁵² See Verba, , Brady, , Schlozman, and Nie, , ‘Beyond SES: A Resource Model of Participation’.Google Scholar

⁵³ Because the causal ordering is less unambiguous than for the skills measures, we omitted from the analysis the other measures of resources derived from non-political institutions: the measures of requests for political activity and exposure to political stimuli. When included in the equation, these measures are also significant and the R² is higher. None of the other relationships is disturbed, however, and the skills measures retain their impact. For further discussion see Verba, , Schlozman, , Brady, and Nie, , ‘Beyond SES: A Resource Model of Participation’.Google Scholar