BLOG: A Behavioral Econometric Assessment: Explaining Differential Electoral Turnout in the UK, across Different Ethnicities

By Amarvir Singh-Bal. Durham University (Josephine Butler College), Dept. of Economics

Within academic literature three schools of thought help to explain turnout: the sociological, rational choice, and Michigan model. The sociological school in particular helps to explain patterns of differential turnout between social group membership and identity, such as age, class, and gender. Models in this school include relative deprivation, civic voluntarism (Milbrath 1965; Verba, Schlozman, and Brady 1995: 269 and Ch. 9) and the social capital model (Becker 1975; Crewe, Fox and Alt, 1977; Coleman 1988; Jackman and Miller 1998; Putnam 2000). When applying these models to British politics, one social group which has not been fully explored is ethnic minorities. This may be because they comprise only eight percent of the British electorate, and are therefore unlikely to change an election. Analysis of electoral data for ethnic minority voters in Britain across local and general elections in 2005 and 2010, show that Asians have the highest turnout, followed by black and then mixed ethnic groups (Electoral Commission, 2005; Parliament, 2014). Although there are many ways to explain differential turnout in Britain, I suggest scholars should look more closely at the role of relative deprivation in explaining patterns of turnout within ethnic minorities. This paper will focus on using an econometric assessment of relative deprivation to explain turnout patterns.

1) Introduction

Explaining low voter turnout amongst black and mixed ethnic groups is important for participatory democracy as a whole. Reduced turnout of one group is a significant problem for democratic pluralism, as a sector of the electorate is not being represented (Parliament, 2014). This may reduce incentives for governments to respond to the interest of non-voters and thus threaten a central tenet of democracy: that every citizen’s preference, irrespective of status, should count equally (ibid). Black and mixed ethnic groups may choose a different way of participating politically, as seen in riots that occurred across various cities during the summer of 2011 (Fielding and Cutts, 2008, p. 533).

Relative deprivation proposes that differences in turnout occur when people’s expectations of a standard of living, to which they believe they are entitled, are thwarted. This leads to frustration and apathy towards the institutions which should alleviate their problems. Typically, people of black and mixed ethnicity consider themselves more deprived than their white counterparts (Peach, 2006). The fear of racial harassment contributes to feelings of deprivation, rendering them less likely to vote than Asians. The relative deprivation model, therefore, helps to explain low turnout amongst the black and mixed ethnic groups.

The model of relative deprivation was introduced in Britain during the 1960s to explain class (Runciman, 1966), but its application to British politics has not since been rigorously discussed. Recent research is almost exclusively limited to the US; conclusions drawn from it cannot conclusively be applied to the British context due to the differences between race relations in the two countries. The US studies also suffered from methodological limitations (Walton, 1985, pp. 78–82), such as the small number of ethnic respondents (Orum, 1966, Mcpherson, 1977; Bobo, 1990).

This paper focuses on low patterns of turnout exclusively for black and mixed voters, not the entire British electorate. It is necessary to use empirical methods to observe if relative deprivation explains some of the differences in turnout among ethnic minorities. I examine recent empirical evidence regarding the relationship between Black and Mixed people’s perceptions of their relative deprivation and their turnout rate, compared to South Asians (hereafter, Asians). A cross-sectional dataset from the 2010 Ethnic Minority Election Study (EMES) provides a large enough sample to test whether the theory of relative deprivation helps to explain the aforementioned differences in turnout. The evidence I present rejects the US findings that relative deprivation actually mobilizes turnout; I posit that the opposite is true for elections in Britain. The effects are modest, but support my hypotheses that, due to perceptions of relative deprivation, (H1) black and (H2) mixed ethnicities are less likely to vote than Asians in British elections.

2) Theory: Relative Deprivation explains low turnout for Black and Mixed ethnicity individuals, relative to Asians, in British elections

2.1) Theoretical Background

Life experience for those of black and mixed ethnicity causes a generalized sense of relative deprivation, compared to the idealized standard reflected within a primarily white society. If comparisons to idealized standards prove unfavorable, feelings of relative deprivation lead to frustration and anger (Walker, Wong, and Kretzschmar 2002, Clarke, 2004, p.217). Members of the black and mixed community evaluate how equitably they are treated by the state, which they hold responsible for alleviating their deprivation. The gap between expectations and evaluations of their current economic and political situation negatively impacts their attitude towards the political system, reducing turnout (ibid, p. 288).

Applying Runciman’s formal definition of relative deprivation to the black and mixed community:

‘A [The B&M community] is deprived of X [social, political and economic equality] when
(i) they do not have X,
(ii) they see another person or other persons, which may include themselves at some previous or expected time, as having X (whether or not this is or will be in fact the case),

(iii) they want X,
(iv) they see it as feasible that they should have X’ (Runciman, 1966, p. 10).

Thus, the core tenets of the relative deprivation model will be examined more closely as being a perception of: (1) general deprivation; (2) economic deprivation; and (3) attribution of responsibility to those elected.

2.2) Perceived General Deprivation

General deprivation involves individuals believing that they have not received an equal share in life, or that political or social arrangements are inequitable and unjust (Clarke, 2004, p.217). Data from the EMES shows that the black and mixed community is more likely than the Asian community to experience relative deprivation at a general level, explaining their low level of electoral turnout. Those of Black and mixed ethnicity are 61.1% (Figure 1) and 62.7% more likely to feel that there is a gap between what people from their ethnicity expect and what they receive, compared to 50.9% of Asians (χ2=72.882, df=6, p<.001).3 One reason for an increase in black and mixed perceptions of relative deprivation, compared to Asians, is that they are more prone to discrimination, relative to whites, within the labor market (Cheung and Heath, 2008), with black men earning £115 per week less than white men (Cabinet, 2002; electoral commission, 2002, p. 17). Whereas 30.1% of Asians are said to have experienced discrimination between the 2005 and 2010 election cycle, 47% and 48.6% of black and mixed ethnicity had experienced the same (χ2=80.532, df=3, p<.001). Furthermore, 54.9% of black people and 55% of mixed race feel that non-white people do not have the same opportunities as white people, as they are held back by discrimination, compared to 49.3% Asians (χ2=4.758, df=6, p<.001).

2.3) Perceived Economic Deprivation

As a consequence of discrimination within the labor market, the black and mixed community is more likely to attribute experience of economic deprivation to the elected, augmenting the case for the relative deprivation model in explaining patterns of turnout across ethnic minority groups. Economic deprivation is defined as a judgement that one’s household financial condition has deteriorated (retrospectively) or is unlikely to improve (prospectively) (Evans and Anderson, 2005, p.208). Retrospective feelings of economic deprivation in the past twelve months are reported by 52% of black and 53.3% of mixed ethnicity citizens, compared to 48% of Asians (χ2=29.389, df=12, p<.082). 36.55% of black and mixeds believe that their household financial situation will worsen, compared to only 33.1% of Asians (χ2=29.389, df=12, p<.003). Being more likely to feel that their financial situation has deteriorated and will worsen, those of black and mixed ethnicity are more likely to hold the government accountable, resulting in a lack of trust in the political system. (Clarke, 2004, p. 217).

2.4) Attribution of responsibility to elected government, as a result of relative deprivation, leads to a fall in turnout

The black and mixed community is more likely to hold the government accountable for their relative deprivation, resulting in low electoral turnout. 34.3% and 28.44% of those of black and mixed ethnicity (compared to 13.5% of Asians) felt that Westminster politics treated their ethnic group unfairly (χ2=236.197, df=6, p<.001). 69% and 64.1% of black and mixed groups believe that the police can arrest them for no reason, compared to 52.2% of Asians (χ2=30.169, df=6, <.001).

The experience of black and mixed people in Britain, tainted by perceptions of relative deprivation, results in a low turnout compared to Asians. 38.7% of black and 55% of black and mixed people, compared to 24.1% of Asians (χ2=59.927, df=9, p<.001) expressed dissatisfaction with the democratic system in Britain. 31.2% and 34.6% of black and mixed people expressed mistrust of the police, compared to 17.3% of Asians (χ2=17.3%, df=6, p<.001). 50.7% and 61.6% of black and mixed people, compared to 38% of Asians, claimed to distrust politicians (χ2=59.927, df=6, p<.001), and 43.5% and 58.3% of black and mixed people, compared to 30.3% of Asians, feel resentment towards parliament (χ2=84.644, df=6 p<.001).

Black and mixed people are more likely than Asians to believe that they have not and/or will not receive their ‘just desserts’ from the political system, and are more inclined to feel frustrated and disillusioned by it (Sanders et al., 2014, p.125). Their sense of disillusionment renders them less likely than Asians to participate in elections. This leads to my hypothesis that:

Due to perceptions of relative deprivation, people of (H1) black and (H2) mixed ethnicity are less likely to vote than Asians in British elections ceteris paribus.

3) Methods

3.1) Data: The Ethnic Minority British Election Study (EMES)

The EMES has been used to test if relative deprivation theory explains the relationship between low electoral turnout of people of black (N=1127) and mixed (N=107) ethnicity compared to Asians (N=1541). People from ‘other ethnicities’ are also included (N=11).

Variable Table

3.2) Estimations

Having operationalized the variables, I now test if the relative deprivation model can be used to explain low turnout for black and mixeds through the use of a multivariate model. Before moving onto more complex multivariate analysis, it is worth testing (in bivariate form) whether there generally has been a low electoral turnout between black and mixed people and Asians across the local 2005 and 2010 general elections. If the simple bivariate relationship postulated does not appear (that black and mixed people are less likely to vote than Asians), it is unlikely that my two hypotheses are correct. Thus, I first estimate the relative electoral turnout of black and mixed people across the three elections. Next, I perform a more complicated multivariate to estimate if the relative deprivation model can be used to explain low turnout between black and mixed voters and Asians. I test all estimations first without controlling for socio-demographic variables, followed by a test with controls. To ensure comparability, both models, the bivariate and the multivariate, include the same respondents.

Electoralturnout = α + β1Black + β2Mixed + β3Otherethnic
 + β4RelativeDeprivation +β5Female
 + β6Englishlanguage
 + β7Newspaper
 + β8Internet
 + β9Knowledge
+ β10Qualifications
 + β11Employment
 + β12Age 
+ β13Income
 + ε

When testing the bivariate relationship of electoral turnout between black and mixeds and Asians, the coefficients of interest are β1. Here, I expect a negative and statistically significant coefficient, denoting that people of black and mixed ethnicity are less likely to vote compared to Asians.

Moving on to the multivariate analysis, I control for the effects of relative deprivation (β4) by adding it to the model. This is done to observe whether relative deprivation, broadly, explains the difference in turnout between those of black and mixed ethnicity and Asians. If it does, I expect the effect of low turnout amongst black and mixed people and other ethnicities to be reduced. Two outcomes are anticipated: firstly, the β1−3 coefficients become closer to 0. This means that once I control for the relative deprivation variable, I expect the negative coefficient value between black and mixed turnouts to reduce substantially closer to 0. Secondly, I expect a reduction in the statistical significance of β1−3 in regression 1 to become no longer statistically significant in regression 2. This is due to the null hypothesis only being true on occasion, which means that there is now no probable difference between black and mixed and Asian turnout when controlling for relative deprivation. If these two outcomes result, my two hypotheses will have empirical credibility. To test if this relationship occurs ceteris paribus I also add controls in the models using the socio-demographic variables expressed in β5−13.

In summary, within my bivariate estimations, I predict statistically significant and negatively correlated coefficients, showing that black and mixed people are less likely to vote in local, as well as in the 2005 and 2010 general elections. Moving on to the multivariate models, I expect the correlation coefficients β1−3 to become closer to 0 and become less statistically significant. Within all my models I first run the analysis without controls and then with socio-demographic controls.

4) Results

The empirical results provide some evidence for the two hypotheses that, due to perceptions of relative deprivation, people of (H1) black and (H2) mixed ethnicity are less likely to vote than Asians in British elections. However, these findings must not be overstated, as the variation in the correlation coefficients is not substantial after controlling for relative deprivation. Evidence of relative deprivation explaining differences in turnout is seen with the reduction of the values within the negative correlation coefficients of black and mixed turnout in elections, after controlling the relative deprivation in the second model. The hypothesis is strengthened by a fall in statistical significance within the same coefficients. From a broader perspective, this provides evidence for the view that relative deprivation does, and therefore should, explain some minor patterns of turnout amongst people of different ethnic groups.

In order to present these findings, the bivariate relationship for electoral turnout for black and mixed people will be discussed. On average, this was lower compared to Asians across local, as well as the 2005 and 2010 general elections. I then discuss the impact of relative deprivation, as an explanation, on electoral turnout amongst different ethnic groups.

Regarding the bivariate estimations in Table 1, I find there is sufficient evidence of a negative relationship between black and mixed and turnout in comparison to Asians. Table 1 presents the estimations of turnout for black, mixed and other ethnicities, whilst Table 2 presents estimations for people of mixed ethnicity. Columns (2), (6) and (10) illustrate the coefficient estimates of bivariate regressions. Columns (4), (8) and (12) present the coefficient estimates of controlled multivariate regressions. Overall, the coefficient remains stable between the bivariate and multivariate regression estimates. Thus, the discussion is largely limited to the controlled estimations in columns (4), (8), and (12) in Table 1 and 2 throughout this subsection.

Both ethnic groups are less likely to vote than Asians. For example, people of black ethnicity were 8.6%, 6.5%, and 6.6% less likely to vote in local, 2005, and 2010 general elections, respectively, compared to Asians. Interestingly, people of mixed ethnicity were 17.2%, 14.6%, and 22% respectively less likely to vote throughout the same elections. The same relationship, although not statistically significant, is seen for ‘other’ ethnicities across the elections. Nonetheless, the difference in turnout remains statistically significant in all but the 2005 election, indicating that there is some evidence to conclude that these estimations are grounded on compelling evidence. As these are average effects, I provide evidence that the theory of relative deprivation, to some extent, explains low voter turnout amongst people of black and mixed, compared to Asians, across these three elections.

4.1) Evidence that Relative Deprivation explains low turnout amongst people of black and mixed, compared to Asians

I present the effects of relative deprivation to explain differences in electoral turnout amongst black and mixed people and Asians. After controlling for the effect of relative deprivation, there is evidence to suggest that relative deprivation explains some differences in turnout amongst people of black and mixed ethnicity (Table 2). This is first apparent in the reduction in the size of the coefficient estimates (being closer to 0), and second through the reduction in statistical significance. Both these effects are consistently found across people of black and mixed and ‘other’ ethnicities. Despite the percentage differences of relative deprivation across ethnicities being small, I have nevertheless produced empirical credence to the claim that the theory of relative deprivation explains some lack of electoral turnout of black and mixed people within recent elections in Britain, compared to Asians.

4.2) Fall in the value of correlation coefficients, after controlling for relative deprivation

The analysis is restricted to the controlled effects, which supports my two hypotheses. After controlling for relative deprivation, the probability of low turnout for black people reduced from 8.6% to 6.5% in local elections, 6.5% to 5.9% in the 2005 general election, and 6.6% to 4.6% in the 2010 general election. A similar effect is also witnessed when observing the effect of relative deprivation on turnout for people of mixed ethnicity. The correlation coefficients fall from 17.2% to 14.6%, 14.6% to 13.5% and 22% to 19% respectively. relative deprivation provides an even greater explanation for a lack of turnout amongst people of ‘other ethnicities’. After controlling for relative deprivation, the correlation coefficients reduce from 36.3% to 26.4%, 35.4% to 32.1% and 42.2% to 33.2% respectively. In short, after controlling for relative deprivation, the value of the correlation coefficients moves closer to 0 across all ethnicities for all elections.

4.3) Fall in the confidence interval, after controlling for relative deprivation

Also supportive of my two hypotheses is the reduction in statistical significance of the correlation coefficients across all ethnicities, after controlling for relative deprivation. When controlling for relative deprivation, the p value for black people reduced from p<.050 in both local and general 2010 elections (Table 1) to there being no statistical significance across all elections (Table 2). For people of mixed ethnicity, the confidence interval reduced from p<.050 in the local election to their being no statistical significance. In the 2010 general election the statistical significance reduced from p<.001 to p<0.010. There was no statistical significance for people of ‘other ethnicity’ in either Table 1 or 2. Overall, the overwhelming fall in statistical significance for all correlation coefficients, after controlling for relative deprivation, illustrates that the null hypotheses are true, meaning that there is no probable difference between individuals of black and mixed ethnicity and Asian turnout when controlling for relative deprivation.

4.4) Summary

Within Section 2, I predicted that in order for the two hypotheses to be correct I must control for relative deprivation and observe, firstly, that the correlation coefficients become closer to 0 and, secondly, a drop in the degree of statistical significance. This proves to be true, except in 2005 where there was no statistical significance in the bivariate estimations to begin with. However, I must not overstate my findings. After controlling for relative deprivation, the reduced effects of low turnout were not large enough to reduce the ethnic coefficients in Table 2 all to 0. However, the fact that the negative coefficients were reduced, even to a minor extent, and that there was an overall fall in statistical significance, is enough for me to conclude the theory of relative deprivation does account for some low turnout in Britain amongst certain ethnic groups. Therefore, it should be used to explain patterns of turnout in British elections.

5) Conclusion

This paper has aimed to show why the relative deprivation model should be used to explain patterns of turnout amongst ethnic minorities in British elections. Specifically, feelings of disenfranchisement, discrimination and entitlement are the key factors in explaining why black and mixed groups vote less than Asians. This essay contributes to the turnout literature in several novel ways. Following previous studies of relative deprivation, (in 1960s Britain, and on the role of relative deprivation in mobilizing black and mixed turnout in the US) I used a cross-sectional dataset allowing me to include perceptions of relative deprivation-related questions specific to ethnic minorities, which were absent in the existing literature. Applied to Britain, relative deprivation does provide some explanation for differential turnout between ethnic minority groups. These results shed some light on a key relationship in the literature that, contrary to the US, relative deprivation actually decreases turnout. The overall effect is modest, and other models of voting may provide a more substantive explanation for voter turnout amongst ethnic minorities. It would be interesting to compare the effect of relative deprivation on turnout alongside other sociological models, such as civic voluntarism and social capital theory, providing insight into why differences of turnout occur between different ethnic minorities in British elections.

Participation in elections can be seen as a way of citizens demonstrating their engagement, but also as a mechanism through which feelings of deprivation attributed to those elected are expressed (Ezrow and Xenosakis, 2016, p.10). A percentage of black and mixed citizens feel disenfranchised and do not express their feelings of deprivation during an election, signaling a concern for democratic plurality. Ideally, the ballot should not only be engaged with by those who feel empowered by their position and status in society, but also as a safety mechanism (ibid, p.11) for instances of discontent within the disadvantaged community.

This specific study of turnout patterns across ethnic minorities has helped me to explain one area of turnout within a sea of theoretical and empirical observations. Having argued, and shown, that relative deprivation should be one explanation for different patterns of turnout, perhaps my use of the model can be used to explain more substantive differences amongst other socio-demographic groups. In conclusion, the implications of this paper provide a springboard for further academic discussion of theories of turnout, such as foregrounding the extent to which ethnicity has overridden social class in shaping political outlook and action (Saggar, 1999, p. 103). If the influence of factors such as age and education is outstripped by ethnicity, then I advocate that sociological theorists of turnout should place greater emphasis on different ethnicities as a predictor.

6) Bibliography

6.1) Non-academic Sources 

Cabinet Office (2002), Ethnic Minorities and the Labour Market, Interim Report, (London: Cabinet Office, Policy and Innovations Unit)

Parliament (2014), ’Voter engagement in the UK—Political and Constitutional Reform’, Online: https://www.publications.parliament.uk/pa/cm201415/ cm- select/cmpolcon/232/23207.html (accessed March 2017)

Purdam, J. and Fieldhouse, E. (2002), Voter Engagement among Black and Minority Ethnic Communities, London: Electoral Commission, Online: http://www.electoralcommission.org.uk/_data/assets/electoral_commission _pdf_file/0020/16094/Ethnicfinalreport_11586-6190_E_N_S_W_.pdf (accessed March 2017)

Electoral Commission, (2005), Black and Minority Ethnic Survey: Research Study Conducted for The Electoral Commission,
 Online: http://www.electoralcommission.org.uk/_data/assets/pdf_file /0019 /47260/ECblack and mixedEReportFINAL2_18810-13883 _E_N_S_W_.pdf (accessed March, 2017)

6.2) Academic Sources

Abramson, P.R. (1972), ‘Political Efficacy and Political Trust Among Black School-children: Two Explanations’ in Journal of Politics Vol 34, pp. 1243-69

Abramson, Paul R. (1983), Political Attitudes in America (San Francisco: Freeman and Co.)

Becker, G. S. (1975), Human Capital: A Theoretical and Empirical Analysis (New York, NY: National Bureau of Economic Research)

Bobo, L. and Gilliam, F. D. (1990), ‘Race, Sociopolitical Participation, and Black Empowerment’ in American Political Science Review Vol 48 (2), pp. 377-93

Clarke, H.D., Sanders, D., Stewart, M.C., Whitely, P. (2004), Political Choice in Britain (Oxford University Press)

Coleman, J.S. (1988), ‘Social Capital in the Creation of Human Capital’ in American Journal of Sociology, Vol 94 pp. 95-120

Crewe, I., Fox, R. and Alt, J. (1977), ‘Non-voting in British general elections 1966-October 1974’ in Crouch, C. (ed.), British Political Sociology Yearbook (London: Croom Helm), pp. 38-109

Downs, A. (1957), ‘An Economic Theory of Political Action in a Democracy’, in Journal of Political Economy, Vol. 65(2), pp. 135-50

Evans, G., and Anderson, R., (2005), ‘The Political Conditioning of Economic Perceptions’, in Journal of Politics, Vol. 68(1). pp. 194-207

Ezrow and Xenosakis, (2016), ‘Satisfaction with Democracy and Voter Turnout. A Temporal Perspective’, in Party Politics, Vol 22 (1), pp. 3-14

Guterbock, T. M. and London, L. (1983), ‘Race, Political Orientation, and Participation: An Empirical Test of Four Competing Theories’, American Sociological Review, Vol 48 pp. 439-53

Heath, A.F. and Cheung, S.Y. (2009), ‘Unequal Chances: Ethnic Minorities in Western Labour Markets’ in European Sociological Review, Vol 26 (4) pp. 265-67

Jackman, R.W. and Miller, R.A. (1995) ‘Voter Turnout in the Industrial Democ- racies During the 1980s’ in Comparative Political Studies, Vol 27 pp. 467-92

Long, S. (1978), ‘Personality and Political Alienation Among White and Black Youth: A Test of the Social Deprivation Theory’ in Journal of Politics Vol 40, pp. 433-57

Martin, N.S. (2016), ‘Do Ethnic Minority Candidates Mobilise Ethnic Minority Voters? Evidence from the 2010 UK General Election’ in Parliamentary Affairs Vol 69 (1), pp. 159-80

McPherson, J.M. (1977), ‘Correlates of Social Participation: A Comparison of the Ethnic Community and Compensatory Theory’ in Sociological Quarterly Vol 18 pp. 197-208

Milbrath, L. (1965), Political Participation: How and Why Do People Get Involved in Politics? (Chicago, IL: R and McNally)

Orum, A.M. (1966), ‘A Reappraisal of the Social and Political Participation of Negroes’ in American Journal of Sociology Vol 72, pp. 32-46

Peach, C. (July 2006), ‘Muslims in the 2001 census of England and Wales: gender and economic disadvantage’ in Ethnic and Racial Studies, Vol 29(4), pp. 629-55

Putnam, R. (2000), Bowling Alone: The Collapse and Revival of American Community, (New York, NY: Simon Schuster)

Rodgers, H. (1974), ‘Toward Explanation of the Political Efficacy and Political Cynicism of Black Adolescents: An Exploratory Study’ in American Journal of Political Science, Vol 18, pp. 257-82

Runciman, W.G., (1966), ‘Relative Deprivation and Social Justice: A Study of Attitudes to Social Inequality’ in Twentieth Century England (Penguin: London)

Saggar, S. and Heath, A., (1999) ‘Race: Towards a Multicultural Electorate?’, in Evans, G. and Norris, P. Critical Elections: British Parties and Voters in Long-Term Perspective (Essex: Sage Publications)

Sanders, D., Fisher, S.D., Heath, A. and Sobolewska, M. (2014), ‘The democratic engagement of Britain’s ethnic minorities’ in Ethnic and Racial Studies Vol 71 (1), pp. 120-39

Shingles, R.D. (1981), ’Black Consciousness and Political Participation: The Missing Link’ in American Political Science Review Vol 75 pp. 76-91

Taylor, M.C. (2002), ‘Fraternal Deprivation, Collective Threat and Racial Resentment’ in Walker, Iain and Smith, H.J. Relative Deprivation: Specification, Development, and Integration (Cape Town: Cambridge University Press)

Verba, S., Schlozman, K.L., and Brady, H.E. (1995), Voice and Equality: Civic Voluntarism in American Politics (Cambridge, MA: Harvard University Press)

Walker, I., Wong, N.K., and Kretzschmar, K. (2002), ‘Relative Deprivation and Attribution: From Grievance to Action’ in Walker, Iain and Smith, H.J. (eds.), Relative Deprivation: Specification, Development, and Integration (Cape Town: Cambridge University Press)

Walton, H. (1985), Invisible Politics (NewYork: State University of New York Press)

Worrel, J. (2002), Encyclopaedia of Women and Gender: Sex Similarities and Differences and the Impact of Society on Gender (Association Press: California)

7) Appendix

Figure 1

Table 1

Table 2

Advertisements