The statistical correlations between feelings of pride and the level of involvement in soccer were measured according to two dimensions of involvement. The first dimension relates to the distinction between the local–ethnic level and the general Israeli level. The second dimension relates to the distinction between immediate (physical attendance) and mediated involvement. These two dimensions produce four modes of involvement: attendance at local stadiums, attendance at the Premier League stadiums, consumption of soccer media in Arabic, and consumption of soccer media in Hebrew. Media consumption is composed of diverse practices and for this I performed a factor analysis that confirmed the existence of the above-mentioned categories (table 3.a).

The five elements of the mediated consumption in the general sphere were aggregated into one index by summation of the standardized score of each variable for each interviewee. This index has a high internal validity (α = 0.78). The components of the mediated “Arab sphere” were aggregated the same way into one index. This index has a medium internal reliability (α = 0.48).

Table 3.b introduces the correlations between involvements in soccer in the different spheres and different modes, and the pride in belonging to certain groups. The table shows 9 logistic regressions in which the dependent variable is pride (1 = the identity was chosen by the interviewer, 0 = the identity was not chosen). Namely, there is one equation for each identity.