I distinguished the respondents also by gender and social status.
I grouped them into a lower and an upper class, on the basis of
their occupation and educational attainment. Those with an educa-
tion not higher than a high-school diploma, and with manufacturing/
semiskilled oocupations, were classified as lower class: barbers,
bricklayers, tailors, auto mechanics, and so on.

The upper-class group comprises subjects with at least some
college education and those holding professional, white-collar occu-
pations. It is important to note that lower-class subjects are some-
times referred to in the text as "low-middle class"; similarly,
higher-class subjects are sometimes referred to as "high-middle
class." This is done for the sake of accuracy in the presentation of
specific respondents, and in order to better convey the logic of their
discourses and the nature of their ideological stance. From an
analytical standpoint, however, low-middle class and high-middle
class subjects do not comprise separate categories and should
therefore be considered as members of the lower-class and upper-
class groups, respectively.

Income was not used as the primary variable in determining the
class of the respondents. For instance, owners of small auto-parts
or food businesses were classified as members of the lower class,
regardless of their income, if their educational level defined them
as such. It was clear, in fact, that differences in economic resources
were never so extreme among subjects of comparable education and
occupational categories as to foster different social and cultural
experiences.

The distinction between middle and upper-middle class was
made primarily on the basis of the interviewees' educational level,
which, in the majority of cases, was consonant with their social
standing and their level of occupational prestige.

In the case of female respondents with no occupation--in all
classes--class differentiation was based on their educational level.
For the married women, this corresponded with their husbands'
educational level and corresponding occupational category.

In sum, respondents were divided into three major ethnic gen-
eration groupings; each of these comprises two class subgroups,
with respondents in each subgroups distinguished by gender. More

-8-