3.6. Discussion
The two regressions discussed above support Hypotheses 3 (that members of the general population will choose a target gender biased toward their own gender; that is, that target gender and own gender are related for this group) and 4 (that members of the general population will choose a target age biased toward their own age; that is, that target age and own age are related for this group), but disprove Hypotheses 5 (that qualified managers working in the marketing department are less likely to choose a target gender biased toward their own gender than the general population; that is, that target gender and own gender are less related for this group) and 6 (that qualified managers working in the marketing department are less likely to choose a target age biased toward their own age than the general population; that is, that target age and own age are less related for this group), making it clear that qualifying as a manager within a marketing department does not automatically eliminate or even lessen the incidence of self-reference bias. Self-reference bias starts across all populations at a very early age, being evident in the wider population and despite intervention by marketing educators, appears to persist in students. The strong correlation between target gender and own gender, and target age and own age for milk shown by all respondents except for the general population suggests that the presence of a moderately strong consumer stereotype attached to a product is not sufficient to override the effect of self-reference bias either for students or for managers; since the general population only showed lack of correlation between target age and own age for milk, with the result not being replicated for gender, this cannot be reliably attributed to the influence of the relevant consumer stereotype.