What types of cultural activities does each social group consume? How does it consume them? Although the above enables the describing of the average consumer for each cultural manifestation, to identify what barriers are relevant we considered the different cultural manifestations and their form of access together. To do this, we classified individuals according to the way in which they participate (physical/digital) and we implemented a statistical model that relates this with personal factors (age, gender, education, etc.). Also, the technique employed enabled identification of the indirect effects that digital participation can have on physical participation, and physical participation on digital. Table 1 groups together the main results, distinguishing between physical and electronic participation.

The first difference observed is age. In both cases, physical and digital participation, we find an impact that is initially positive (the higher the age, the greater the participation) and that, after reaching a maximum peak, starts to decline (the higher the age, the lower the participation). The difference lies in the turning point, which is higher in the case of traditional consumption (47-49 years) than in online consumption (29-35 years). This question could point to a broader generation and consumer habits gap, as well as be connected with the existence of the digital divide, linked to age.
A second difference is the gender gap for physical consumption that disappears with digital consumption. The higher female participation in elite culture has been profusely analysed (see, for example, Christin, 2012) and is usually attributed to early socialisation in the arts and/or differences in the labour market. The disappearance of the gender gap in online participation could hide the existence of a digital gender gap. Observing the different users of the Internet (figure 5), lower female participation is confirmed in the consumption of cultural contents and in their recreational and informational use. The only exception arises in those activities related with the generation of contents (uploading contents, participating in forums, chats and social networks).

In third place, level of education has a positive impact on participation in both physical and digital media. Education can reflect both the set of skills necessary for the correct interpretation and enjoyment of certain manifestations of high culture, and the socioeconomic level.
In fourth place, the situation with respect to the labour market is relevant. It has already been pointed out how unemployment, linked to restrictions affecting the resources of individuals, increases digital and reduces physical consumption. In the same direction, having some kind of disability affects consumption, while being a student increases consumption in both spheres.
In fifth place, the number of people in the household reduces the possibility of participating in both spheres. This could be related either with reduced availability of time, as in the case of parents with children under their care, or with restrictions of income, as in the cases of adult individuals living with their parents.
In sixth place, the spatial distribution of the offering is important for explaining physical consumption. The size of the municipality of residence determines the probability of attendance. Regional differences in consumption are also observed, so that the autonomous community of residence influences the probability of participation, an aspect that can be associated with the geographical concentration of the cultural offering. Surprisingly, these two factors are equally relevant for explaining online consumption, a result that suggests limitations in access to the Internet according to residence.
Finally, it should be pointed out that a tendency is observed towards consuming high culture that underlies all the individuals, independently of the way in which they consume it. This suggests that those 2.6% of individuals who only consume it in digital format could, if the barriers that they face lose relevance, also be physical consumers.