Without knowing the details of the study, my first reaction to this analysis is that macro findings say very little. A micro-analysis that reveals specific demographics would provide a deeper description. Even the question itself leads to ambiguity. For example, I could choose books as my first choice and podcasts as my second. But these choices do not indicate degree - do I think books and podcasts will help me about the same in the future or do I think there will a be a big different between the two? And what does each learner interpret as "efficient" learning? And the socio-economic aspect of this analysis is not phase 2 but rather an essential part of the main analysis, again at the micro level.
Your question is a legitimate on: "Do we tend to prefer learning using the tools and methods we grew up with?" Was this question part of the study? I've seen literature that supports the notion that we tend to teach the way we were taught, so I would suspect the same goes for learning over time. Regardless, since everyone learns differently, it's hard to draw conclusions even if we know we tend to stick to the same tools and behaviors as we've done in the past.
I'd be interested in knowing how learners currently take advantage of learning affordances now (to learn another language) and how do they forecast learning affordances for the future. A third question addressing past affordances would also provide a historical perspective.
Human behavior (e.g., learning) is simply too complex to generalize quantitatively in an analysis of this type - there are simply too many concomitant variables at play. The amount of detail what is required to draw any sound conclusions would require detailed information, information that I suspect Busuu is not willing to provide. :)
Note: It's been my belief that efficient and engaging learning (and the teaching that allows this to happen) looks about the same whether being delivered f2f, through blended courses, or at a distance.