What evidence does Tannen offer for the rapport-/report-talk distinction?

Question

Regarding gender differences in conversation style, Tannen draws bold conclusions which seem to get at what men and women want out of conversation, how they approach the world, and so on. For example:

I now see that my husband was simply approaching the world as many men do: as a place where people try to achieve and maintain status. I, on the other hand, was approaching the world as many women do: as a network of connections seeking support and consensus. (Tannen)

I often see the conclusions talked up with little evidence discussed. As far as evidence goes, when talking about Tannen's views in their respective textbooks, Trudgill brings up frequency of use of indirectness and of interruptions and Iwata et al. bring up frequency of jokes and of taking the floor in conversation. Surely there must be more evidence to back up the bold conclusions.

Question: What empirical evidence does Tannen offer for her conclusions regarding men and women's different approaches to conversation and the world?

Answer 1

In Gender and Discourse (1996), Tannen states:

The theoretical and methodological approach found here derives from the work of Robin Lakoff and John Gumperz, who were my teachers at the University of California, Berkeley. It was Lakoff (see especially Lakoff 1975, 1979,1990) who introduced me to the concept she calls communicative style ... and the notion that misunderstandings can arise in conversation, both cross-cultural and cross gender, because of systematic differences in communicative style.

Lenora Limm, a member of the review committee the University of California, reviewed Lakoff 1975, and states:

Methods. Lakoff states in no uncertain terms that her claims about women's language are based on data "gathered mainly by introspection: I have examined my own speech and that of my acquaintances, and have used my own intuitions in analyzing it" (p. 4).

Take that as you will. If you wish to dig further, Tannen gives many references at the end of her work.