Most social science research about beliefs excludes atheist participant samples . . .

No, apparently I'm not being clear. "Knowing" is a word which fits inside "believing", like in a circle graph having one circle being entirely encompassed by another. "Believing" has a much broader definition, and apparently a number of people want to think of the word more strictly and less broadly than I do. Hence my inquiry about the word usage changing.
Well we have a problem with what we mean by these words. I regard them as separate states of mind with believing being inferior. Knowing requires a high degree of evidence while believing may not require any. Many times belief is nothing but accepting what one was told as a little child and that is why these religions propagate through generations and are local to cultures. psik