Several years ago, Alan McConchie created the Pop Vs. Soda project that attempted "to plot the regional variations in the use of the terms "Pop" and "Soda" to describe carbonated soft drinks." I wonder if usage has shifted over time.
The primary source of data for this study will be submissions from readers of this web page. Obviously, this may not be a completely random sampling, but since the primary objective of the study is to map the regional distribution and not the population distribution per response, this sample should suffice. Also, since a large percentage of internet users are college students who may use dialects not local to their current place of residence, this survey asks for the respondent's "home town" and the beverage-term used by most of the population there. This data will be imperfect at best, but should be the most accurate possible without actually going into the field, and certainly the most comprehensive study of the field to date.