Try to run a MANOVA with other explanatory variables before you make any conclusions of a statistical correlation. Way too many variables here to simply break it down between post-2007 and pre-2007 (how did you pick how far back you went?). Anyway, have fun!