None of these methods of studying the attributes of innovations is an ideal means for predicting the future rate of adoption of innova- tions. But when they are used, especially in concert, they are better than nothing. And in any event, research on predicting an innovation's rate of adoption would be more valuable if data on the attributes of the innovation were gathered prior to, or concurrently with, individuals' decisions to adopt the innovation (Tornatzky and Klein, 1981, p. 5).*