THE MESSAGES REPLICATION FACTOR - METHODS TAILORED TO MESSAGES AS OBJECTS OF STUDY

In research on effects of message variables, it is generally necessary to examine responses to actual messages that represent, embody, or in stantiate the values of the variable of interest. Researchers have lat ely become attentive to problems of confounding in the use of individu al concrete messages to represent abstract theoretical contrasts, and replicated treatment comparisons are increasingly common in communicat ion research. How to treat the replications factor in the statistical analysis remains controversial. Whether to treat replication factors a s fixed or as random hinges on what is assumed about the relationship between abstract treatment contrasts and their concrete material imple mentations. We argue that reflection on this relationship justifies a general policy of treating replications as random. Two circumstances i n which fixed-effects analyses might seem attractive (the case of matc hed-message designs and the case of experimental manipulations occurri ng outside of messages) are considered, but it is concluded that these situations also require random-effects analyses.