Abstract
Messages pose fundamental challenges and opportunities for empirical communication research. To address these challenges and opportunities, we distinguish between message variability (the defined and operationalized features of messages in a given study) and message heterogeneity (all message features that are undefined and unmeasured in a given study), and suggest approaches to defining and operationalizing message variability. We also identify alternative message sampling, selection, and research design and analysis strategies responsive to issues of message variability and heterogeneity in experimental and survey research. We conclude with recommendations intended to advance the study of messages in communication research.