Analyzing data from the 1982, 1985, and 1992 Surveys of Public Participation in the Arts (SPPA), this study defined arts participation as involvement in a listed activity at least once in 12 months. Respondents were considered crossover participants if they were involved in a single art form in more than one way or if they were involved in more than one art form in any manner listed in the surveys. What constituted crossover and the characteristics of multi-participants as compared to non-participants were also part of this study. Four types of statistical procedures were used. An analysis of relationships among participants in arts activities was made through the use of correlations cluster analyses and factor analyses. Six major findings indicated that: (1) significant crossover was limited to groups of activities that were related in some manner; (2) decisions to cross participate largely involved consideration of similarities among activities; (3) the average multiple arts participant was over 40 years old, female, college educated, earning more than 20 thousand dollars annually, white, and residing in or near urban area; (4) the average age of participants was rising; (5) strong relationships among groupings of activities were based on definable similarities; and (6) the primary explanatory factor in crossover participation was the ability to view the arts on television or video. These themes suggest a strategy that includes use of media outlets (especially television) as the vehicle for involving people in participation of more than one type of art. (MM)