The research tested the hypotheses that the frequency of sex and violence was based on the gender of the primary artists, the race of the primary artists, and the style of the music videos. The videos were randomly selected and represented approximately 13 percent of the 1,500 videos released in 1994. Sixty-eight variables were coded. The data were analyzed by means of the chi-square tests. Results were mixed for each hypothesis. Although race, gender, and style could be used to predict differences in the frequency of occurrence of sex and violence in music videos, these differences were limited to certain dimensions of sexual behavior and violent activity. Of the three main variables studied, music style appeared to be the strongest predictor of the prevalence of certain elements of violence and sex. Findings also supported Baxter's earlier finding that sex and violence were neither very explicit nor the most prevalent element of music video content. Although the videos provide less cause for alarm than often thought, they do contain elements of violence, and rap language should remain a source of concern. Tables and 25 references