Adolescent Judgment of Sexual Content on Television: Implications for Future Content Analysis Research

Abstract

Many studies of sexual messages in media utilize content analysis methods. At times, this research assumes that researchers and trained coders using content analysis methods and the intended audience view and interpret media content similarly. This article compares adolescents' perceptions of the presence or absence of sexual content on television to those of researchers using three different coding schemes. Results from this formative research study suggest that participants and researchers are most likely to agree with content categories assessing manifest content, and that differences exist among adolescents who view sexual messages on television. Researchers using content analysis methods to examine sexual content in media and media effects on sexual behavior should consider identifying how audience characteristics may affect interpretation of content and account for audience perspectives in content analysis study protocols when appropriate for study goals.

The project was supported by Grant No. 5R01HD044136 from the National Institute of Child Health and Human Development (NICHD). Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the NICHD. We thank Dale Kunkel and L. Monique Ward for their contribution of study materials to our project. We also thank Dale Kunkel, L. Monique Ward, and anonymous reviews from the Journal of Sex Research for their helpful comments regarding this manuscript.

Notes

Note. ✓ = at least 70% of teens checked this category and researchers checked it as well OR less than 30% of teens checked this category and neither did researchers (indicates adolescent/coder agreement).

* = at least 70% of teens checked this category but researchers did not.

— = less than 70% of teens checked this category but researchers did check it.

Blank cells indicate that 31% to 69% of teens checked this category, but researchers did not (indicates adolescent disagreement for categories not checked by researchers).

Category descriptions listed in the table have been condensed from those used in the survey.

¹Throughout the remainder of the article, we will refer to the coding schemes as Kunkel, Ward, and IM. However, it is important to note that each coding scheme was modified from its original version for ease of administration in a survey for adolescents. The coding schemes are not referenced given that they are revised from their original sources, and Kunkel, Ward, and IM are instead used as descriptors to identify a particular set of coding categories.

Note. ✓ = at least 70% of teens checked this category and researchers checked it as well OR less than 30% of teens checked this category and neither did researchers (indicates adolescent/coder agreement).

* = at least 70% of teens checked this category but researchers did not.

— = less than 70% of teens checked this category but researchers did check it.

Blank cells indicate that 31% to 69% of teens checked this category, but researchers did not (indicates adolescent disagreement for categories not checked by researchers).

Category descriptions listed in the table have been condensed from those used in the survey.

Note. ✓ = at least 70% of teens checked this category and researchers checked it as well OR less than 30% of teens checked this category and neither did researchers (indicates adolescent/coder agreement).

* = at least 70% of teens checked this category but researchers did not.

— = less than 70% of teens checked this category but researchers did check it.

Blank cells indicate that 31% to 69% of teens checked this category, but researchers did not (indicates adolescent disagreement for categories not checked by researchers).

NA = No other people were listed by researchers or respondents so this response was not applicable but was considered agreement.

^a Two marks are indicated. The first is for the male character and the second is for the female character, since outcomes were evaluated separately for each character involved in the scene.

Category descriptions listed in the table have been condensed from those used in the survey.

Note. N = 89.

^a Because three or fewer participants marked “not sure” for each program, we combined “not sure” with “never.”

²A variety of ranges are recommended for use as cutoff points for assessing which variables have achieved successful intercoder reliability in content analysis studies. We used .70 given the exploratory nature of this research and the recommendation of this value by Riffe, Lacy, and Fico (1998).

Jennifer A. Manganello, Vani R. Henderson, Nicole Trentacoste, and Suzanne Martin were with the Annenberg Public Policy Center at the University of Pennsylvania at the time of the study.