Video Games and the ESRB

Click to increase image sizeClick to decrease image size

ACKNOWLEDGEMENTS

The authors thank Dan Barrett, Deborah Linebarger, and Amy Bleakley for their helpful suggestions on earlier versions of this manuscript.

Notes

1. While the term “parent” is used, the survey explicitly included caregivers as well.

2. This study was conducted prior to the introduction of the E10+ category.

3. The response rate that we obtained is similar to that of other survey research. For example, the Federal Trade Commission's (2007) report on video games garnered a 35 percent response rate. To investigate the quality of the sample, we compared the demographic attributes of the survey respondents to census data for the geographic areas where the survey was conducted. In general, the demographics associated with our sample were comparable.

4. Two additional attitudinal beliefs were measured, but eliminated from the scale because they reduced Cronbach's α to .69: (1) the ratings would lead to arguments with my child and (2) using the ratings would make me feel overprotective.

5. Though the ratings displayed on the game boxes contain a textual description of the symbol, this method of asking about knowledge of the ratings is similar to the Federal Trade Commission (2007) strategy of asking respondents to list any ratings symbol of which they may be aware. Further, ratings are prominently displayed while the textual descriptions are more subtle, giving an advantage to parents aware of the meaning of the ratings. Items asking respondents to identify the ratings scaled well with the item asking respondents whether the ESRB ratings included both ratings and content descriptors, further indicating the utility of this type of knowledge question.

6. Interestingly, those with more knowledge of the ratings were more likely to perceive the system as accurate (r = .21, p<.05). It is not clear, however, whether this is a result of embarrassment over low levels of knowledge (a potential survey effect), or whether those who don't perceive the system as accurate do not feel it necessary to learn about the system (a potential parental explanation). Additional research is warranted.

7. Philadelphia parents were more concerned about negative effects than Helena parents (p<.01) and were more likely to use the ratings (p<.05).

8. Multicollinearity was a potential concern in the regression analysis; tolerance levels corresponding with norms, efficacy, and attitudes in predicting intentions were .63, .39, and .38, respectively. Analysis was repeated without efficacy and without attitudes. Results continue to support the conclusion that intentions are driven by attitudes.