s or percentages being compared, there is a 5 percent chance of getting a significant t value of 1.96 from sampling error alone. As the number of comparisons increases, the chance of making this type of error also increases.A Bonferroni adjustment was used to correct significance tests for multiple comparisons. This method adjusts the significance level for the total number of comparisons made with a particular classification variable. All the differences cited in this report are significant at the .05 level of significance after a Bonferroni adjustment.AcknowledgmentsThe authors wish to thank the following staff members of the National Center for Education Statistics for their thoughtful review of earlier drafts of the report: Marilyn McMillen, Michael Cohen, Daniel Kasprzyk, Carol Sue Fromboluti, and John Ralph. The report was reviewed by Joan McCord of Temple University and Denise Gottfredson of the University of Maryland, who also provided valuable comments. The report reflects many suggestions from these reviewers, and their assistance is greatly appreciated.Endnotes1 Students' answers to questions about beer or wine and about liquor were combined into a new variable describing the availability of alcohol. If either beer or wine and/or liquor were reported to be very easy or fairly easy to obtain at school, alcohol was coded as easily available. Otherwise, it was coded as not easily available.2 For additional information on telephone coverage issues and estimation procedures to correct for coverage biases, see J. M. Brick and J. Burke, Telephone Coverage Bias of 14- to 21-year-olds and 3- to 5-year-olds. Washington, DC: U.S. Department of Education, National Center for Education Statistics, report number NCES 92-101.ReferencesChristian Science Monitor. 1996. "Kids and Drugs." Editorial, August 29.Dryfoos, J. 1990. Adolescents at Risk: Prevalence and Prevention. New York: Oxford University.Dusek, D.E., and Girdano, D.A. 1987. Drugs:...
