Decreasing the high school dropout rate in the United States is gaining a great deal of national attention. One of the key strategies that has been identified to decrease dropout rate is to offer at-risk students alternatives to traditional school. This study contributed to our knowledge about alternative high schools for at-risk students by using data from the large-scale, nationally representative 2003-04 Schools and Staffing Public School Survey. Using an effective schools conceptual framework, the author used logistic regression analyses to compare the characteristics of regular and alternative public high schools. An overall model was created that could predict two thirds of the variance associated with the dependent variable. The author also conducted multiple regression analyses to determine which of the five constructs of school inputs, school structures, classroom structures, student support, and program rigor and relevance, were statistically significant predictors for the outcome measures, which included graduation rate, percent attendance at 2-year college, percent attendance at 4-year college, and average daily attendance at public alternative high schools for at-risk students. Several significant findings were discovered, for example, classroom structures and processes were found to be associated with graduation rate at public alternative high schools. Implications of the findings were discussed. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://www.proquest.com/en-US/products/dissertations/individuals.shtml.]