Check out the NY Post listing on the "top" 50 NYC high schools here: The top 10, 11 - 20, 21 - 40, 41 - 50. The data for all 400 plus high schools, including graduation rates, average SAT scores, etc. is available here.
The Post used the DOE progress reports, plus other relevant outcome data, to calculate this list.
The Post says that the graduation rate in their listings reflects the percent of 9th graders who end up graduating after four years plus a summer; but that is not true. Many high schools discharge significant numbers of students before they even reach the 12th grade, and many if most of these students end up dropouts, but are never counted as such.
Even in the case of one of top schools on the list (Bard), the reporter notes that “about 20 students in each class transfer out.” In the case of Bard, a highly selective school, they probably enroll in other regular high schools, but for other schools, discharged students often end up in alternative high schools, GED programs or sometimes nowhere at all. The DOE used to make the discharge data by each individual school available in their graduation reports, but no longer does, ever since Jennifer Jennings and I produced a report on the rising discharge figures under Bloomberg and Klein.
[Correction! Updated reports for the classes of 2008 and 2009 do contain discharge data by school, at least for general ed students; see Appendix B at the links above, which reveal egregiously high discharge rates at many schools, with twice the number of official "dropouts" in many cases. What the city no longer seems to report on are discharge rates for D 75 and self-contained students.]
The NY Post's listing does not include any data on the growing practice of credit recovery, which is another manner in which many schools are artificially inflating their grad rates, (see this article by the same Post reporter on the phenomenon.) The DOE refuses to release any data on credit recovery, so it is impossible to know just how widespread this practice is. The class size data are also are not fully reliable; and tend to underestimate the actual size of classes in many high schools, since inclusion (CTT) classes are commonly reported as two separate classes.
I also don't trust the college-going rates in the listings; and the SAT scores don't include information as to what percent of the class actually took the SATs. Finally, the ratings may reflect more than anything else the socio-economic background of the students rather than what the schools actually bring to the table.
Nevertheless, as parents have a right to see this information, I have now posted a spreadsheet with SAT average scores for every school, for 2008 and 2009, as well the schoolwide class size averages for 2009-10 school year, as calculated by the DOE (as opposed to class size averages in each school by grade and subject, that are available here.). Neither of these files are on the DOE website, as far as I know.