Confidential data about a student group are suppressed if the number of students enrolled in the group is 5 or fewer. For example, if we are reporting dropouts for Hispanic students in a school, then there would need to be more than 5 Hispanic students enrolled in that school.
We are concerned about both direct and indirect disclosure of confidential data about small groups, whether through single or multiple releases, and taking into account other reasonably available data. Indirect disclosure occurs when data are reported both for all students in a group (e.g. all students enrolled) and for a large subset of this group (e.g. students without disabilities) leaving only a small subset (e.g. students with disabilities) not reported. A group is considered small if the number of students in the group is 1-5 students.
Priority is given to maximizing public reporting at the school level. For example, if data are suppressed for a single school in a district, then rather than suppress data for a second school, district level data may be suppressed.
Each disaggregation factor, such as English proficiency, migrant, disability, or economic status, typically has two categories or student groups for public reporting purposes. The rules applied to protect privacy are the same for each of these two-category factors. To understand these rules consider, for example, that economic status has two categories: economically disadvantaged students and students who are not economically disadvantaged. If the number of students who are not economically disadvantaged is ten and the number of economically disadvantaged students is five, then we would not report data for either group, but we would report data for the total number of students enrolled (both groups combined). If both the number of students who are not economically disadvantaged and the number of students who are economically disadvantaged include more than 5 students then we would typically report data for both groups and for the total. All two-category factors would be handled this way on school reports, although additional topic-based rules sometimes apply.
Factors with More than Two Categories
Factors with more than two categories are handled differently from two-category factors. Race/ethnicity, for example, is a factor with more than two categories (i.e., A, B, H, I, W, missing/invalid, and beginning in 2010-11, P, and T), and special rules apply.
For race/ethnicity, school data for all categories where enrollment counts are small (5 or fewer) are combined in a category called "combined groups, small number." If the "combined groups, small number" category still does not include at least 6 students then we add in data for the "missing/invalid" category if the data for the "missing/invalid" category has not already been included in the "combined groups, small number" category. If the "combined groups, small number" category still does not include at least 6 students then we add in data for the next largest racial/ethnic group. If there is a tie, then we add in data for both of the tied groups.
Full Academic Year vs. Total Enrolled in School
For test scores, the highest priority is given to publicly reporting summaries for all students enrolled in a single school for the full prior academic year (FAY). This is because schools are held accountable for the performance of these FAY students. In some cases the difference between the number of "Students in This School" and number of "Students Enrolled FAY" will be five or fewer students. If data were reported for both groups then someone might be able to determine the performance of the small group of mobile students. In this case we report only the data for the "Students Enrolled FAY" and suppress the data for "Students in This School." It will be rare that a school has fewer than 5 FAY students enrolled in a grade. But it does happen. In that case the total enrolled FAY data and all disaggregated FAY data would be suppressed.
Privacy and District-level Reporting
Privacy rules are relatively simple at the school compared to the district level because the only consideration at the school level is the number of students in each group at that school. Note that for district level reporting both the sizes of groups in the district and in each school of the district need to be considered. Rarely does this result in suppression of district summary data (all students combined) or, for test scores, district summary FAY data. More often it results in suppression of certain data by demographic group.
District summary data (all students combined) might not include data for a school if the school data have been suppressed because the number of students enrolled is small (5 or fewer students). Exclusion of data for small schools from district totals prevents total suppression of district level data for districts with a single small school. The most common "small school" example is the multi-district charter school created under 118.40(3)(c) where students are counted as part of the sending district. For test scores, district summaries focus on FAY students. In a tested grade, the number of intradistrict mobile students and/or the number of FAY students in a single school is sometimes small (5 or fewer students). When this occurs, district FAY data may not include scores for the small group.
Generally, no results will be reported at the district level for a demographic group (e.g. limited English proficient students) if data for that group is suppressed in one and only one school. For test scores, results may be reported, but, to protect privacy, the results for that demographic group do not include data for all students in the group. The number of students included in results that are publicly reported will always equal all students in the group unless there are privacy concerns. Results should be interpreted with caution if the "number included in percents" is not equal or approximately equal to the number "enrolled" in the group.