National AYP and Identification (NAYPI) database

NAYPI Database Overview

American Institutes for Research (AIR) (Contacts: James Taylor or Yu Zhang ) created the National AYP and Identification (NAYPI) database to facilitate analyses for two studies funded by the U.S. Department of Education, Policy and Programs Studies Service: the State Study of the Implementation of Accountability and Teacher Quality under No Child Left Behind (SSINCLB) and the National Longitudinal Study of the Implementation of No Child Left Behind (NLS-NCLB) (Contacts: Stephanie Stullich and Elizabeth Eisner).

Data

Data were collected from state education agency officials and consolidated state performance reports (CSPR) and the data were then put into a common standardized format enabling analyses across states and across the nation.The database contains nearly 90,000 public schools in 15,000 districts across 50 states and DC.

The NAYPI database contains detailed information on whether each school met each of its 37 potential AYP targets including reading proficiency, math proficiency, reading test participation, math test participation, and the other academic indicator for the “all students” group and each of eight student subgroups. The database focuses on simple “Yes, No, N/A” results on each AYP target (reading proficiency, math proficiency, reading test participation, math test participation and the other academic indicator) for all students in the school and importantly for each subgroup calculated under NCLB accountability (i.e., five racial/ethnic categories, students from low-income families, students with disabilities, students with limited English proficiency). Some data elements such as the applicability of subgroups were not available for all states (download the Availability of 2003-04 AYP data Excel sheet for detail). The database also includes a variety of demographic variables drawn from the Common Core of Data (CCD) to describe the schools such as enrollment, grades levels, minority population, and poverty variables. The database does not contain percentages or numbers of students scoring proficient (for such data see the AIR-developed National Longitudinal School-Level State Assessment Score (NLSLSAS) database at

Years

The current public release is for accountability testing years 2003-04 and 2004-05. The AYP variables represent school AYP performance in those years. However, because the identification status, supports, and sanctions that result from 2003-04 testing and from prior years’ testing, occur in the following school year, the identification variables are labeled as 2004-05 and 2005-06, respectively. There will be a subsequent release for accountability testing years 2005-06, which is currently being created using a combination of state-provided data and data from the Department of Education’s ED Facts/EDEN system. NAYPI data will be able to be linked to future years of ED Facts/EDEN AYP and identification data because that system has been refined to collect detailed AYP and identification data. Together with ED Facts/EDEN the AIR NAYPI database will create a comprehensive longitudinal database spanning 2003-04 to the current year.

Numbers

The NAYPI Database for 2003-04 contains the AYP status and identified for improvement (IFI) status resulting from 2003-04 accountability testing for 88,160 schools (including both Title I and non-Title I schools) in 50 states and the District of Columbia. The database does not include public schools in Puerto Rico (approximately 1,558 schools).When compared to the CCD, which contains approximately 96,000 schoolsin 50 states and the District of Columbia, the 2003-04 database does not includeapproximately 2,500 schools for which states reported AYP and identification as “not determined” and does not include approximately 5,500 schools because they were not included in state-provided data files. These 5,500 schools do not have uniform characteristics, but many are coded as “Other/Alternative” type schools or reported zero students enrolled according to the CCD.The database contains a smaller number of schools with valid AYP status for 2003-04 (80,691 schools) because(1) New York provided IFI data and a summary count of schools making AYP, but no school-by-school AYP data (4,904 schools missing on AYP0304), (2) some schools’ AYP status was reported as not determined, no data, or not available (1,987 schools coded n/a on AYP0304), and (3) AYP and IFI data for all schools could not be merged due to a lack of common IDs (578 schools missing on AYP0304).

The NAYPI Database for 2004-05 contains the AYP status and identified for improvement status resulting from 2003-04 accountability testing for 89,813 schools (including both Title I and non-Title I schools) in 50 states and the District of Columbia. AYP status was reported as “n/a” (not determined, no data, or not available) for 2,785 of these schools and is missing for 250 schools.

Use

The database is generally intended for use by researchers and is not “user-friendly” for the general public and we do not have funding to provide technical assistance to users. Please note that the database has not undergone the same degree of quality assurance that AIR typically conducts for a public release of data files. In that sense the database is delivered “as is” at the direction of the Department of Education to facilitate analyses of these data. Individual states are always the most authoritative, detailed, and current source for their own AYP and identification data; this database provides a simple way to look in a standardized way at the larger national picture.

There are several caveats and cautions to the database such as that some states’ data did not include all subgroup level data and that the data were collected within a certain window of time and states may have revised AYP and identification statuses after submitting “final” data to the NAYPI database. Additionally, the database cannot produce the exact state-by-state counts of

identified schools[1] and Title I schools[2] without referencing additional data sources. The database also omits some state-specific categorizations (e.g., use of separate Asian, Filipino, and Pacific Islander subgroups in CA, subject specific identification in reading or math in MA) in favor of standard national categorizations.

The database has proven useful for the analyses conducted as part of the SSI-NCLB and NLS-NCLB as well as other analyses conducted for ED including those used in the National Assessment of Title I. The database should be useful to those wanting to examine AYP and identification across several different states, perform national analyses of AYP and identification for school improvement or NCLB policy, those wanting to supplement their multi-state data collections with AYP and identification data, those designing studies and wanting to select schools with particular NCLB identification statuses such as Corrective Action or Restructuring or missed AYP due to the achievement of the students with disabilities subgroup, those needing a comprehensive and reliable source of historical data on AYP and identification under NCLB.

Documentation

The documentation provided isrudimentary and there are several aspects of the database that are not self-evident (e.g., states with unusable data for certain analyses, establishing the correct denominator, missing data). Please find an accompanying Word document that lists the variables in the file and an Excel spreadsheet entitled “AYP data availability by state” that indicates which states are currently included in the database and which data elements each state’s data contained. This spreadsheet is needed to exclude certain states from certain analyses if they lack usable data for particular data elements such as subgroups.

Citation

Please reference the source of these data as the “National AYP and Identification(NAYPI) Databasecreated by American Institutes for Research forthe State Study of the Implementation of Accountability and Teacher Quality under No Child Left Behind (SSINCLB) for U.S. Department of Education. Contacts: James r Yu .”

[1]Database shows 11,179 schools in IFI for 2004-05. However, there are six states where the total number of schools in IFI is known but the database does not contain data on which non-Title I schools are IFI. These six states are: GA, LA, OK, MO, NJ, IL Therefore, the database alone cannot produce accurate state-by-state counts of identified Title I and non-Title I schools and must be combined with reported numbers of identified schools from each state to produce accurate state-by-state counts. Given the confirmed state-by-state totals, the database as a whole has 438 too few non-Title I schools coded as identified. These non-Title I schools are currently included in the database within the 88,160 schools but they are coded ificat05 = 0 when they should be non-zero. Until we have a school by school list from each of these states though we cannot know which schools have the wrong statuses.Puerto Rico was at one time included but now it is not, it had 638 IFI schools (title I and non-title I).

[2]Title I variables are somewhat flawed so tables produced for only Title I schools will be somewhat off. There are several reasons for error in the Title I IFI numbers in certain types of tables created by just simply running the database: (1) The CCD 0304 titlei03 variable has missing data and might be considered to have an ambiguous meaning. (2) We know a certain number of schools in a state are IFI but we do not accurately know their IFI level e.g., DC and NJ. Also the database has some schools misclassified to IFI level (usually too many as IFI Level 1). This is due to non-standard state coding systems being standardized based on incomplete information. (3) An incomplete list was provided in the CSPR or state-provided file. (4) The timing of the best available number of identified schools (based on CSPR, interview, website, press release, etc.) and the list we have differs.