Spanish Reading/Language Arts California Spanish Assessment (CSA)

CSA Research Files

These research files contain results from the administrations of the California Spanish Assessment (CSA) and contain the same information presented in the Detailed Test Results section of this site. These files are provided to allow for more complex analyses and customized reporting of the data.

To protect student confidentiality, no scores are reported (or included in the research files) for any group with fewer than 11 students.

PLEASE NOTE:

• The "statewide" files include the data for the state, and all counties, districts, and schools. Files can also be downloaded for any single county or district.
• The "county" files include the data for the selected county and all districts and schools associated with that county.
• The "district" files include the data for the selected district and schools associated with that district. "School only" files are not available.
• Independent charter schools (direct funded) are treated as a district. Test results for these schools are included in the numbers for the state, county, and school.
• Research file formats may vary across years. Refer to the research file layout for more details.

Downloading CSA Research Files

Use of these research files requires some expertise in the handling of data and advanced data management skills. Many of the district and county research files are very large (up to 130MB) and may be too large for spreadsheet applications. Database applications such as MS Access, SAS, or SPSS are required to manage these files. The file size is indicated in parentheses and is shown only when the size is greater than one megabyte.

How to Import a Research File into Microsoft Excel (PDF; 1MB)

Research File Formats, Layouts, and Lookup Tables

Getting Accurate Results from the Research Files

Achieving accurate results when working with these research files requires an understanding of the structure and content of the two primary tables: the entities table and the test data table. The research files have many rows for each entity. There are records for each combination of grades, tests, and student groups. This means that there are hundreds to thousands of records for each entity, with an average of approximately 900 records. To correctly work with the data, you must use constraints to limit the data you are reporting. These constraints are discussed next.

Entities Table

This table is composed of the state, all counties, districts, and schools in California. Because there are both school-level and district summary records as well as county and state summary records, it is critical that in any analysis, a "Type ID" record type be selected. This will help avoid the double or triple-counting that will occur when a school count is also counted in the associated district record.

Test Data Table

This table is composed of the school, district, county, and state aggregate CSA counts and scores. To accurately analyze and report from these research files, the appropriate constraints must be applied to the following elements:

• CDS code—The research files contain summary district and county records. A district summary record will have a "school" code of "0000000." When working with the file, be sure to include the county, district, and school codes. Failure to include all three data codes will result in double-counting in any summary calculations.
• Test type—Identifying the desired test (Spanish Reading/Language Arts) will help to provide clear query results.
• Student Group ID—Each student will be included in both the "All Students" student group aggregation and each of the appropriate student group aggregations. Consequently, an individual student group must be selected to avoid duplicate counts.
• Test ID—In general, each student will take a number of tests. A specific test should be selected to avoid confusion.

Providing accurate and meaningful reports from the research files generally requires the linking of the Entities and Test Data tables. Additional efforts might include linking to the lookup tables. Working with these tables requires an understanding of relational data tables and their manipulation.