K-12 Staff Demographic, Employment and Assignment Data

The K-12 staffing data, also known as the Registry for Educational Personnel, contain individual-level records on all education personnel for the academic years 2004-05 through 2019-20. While these data include information on employees from any educational entity in Michigan, the level of information recorded is not uniform for all employees. Private schools, for example, are able but not required to participate in the data collection. Additionally, for some positions (i.e. non-instructional) only a limited number of fields are required to be reported, regardless of the type of entity. These data are contained in nine datasets, which we categorize into the following three broad topical areas: Assignment, Employment, and Demographics; Education and Certifications; and Mobility. Note that some of these datasets could be used by researchers as a standalone product, but others should not due to differences in how and when districts record terminations and leaves of absence.

The Staff Demographics, Assignment, and Employment datasets comprise a significant portion of K-12 staffing data available through the Michigan Education Data Center. They contain information on who is employed, where they are employed, and for what function each semester. Researchers should exercise caution in using any of these datasets separately as there is some variability in how information is reported, and note that information on non-instructional employees is more limited. For example, districts may adopt different policies for recording staff members who are employed for partial semesters or for vacant positions. As a result, there may be some staffers included in the data that do not actually work in a given semester (e.g. substitute teachers).

Demographics

Highlights

Record Level: Employee-level 
Record Count: 1,021,070 observations 
Years Covered: 2004-Present 
Population Coverage: Part-time and full-time employees of educational entities in Michigan

The demographics file is a time-invariant file for all staff members included in REP, uniquely identified by the researcher personal identification code (RPIC). Note that while there should in theory be one observation per staff member, there are 2,341 RPICS with at least one duplicate observations. These duplicates are due to the fact that date of birth and gender are stored in the State’s personnel master file, while race/ethnicity is stored at the district level. Thus, if a staff member is employed by multiple districts and the districts have conflicting values for race/ethnicity, the staff member and their associated RPIC will have multiple observations in the base file. The variables included are RPIC, the old ID number, race/ethnicity, gender, and date of birth.

Researchers should be aware that in 2022, changes were made to how a staff member's race/ethnicity is aggregated when differing values reported in separate systems. Educators self-report their own race/ethnicity in MOECS, and districts report their employees’ race/ethnicity in the REP. MOECS and REP are two separate database systems. The race/ethnicity variable in the staff base file is created by first taking the self-reported race from MOECS and then, for educators who did not have race data in MOECS, they use the most recent race reported for that person in the REP. This has always been the case. What changed is how CEPI handles individuals whose race is reported as “unknown” in MOECS. Previously, CEPI would only extract race data from the REP if there was a missing value for their race in MOECS. This means that if the race was listed “unknown” (which is a non-missing value), they would not extract race data from the REP. They changed this in calendar year 2022. Now they extract race data from the REP for anyone with either a missing value or “unknown” race/ethnicity in MOECS. These changes apply to all staff members listed in the staff demographic file.

Employment

Highlights

Record Level: Employee by collection by school district 
Record Count: approx. 330,000 observations per collection 
Years Covered: 2004-Present 
Population Coverage: Part-time and full-time employees of educational entities in Michigan

The employment dataset contains information on the tenure of each staff member’s employment with a given district. Note that a staff member may work across multiple districts, and there may be conflicts in information reported across districts for the same staff member (e.g. highest education level). Researchers should be cautious in using the annual salary variable as not all districts are required to report this. The Employment data are theoretically unique at the RPIC x Collection X District level, but there are 97 RPICs with at least on duplicate record which may be due to having multiple hiring or termination dates in a given semester or multiple highest education levels.

Variables included are hire date, termination date, whether the position is funded, whether the position is filled by a permanent employee, employment status with the district, the highest educational degree earned, and annual salary (only required for Title 1).

Assignments

Highlights

Record Level: Employee by year by school district by assignment 
Record Count: 6,622,529 observations 
Years Covered: 2004-Present 
Population Coverage: Part-time and full-time employees of educational entities in Michigan

The assignment dataset contains detailed information about the job function(s) a staff member performs each semester in a given location. Data are reported per semester, but employees may have multiple rows per semester for reasons including performing multiple jobs in the same school, the same job for multiple grade levels, or the same job and grade for multiple schools and/or districts. Researchers should note that some assignment codes have changed over time, but we retain a variable with a description of the position which should be used to check for this issue. Researchers should also exercise caution when trying to use FTE or generate a total FTE as some staff members may perform the same position across multiple grades or settings.

The variables in the Assignment dataset can be categorized into variables about the staff member and about the assigned school. Note that staff members employed at the district level, and thus not associated with a particular school (e.g. district superintendent), will have missing values for the school variables. The school variables included are entity type, school code and name, district code and name, ISD code and name, and school open and close dates. The inclusion of the school, district, and ISD identifiers allow researchers to merge on additional information from their respective datasets. The staff variables include information on the staff member’s highly qualified status, effectiveness rating, FTE, assignment code and description, whether funded by a federal grant, flags to identify as a teacher or principal, type of position, funding code and description, grade(s) working with, and special settings.