To learn more about the impact of a 2017 law that required many Illinois schools to test their water for lead, the Tribune requested the results officials had sent to the Illinois Department of Public Health.
The agency turned over more than 2,100 separate spreadsheets containing information about thousands of water samples collected at public, private and charter schools throughout the state, including the amount of lead detected in parts per billion, the location of the fixture, the type of fixture and whether the sample was a first draw or was collected after a 30-second flush of the fixture.
The samples sent to the state were collected between July 2015 and September 2020, according to the data. Schools that conducted testing before the law’s passage could apply for a waiver and submit those earlier results.
Although the law required lead testing only at schools that serve children up to fifth grade and located in buildings built before January 2000, the spreadsheets from the state included information from some districts and schools that chose to test at additional facilities.
Results from the Chicago Public Schools, the largest district in Illinois, were formatted differently than other results submitted to IDPH because of Chicago’s robust water testing program, which involves collecting five samples from each fixture and, often, multiple dates for the sampling process at each school.
Because of these differences, data from the spreadsheets was organized into two databases: one for the results CPS submitted to the state and the other for results from the rest of Illinois.
A separate records request was made to the Chicago Public Schools in September for all available results beginning in 2016 from its ongoing testing program. The district provided results from several rounds of testing over a six-year period, from May 2016 to August 2022.
Using the results from CPS and the rest of the state, the Tribune created an interactive tool that allows parents or community members to explore the results at their school or district. The tables in the lookup tool include fields that summarize the results for each school, including the highest water lead level identified, whether any water lead levels exceeded 5 parts per billion, and how many samples were identified within particular lead ranges.
In reviewing the data on Chicago schools, inconsistencies emerged between the results initially provided by CPS and the Chicago results provided by the state health department, including slightly different results from the same fixtures drawn on the same dates. After reporters spoke with Chicago Public Schools officials, the district sent updated information to the Department of Public Health and the Tribune.
After identifying the discrepancies in the Chicago data, the Tribune decided to run a check on the accuracy of the statewide data. The Tribune randomly selected 10% of the public school districts that had identified elevated lead levels and asked district officials whether the Illinois Department of Public Health’s results matched their own records. A small number of districts could not answer the question because they could not find records related to their testing, and results from two schools had been incorrectly labeled as belonging to different schools, but no major issues arose that called the state data’s integrity into question.
In order to analyze the results of the lead testing, reporters had to clean up both the CPS and statewide data sets. The results were listed not by name but by school or facility ID, so the Tribune used data tables from CPS and the Illinois State Board of Education to identify the schools where the water samples were taken. For some results, the identified schools did not match the descriptions of the schools that were included in the results data; these cases were investigated and either amended or eliminated. Duplicate rows of information were removed when possible.
To examine whether the state was in possession of all testing results required by the law, the Tribune had to figure out which school buildings were built before 2000. There was no single reliable source of this information, but the Tribune drew on publicly available information from the Illinois State Board of Education and the Chicago Public Schools, including building a data set of CPS building ages from individual facility assessment reports and updating a facility age database maintained by the Illinois State Board of Education using annual statistical reports.
In the end, the Tribune identified almost 200 school districts that appeared eligible for testing but whose data was not in the information shared by the Illinois Department of Public Health. When reporters called a handful of the missing districts, school officials who responded each stated that they had tested the water for lead and were not sure why the state didn’t have their results. An IDPH spokesperson said the agency has now notified hundreds of individual schools that testing should be completed and the results submitted.
For the public schools that did submit testing, the Tribune wanted to know how many fixtures where lead levels exceeded 5 parts per billion would be found at the average student’s school. For each school, the number of students attending classes during the year the water samples were drawn was multiplied by the number of fixtures where an elevated level was found. The average was calculated by adding these numbers together statewide, then dividing that sum by the universe of students who attended the schools that were tested. Similar calculations were carried out for Black, white and Latino students.
In those calculations, the Tribune used only the first-drawn samples on the first date that sampling occurred at each school, so counts of elevated fixtures would not be inflated by any voluntary retesting results submitted to the state or if subsequent samples were not labeled in a manner indicating they were drawn from the same fixture. In addition, although the sampling methods used in Chicago differed from those used elsewhere in the state, the first-draw samples reflected a similar protocol statewide.
The Tribune also was able to use the data set obtained from the Chicago Public Schools, which included unique fixture identification numbers and multiple rounds of testing, to track how testing results at individual fixtures changed over time.
To assess the impact of the law beyond the identification of fixtures at schools with elevated water lead levels, reporters spoke with roughly 40 school districts about the mitigation efforts that followed their testing and received information from an additional seven districts via public records requests. For the hundreds of districts remaining, the Tribune sent school district officials a survey asking schools to elaborate on what steps they took following the testing.
In all, more than 200 districts shared details about what plumbing and maintenance renovations they undertook and what water lead levels they chose as a threshold for action. A summary of what steps districts reported taking following the testing is provided in the lookup tool alongside the school’s results so readers can learn how elevated water lead levels were addressed.
The 2017 law also required that day care facilities test their water for lead. For day cares, testing results are maintained by the Illinois Department of Children and Family Services. The Tribune requested testing results from DCFS in the spring of 2021. The data the agency sent included testing dates and a simple “y” or “n” code, indicating whether any of a facility’s results were above the 2 parts per billion threshold set by the state regulations.
A year after originally requesting the test results, reporters obtained an updated data set and noticed small but substantial changes that made it difficult to draw conclusions about how many day cares had found elevated lead since the law took effect.
In some instances, for example, a day care’s test results on a certain date changed from “y” to “n.” Hundreds of day cares’ results also were missing from one or both data sets, including those for several day cares that confirmed to the Tribune they had completed testing as part of their license renewal. DCFS provided the Tribune with a third data set this spring that had similar issues.
To learn more about the actual test results and the actions day cares took after finding lead, reporters filed public records requests to DCFS for the test results and lead mitigation plans submitted by facilities that had the capacity to serve at least 100 children, as well as those from a smaller number of in-home day cares and other smaller facilities. Reporters read through hundreds of pages of documents related to roughly 100 day care centers.
