{ "id": "2001-julyaugust-teacher-student-exodus", "slug": "2001-julyaugust-teacher-student-exodus", "url": "https://dwillis.github.io/uplink/article/2001-julyaugust-teacher-student-exodus.html", "year": 2001, "month": "July/August", "monthNum": 7, "headline": "Teacher, student exodus", "kicker": "SCHOOLS", "author_name": "David Knox", "author_title": "", "affiliation": "Akron Beacon Journal", "summary": "A reporter at the Akron Beacon Journal used computer-assisted reporting and data analysis to investigate why one school cluster dramatically underperformed others, challenging assumptions that poverty alone explained the gap and focusing instead on teacher transfer policies within school administrators' control.", "keywords": [ "education", "data", "schools" ], "topics": [ "Education", "Investigative Methods", "Demographics & Census" ], "technologies": [ "Excel" ], "full_text": "Computer-assisted reporting is no different than other journalism in an important way: Often the best stories start with a reporter's gut instinct. When Reginald Fields took over the education beat at the Akron Beacon Journal in June 1998, it didn't take him long to learn that the city's public schools weren't equal in quality. Among the district's eight clusters - each composed of a high school and its feeder middle and elementary schools - Buchtel High School was considered by parents, teachers and students far and away the worst.\n\nThe evidence\n\nEvidence backing that poor reputation was easy to find. During the final grading period of the 1999-2000 school year, 36 percent of the students at Buchtel earned a D letter grade - or worse.\n\nComparisons to other schools were equally dismal. Scanning a sorted listing of Ohio's proficiency test scores found almost all the Buchtel cluster schools near the bottom.\n\nBut grades and test scores didn't explain why one group of schools performed so poorly.\n\nFields soon came to question the conventional answer - that Buchtel's students were too economically disadvantaged (read, too black) - to do better.\n\nTrue, Buchtel schools were by far the most segregated in the city - 97 percent black compared to a district average of 47 percent. But the neighborhoods where they were located were much more racially diverse. Furthermore, visits to the Buchtel schools took Fields down streets lined with many tidy, attractive homes near well-kept parks - hardly a portrait of urban blight.\n\nFields also heard a steady drumbeat of complaints from parents and students about the low experience of the teachers at the Buchtel schools. He wondered whether the school district's liberal policy toward teacher transfers - by contract, teachers can ask to move after a single year of service - might contribute toward Buchtel's poor academic performance.\n\n\"Destined to Fail,\" a six-month project sparked by Fields' skepticism, provided a challenging opportunity to use multiple analytical tools and strategies to go behind the issue of poverty and search for other factors under the control of school administrators that can affect educational quality.\n\nThe Ohio Department of Education publishes reams of readily available, easily imported data - almost all useless for the Buchtel project.\n\nAs always, the first task was getting the data.\n\nThe Ohio Department of Education publishes reams of readily available, easily imported data - almost all useless for the Buchtel project because none of it was at the cluster level. With the exception of proficiency tests information, which included the number of students taking and passing each test in each school, all the other \"building-level\" data came in the form of percentages and averages. Those numbers couldn't be brought up to the cluster level because their denominators - the number of students and teachers in each school - varied widely.\n\nWhat was needed was student- and teacher-level data that could be added together, and then calculated for each cluster.\n\nThat information was available from the Akron Public Schools - but mainly as paper records.\n\nCalculating poverty\n\nFor the issue on student poverty, that wasn't a problem. All that was needed to calculate the percent of low-income students was the enrollment and number of students eligible for free or reduced lunch for each of Akron's 48 elementary and middle schools. (High schools were left out because of the tendency of older teen-agers not to apply for free and reduced-cost lunches out of embarrassment. Two other schools were excluded because they are magnet schools that draw from all parts of the city.)\n\nMeasuring teacher inexperience took a little more work. The numbers of years of service for each of the 2,023 teachers in the schools were input into a Microsoft Excel spreadsheet and a percentage of those with less than six years calculated for each building and cluster.\n\nThe results showed that nearly half the 285 teachers in the Buchtel cluster's eight schools had less than six years experience. The cluster average was 9.7 years of experience, compared to the district average of about 14.4 years.\n\nSPSS, a statistical analysis program, was used to gauge the possible impact of teacher inexperience on student performance. A multiple regression analysis found that low teacher experience and poverty together could explain as much as 54 percent of the difference in the proficiency test results among Akron's elementary and middle schools. The probability of the relationship being a coincidence is less than 1 in 1,000. A limitation of regression analysis is that, while it shows correlations, it doesn't prove cause-and-effect - a point stressed by several researchers at two local universities who were asked to review the data. Teacher inexperience may contribute to low test scores, but a lack of experienced teachers also could reflect an exodus of veterans who choose to transfer to better performing schools.\n\nThat scenario would be more likely if poverty is considered an invariable factor - beyond the influence of school policy. That's because a single-variable regression analysis showed poverty alone could account for 48.7 percent of the difference in the test scores, compared to 43.1 percent for teacher inexperience considered as the sole independent variable.\n\nIn other words, the conventional wisdom still works: poverty could be the dominant factor in explaining low test scores.\n\nBut a key finding of the Buchtel data analysis was that the school district's \"open enrollment\" policy, which encourages students to leave their neighborhood schools in search of a better education, might be altering the family-income demographics of the schools. Such student flight might contribute to poor test scores if the students leaving a cluster are more likely to do better on the exams than the classmates they leave behind. To investigate that possibility, another methodology - geocoding - was used to roughly gauge the performance potential of students leaving and staying in each school cluster by determining the household income of their neighborhoods. The Beacon Journal obtained from the school district the addresses of more than 25,000 students - about 84 percent of all students enrolled in Akron's public schools. In accordance with Ohio law, the only addresses omitted were those of students whose parents signed forms requesting the information be withheld.\n\nMicrosoft Word's global search and replace functions were used to massage the text files into delimited columns and rows for import into a database program, Microsoft Access 97.\n\nInitially, school officials said the data, which included which school each student attended and grade level, was available only on paper because the district couldn't afford the cost of programming. Officials relented after the paper pointed out that redirecting the ASCII text output of the report program to a floppy disk entailed no additional cost.\n\nMicrosoft Word's global search and replace functions were used to massage the text files into delimited columns and rows for import into a database program, Microsoft Access 97.\n\nThe addresses were then geocoded into census block groups that included 1999 estimates of median household income, as provided by Claritas, a national marketing research service used by the Beacon Journal's advertising department.\n\nThe goal was to determine where the students would go to school in the absence of the open enrollment program by matching their addresses to the district's 317-page directory of school attendance zones.\n\nThe job was made manageable by sorting the database on street number and name - allowing a dozen or so records to be updated with their assigned schools in a single query.\n\nThe completed database had all the information needed to determine the impact of the district's open enrollment program: what school each student was assigned and the school actually attended.\n\nThe analysis found that 20 percent of the children in the Buchtel district were going to schools in other clusters, compared to a district average of 13 percent. At the high school level, one of every three Buchtel students had left the cluster.\n\nComparing the average neighborhood incomes of students staying and leaving provided evidence that the student flight might contribute to keeping test scores down. Students fleeing the Buchtel cluster were found to have neighborhood incomes more than $3,000 higher than those who remained. Only one other cluster - Central-Hower, the next poorest performing cluster - showed a net loss of income - about $1,800. And what was the reaction of the school officials to the series? They didn't challenge the findings of the statistical analysis. That changed soon after \"Destined to Fail\" was published, December 10-13. Within a week, the school board president called for a panel of parents, school administrators and community leaders to address issues raised in the series.\n\nWithin a month, the school administration presented the board with over a dozen recommendations to better the Buchtel cluster, including several proposals for upgrading the quality of teaching in the schools.\n\nDavid Knox can be reached by e-mail at Dknox@thebeaconjournal.com", "provenance": { "extracted_by": "anthropic/claude-sonnet-4-6", "extracted_at": "2026-07-05", "verified": true }, "similar": [ { "slug": "2004-july-august-new-teacher-data-boosts-narrative-project", "score": 0.242 }, { "slug": "1997-august-back-to-school", "score": 0.229 }, { "slug": "1994-march-akron-completes-yearlong-car-series", "score": 0.18 }, { "slug": "2001-julyaugust-education-tipsheets", "score": 0.169 }, { "slug": "2002-september-october-plotting-school-data", "score": 0.169 } ], "html_url": "https://dwillis.github.io/uplink/article/2001-julyaugust-teacher-student-exodus.html", "json_url": "https://dwillis.github.io/uplink/article/2001-julyaugust-teacher-student-exodus.json" }