In the last few weeks, several incidents of teachers cheating on tests used to compare schools were revealed in Lancashire in England, in Washington DC and the US state of Georgia.
They add to the stream of similar incidents in both countries since school performance comparisons were introduced. They demonstrate the extreme pressure placed on teachers and principals where test results are used to compare school performance. Cheating also further reduces the reliability of school results as a measure of school and teacher quality.
Last week, three teachers at a top ranking specialist language college in Lancashire were suspended over claims they helped pupils with questions in the Year 11 General Certificate of Secondary Education (GCSE) language exams (BBC 10 August); The Times 10 August). School comparisons of GCSE exam grades are used in to rank secondary schools in England. An investigation into the claims is being conducted and it will report to the exam board later this month which will consider what action to take.
Another cheating incident was also revealed last week in a Washington DC charter school ( Washington Post 12 August). According to the report, an administrator and two teachers were dismissed and 27 4th and 6th grade students had their test scores invalidated. As a result, the school had its funding cut by $10,000 and the students’ results were counted as performing ‘Below Basic’ in the computation of the school’s Adequate Yearly Progress under the federal No Child Left Behind law.
Yet another case occurred recently in Georgia, where it was revealed that student answers on state tests had been changed in four elementary schools ( Education Week, 9 July). A state audit showed students’ answers on the math portion of the 5th grade test had been changed after the exams were completed. The audit found the altered answer sheets had up to 40 erasures, compared with the average of two per student on other answer sheets. Most of the answers were changed to make them correct.
As a result, one school principal has resigned and an assistant principal fired. Both have been charged with tampering with state documents, which is a felony. The Georgia State Board of Education tossed out the results of the four schools, so that the schools no longer meet federal standards and are not eligible for federal funding.
Cheating by schools can occur at different stages of the testing process. Pre-test cheating occurs where teachers tell student the questions or prepare them knowing what the questions are. Cheating during tests occurs where teachers directly help students with their answers. Post-test cheating occurs when teachers change answers before submitting test sheets for marking.
Education departments across the US are now spending millions of dollars in trying to monitor and deter cheating designed to bolster school results. Several states have hired test security companies to do audits of tests to check for cheating.
Even with the wide availability of security measures, some researchers are pessimistic that cheating can ever be completely eliminated. They say that schools and teachers, rightly or wrongly, feel that they are being judged on the basis of these tests and some of them cave in under the pressure to improve results. Robert Schaeffer, public education director for the National Center for Fair & Open Testing said:
The pressure on teachers and administrators to boost scores is so heavy that some people crack. When the pressure grows strong enough, people cross the ethical line. There’s more pressure to use the eraser or to fill in the empty bubble. ( Education Week, 9 July)
The incidence of teachers cheating on tests by helping students with answers or changing answers on test sheets generally seems to be low.
One sophisticated study published in the Quarterly Journal of Economics in 2003, using data from the Chicago public schools, estimated that serious cases of teacher or administrator cheating on standardized tests occur in a minimum of 4 to 5% of elementary school classrooms annually. A statistical analysis commissioned by the Texas Education Authority in 2007 found unusual patterns in test score gains in 9% of the state’s schools.
A report by the UK Qualifications and Curriculum Authority published in March this year on maladministration of tests found a small number of instances of teacher-aided cheating in the tests used to draw up league tables for 2008.
However, some testing experts say that it is very hard to determine how much test cheating is actually taking place in schools. Often strategic cheating can occur without triggering statistical and other devices used to identify potential cheating.
One test expert, Tom Haladyna, professor emeritus at Arizona State University, told the Atlanta Journal-Constitution following the recent Georgia incidents that: “It’s just the tip of the iceberg, I think. The other 80 percent is being hidden” [21 June 2009].
Whatever its extent, cheating inevitably occurs where tests are used to measure school and teacher performance. It reduces the reliability of comparisons of school results and league tables, along with measurement and sampling error and influences of factors outside the control of schools such as differences in the socio-economic and ethnic composition of schools.
Julia Gillard has failed to take account of overseas experience with reporting school results. She has failed to acknowledge the harm done to education by league tables. She has failed to understand that school results on standardised tests are unreliable as an indicator of school and teaching quality for many reasons, least of all because they can be manipulated to artificially boost results.
Cheating by schools is just one of these ways. No doubt there will be a sense of righteous outrage from government education ministers and the media when instances appear in Australia in the future under the new regime of publishing school comparisons.
However, governments are to be condemned for instituting a regime that imposes so much external pressure on schools to raise results that some schools and teachers succumb and resort to easy ways to boost results.
Trevor Cobbold