Inference with the crct cheating Scandal
Download 23.25 Kb.
|
- Navigate this page:
- CRCT cheating details revealed
Inference with the CRCT Cheating ScandalStudents will use methods of inference to investigate which hypothetical schools may have cheated on the CRCT.I start with the AJC article then give the students information regarding the scandal and data for hypothetical schools. They will use what they know about inference for means to analyze the data.Each statistical team will create a report detailing their findings regarding alleged cheating on the CRCT. The report should contain their analysis and an explanation of the analysis written for those not familiar with statistical techniques. Class Discussion: Why are significance tests important? How can they be used in real world settings? Do you think this is something you could see yourself doing or being involved in? Do you think you gain more insight into this particular situation than someone who doesn’t understand statistics? Could you explain this to others who may be wondering how the schools were analyzed? After the students have summarized their findings and written their reports, the teacher can share the official results from the Governor’s Office of Student Achievement found at www.gaosa.org. The students may find it interesting to see the actual report. They may also be impressed at how much they are able to understand based on their present knowledge of statistics. Looking at the real report should impress upon them the importance of being able to communicate their findings with those who are not familiar with statistical techniques. CRCT cheating details revealedAJC Exclusive: Records obtained by AJC provide closer look at how far educators might go to ensure good gradesBy Kristina Torres The Atlanta Journal-Constitution 7:23 a.m. Friday, February 12, 2010 On a late June day two years ago, two DeKalb County school administrators panicked. A few dozen of their elementary school students had just finished high-stakes summer retests — exams first taken in spring but not passed. With just a glance at the answer sheets, Atherton Elementary School Principal James Berry and Assistant Principal Doretha Alexander saw they were in trouble. “We cannot not make AYP,” Alexander said. Not making AYP, or adequate yearly progress, meant not meeting a required federal benchmark. These students, all fifth-graders, also faced being held back if they did not pass. “OK,” Berry answered. He pulled a pencil from a cup on Alexander’s desk. “I want you to call the answers to me.” With that, he began to erase the students’ answers. State officials announced Wednesday that 191 schools — 10 percent of Georgia’s public elementary and middle schools — will be investigated for possible cheating on state tests. It was the second time in as many years that the state’s testing program has come under fire. The first was last year, when Berry and Alexander got swept up in a groundbreaking audit by the state that included an “erasure analysis” of student answer sheets. In the subsequent scandal, officials found tampering at three other elementary schools besides Atherton, including those in Atlanta and Fulton and Glynn counties. The state sanctioned 13 educators, banning them from its public schools for at least 90 days. None of those cases, however, was resolved as fast as Berry’s and Alexander’s. Both were the first to be investigated and both received among the harshest penalties. The state banned Alexander for a year; Berry’s ban lasts for two years, the harshest sanction the state imposed. Neither appealed. Now, their case files are public. Obtained by The Atlanta Journal-Constitution under Georgia’s Open Records Act, the files reveal details for the first time how Berry and Alexander cheated — and how cheating may have occurred in other schools. Both Berry and Alexander signed affidavits. Berry admits to erasing and replacing answers; Alexander says she called out the correct answers to Berry but did not personally alter tests. Both also initially denied that they had knowledge of improper activity during the administration of the tests at their school, according to the files. Berry’s attorney, Jackie Patterson, said Berry acted out of a misguided belief that he was helping his students. “He acknowledges he made a major mistake,” Patterson said. Neither Alexander nor her attorney, Don Samuel, returned calls for comment. Improbable gains Their overzealousness was clear from the start. In the fall of 2008, prior to the state’s investigation, the AJC published an analysis that showed improbable gains at some schools on tests taken first in the spring and then in the summer. One of those schools was Atherton, where half of the DeKalb school’s fifth-graders failed a yearly state test in the spring. When the 32 students took retests, not only did every one of them pass — 26 scored at the highest level. At the time, Berry told the paper that he knew of no problems with test security. Instead, he said he stressed to teachers how important test scores were. He said he also “pulled out every stop known to man.” Summer testing in DeKalb that year occurred on June 25-26, a Wednesday and Thursday, with makeup tests on Friday. DeKalb policy at the time allowed Berry and Alexander to keep the answer sheets at the school over the weekend, as long as they secured them on campus and turned them in by the following Monday. The case files do not make clear whether the two officials changed answers that Friday or during the weekend. The files indicate that, at some point, the two talked in Alexander’s office with the answer sheets in their possession. Berry then began to erase them, working with Alexander for about two hours to get the changes made. “I really didn’t pay much attention to the amount of erasing that he was doing,” Alexander said in her affidavit, according to the files. “I think there was a total of about 32 tests administered. He told me that he needed for 26 students to pass for us to make AYP.” Alexander said Berry did all the erasing while she called out the answers. She said she only “complied with [Berry’s] directive, of my superior.” Berry said they both erased and changed answers. They never talked about that day again, even after state officials came to the school to investigate. Denials, then confessions According to the files, both “adamantly denied” to state investigators that they cheated or knew of any irregularities. Only after the state made its findings public the following June did both confess to DeKalb school officials. No law in Georgia makes it a crime to cheat on state academic tests, although lawmakers late last week filed bills that would make it unlawful to knowingly tamper with state tests or help students or other educators cheat on them. The bills would make test cheating a misdemeanor, violators subject to fines and the loss of their pensions. Berry and Alexander were the only two educators of the 13 in the scandal to also face criminal charges by local authorities. DeKalb authorities charged both with falsifying a state document, a felony that carries a potential two- to 10-year prison term. Berry pleaded guilty to that charge in December. He was sentenced to two years’ probation and a $1,000 fine. Alexander completed 40 hours of community service at a local food bank and faces no further action. Map of GA school districts under suspicion The larger the circle, the higher the percentage of schools being flagged for suspicious erasures. The largest circle in southwest GA is Quitman County School District. It has 100% of it’s schools being flagged, but it should be noted that the district only has one school. 
With the exception of Quitman County, what do you notice about the geographical location of the schools in question? What are some possible explanations for this? Sample Data From Individual Schools Examine the data from the following schools. Explain how you can use this data along with information from the population to decide whether these schools may have cheated on the CRCT. Atlanta Public Schools
DeKalb County
Fulton County
State Summary of Wrong to Right Erasure Counts Examine the following table. It gives the mean and standard deviation of the state wrong to right erasure counts for each grade 1-8 and each CRCT (Reading, Language Arts and Math). Using the previous information about individual schools, select the appropriate values to use in your significance tests.  Download 23.25 Kb. Share with your friends:
|