That statement comes in response to a report that Houston is one of 200 districts nationwide with suspicious test scores.
According to an analysis by the Atlanta Journal-Constitution, test results for some Houston children unexpectedly jumped one year and then plummeted the following year.
According to the paper, it's the same kind of fluctuation that exposed a cheating scandal in Atlanta.
HISD Superintendent Dr. Terry Grier said the district does not tolerate cheating. They even have a hotline to anonymously report incidents: 800-455-9551.
"We are not going to be naive enough to think that some cheating doesn't go on. But we're going to be very clear about it. It's unacceptable," Grier said. "If teachers are caught, we're going to dismiss them."
And they have, he said.
"We have conducted right here over 20 investigations since I've been here. And, quite frankly, we've dismissed over 21 teachers and/or administrators for test irregularities [in the past two years]," Grier said.
HISD's full response to the Atlanta Journal-Constitution inquiry follows:
The Houston Independent School District is committed to ensuring that all students make strong academic progress each year under the guidance of effective teachers and principals. One of the many ways HISD measures teacher and principal effectiveness is by analyzing their students' performance on multiple academic exams. This being the case, it is critical that HISD take strong proactive steps to ensure the integrity of the district's administration of student exams. Adults who participate in cheating are cheating their students out of a quality education.
Some of the test-security measures implemented by HISD in recent years include:
• HISD teachers no longer administer state-mandated exams to their own students. Exceptions are made in rare special cases.
• HISD established the HISD Alert Line (800.455.9551) for anyone to anonymously report suspicions of testing impropriety.
• On each campus, all testing materials are kept in locked rooms that must be certified by a central office staff member. Only the principal and campus test coordinator are permitted to possess keys to these rooms. These rooms are also monitored by video camera where feasible.
• On testing days, HISD dispatches trained test security monitors to witness campus security procedures, observe classroom testing, and look for signs of testing impropriety. These monitors do not work for the schools to which they are assigned. Every campus has at least one monitor. Larger campuses have multiple monitors.
• From 2010 to 2011, HISD received 31 reports of possible testing impropriety.
Twenty-two of the 31 cases were referred for further investigation by an independent law firm. Twenty-two of the 31 cases were referred for further investigation by an independent law firm.
In 9 of the 31 cases, cheating was confirmed to have occurred.
Twenty-one employees investigated for possible cheating have been recommended for termination, non-renewal of contract, or agreed to resign/retire during or immediately following those investigations.
In 17 of the 31 cases, investigators were unable to confirm that cheating happened.
Five of the 31 cases remain under investigation.
HISD has reviewed the results of the testing data analysis performed by the Atlanta Journal-Constitution. It is important to note that the strong achievement gains posted by students in the classrooms of the overwhelming majority of HISD teachers have never been called into question. On the rare occasion when allegations of possible cheating arise, HISD has demonstrated its commitment to investigating those allegations thoroughly and holding accountable those who are found to have participated in cheating.
HISD agrees that the kind of statistical analysis performed by the newspaper can be a useful tool for identifying classes of students whose test scores either increased or decreased more than might be normally expected. Experts caution, however, that such analyses do not provide evidence that testing improprieties occurred. Instead, the results of this sort of analysis should be used as a starting point for further research into the many factors that may have resulted in a significant swing in academic performance. Additionally, HISD has identified some shortcomings in the newspaper's methodology that warrant consideration:
•Urban school districts such as HISD have higher mobility rates than the average Texas school district. HISD's mobility rate is 20.2 percent. The state's average mobility rate is 18.2 percent.
•The percentage of HISD classes flagged is in line or even less than some of the other Texas districts that had at least 90 classes analyzed. Additionally, the correlation coefficient between the number of classes in each district and the number of classes flagged as having unusual growth or decline is 0.799. Put another way, there is a statistically significant correlation between the size of a school district and the likelihood of that school district being flagged in the analysis. Therefore:
Either teachers in classes in large districts are systematically more likely to see large fluctuations in student growth
Or
The larger the district, the more fluctuations in the student body (Houston student mobility is approx. 20.2%); in teacher composition, and in other school-based factors (e.g. leadership), such that changes in performance are more likely from one year to the next. This analysis does not take into account any of these factors.
•The analysis does not compare actual cohorts of students. For example, the makeup of a fourth-grade class one year can be completely different from the following year's fifth-grade class. Specific examples of flagged schools that were likely impacted by this bias include:
Community Services and Contemporary Learning Center: These are alternative schools with short term placements. Rusk is a school that serves a very high population of students from homeless families. The odds of the students actually being the same from one year to the next are very low. Additionally, the prediction model does not take alternative schools or schools serving high populations of at-risk students into account.
The use of a statewide prediction model does not take into account schools that serve special populations. This includes HISD's Vanguard magnet schools for gifted students, such as T.H. Rogers, which serves only Vanguard students and students with severe disabilities. Flagged schools in which 35 percent or more of the students are classified as gifted and talented include: T.H. Rogers, Roberts, Kolter, West University, Barbara Bush, Condit, Horn, Oak Forest, River Oaks, and Twain.
The list of schools flagged in the analysis include two of HISD's "overflow" schools, Gordon and Sugar Grove. These are schools that serve as crowding relief campuses, drawing a different mix of students each year than traditional neighborhood schools. Therefore, these schools' student populations change significantly from year to year.
Bilingual students can transition to English classes at various grades. Campuses may have adjusted their program model to move into English at an earlier grade. This practice may cause a substantial change as students would be testing in English for the first time and dips often occur in this situation. The flip side of that may be that schools realized they were moving students into English-language classes too soon and are now keeping them in bi-lingual classes longer, which would produce higher scores in their native language.
Some schools' student populations changed significantly as a result of closures and consolidations. Schools that were flagged but are now closed include: Allen, Chatham, Douglas, E.O. Smith, Easter, Anson Jones, McDade, Rhoads, Concord, Sands Point, Will Rogers, Fairchild, Hohl, and Stevenson.
Two of the flagged schools (Wilson and Garden Oaks) transitioned from being traditional neighborhood schools to magnet Montessori schools during the years covered by this analysis. Their student bodies changed significantly as a result.
•In some cases, the sizes of the "approximate cohorts" that are used do not match state data files and the average scale scores also do not match what was provided to the district.
In summary, HISD has demonstrated its commitment to taking strong action to ensure the integrity of the testing process. HISD has initiated 31 investigations since 2010 that have confirmed cheating occurred in nine cases. Five of the 31 investigations are ongoing. Twenty-one employees implicated in cheating investigations have been recommended for termination, or agreed to leave the district during or immediately following these investigations. Because of the many safeguards HISD has put in place in recent years, the district is confident that the vast majority of teachers and principals conduct their work with integrity. We are proud of the strong academic progress our students are making under their guidance.
