Network News

X My Profile
View More Activity


Posted at 11:46 AM ET, 09/16/2009

After Cheating Concerns, Should Students Retake Standardized Test?

By Valerie Strauss

My colleague Jay Mathews called me last week and suggested I look at his latest column , which argued that students should retake a D.C. standardized test after questions arose about the number of erasures on exam sheets.

He predicted I would disagree with him. Jay, you were right. I do disagree.

The exam in question is the 2008 D.C. Comprehensive Assessment System test. Post reporter Bill Turque reported recently that more than 40 schools were “flagged” for having erasures several times above the city average, and that D.C. schools officials were slow in looking at the problem after it was raised.

The results of those exams were closely watched as evidence that Chancellor Michelle Rhee’s reforms in the battered school system were improving student achievement. Indeed, scores did go up in math and reading.

The CTB/McGraw-Hill expert who reported on the irregularities said no conclusions should be drawn about cheating--but, of course, doubts remain.

Were the erasures a result of cheating? Do students know the material?

Jay said students should retake the test to prove that they learned the subject matter.

He likened the situation to one in 1982 at Garfield High in East Los Angeles, where 14 students were suspected of cheating on an Advanced Placement calculus exam; 10 retook the exam and passed. Mathews, after years of research, learned that at least nine of the students had cheated on the first test, yet passed the second exam with flying colors.

The result was a realization that academic expectations for inner-city kids were too low. Mathews wrote a book about the Garfield math teacher Jaime Escalante, and a film was made, “Stand and Deliver,” based on the work.

Yet I see big differences in the situations. In the 1982 incident, students were accused of cheating.

In the current D.C. case, the suggestion is that teachers were thought to have been behind the changes on the test. Why? Because in today’s education world standardized test results impact teacher evaluations and even salary. In fact, teachers in D.C. schools whose test scores rose by 20 points received a bonus.

Why should students have to retake a test that adults may have messed up--especially if conditions under which the students take them have not changed?

Any system that pays teachers for improved student test scores is a bad idea. Diane Ravitch, in her "Bridging Differences" blog with Deborah Meier, , provides good reasons, including:

--Creating an incentive for teachers to narrow what the teach to only what is tested: reading and math
--Encouraging cheating and effort to keep low-performing students from taking the test.

Another reason gets to the heart of a related question about whether kids should retake the test: Does a good standardized test score guarantee that a student has learned the material?

The answer, researchers say, is no, for many reasons.

Though the D.C. CAS test is untimed, many standardized tests are given with time limits, and those inherently favor kids who can work fast.

Test scores, too, are subject to all kinds of errors--measurement, statistical--as well as conditions that can affect a student but are never taken into account: sickness, bad lighting and a host of others.

Some readers don’t agree with me. On Monday The Answer Sheet asked whether teacher pay should be linked to test scores. Here are some of the responses:

From jbb3rd: “... A measure that takes into account improvement is tested levels could be devised for both the individual teachers and the school administrators as well. Student achievement levels depend on more than just good teaching, but teachers should be held accountable.”

Patrick Ewing: Among the many problems with teaching is an objective way to measure the quality of instruction. Relying solely on student scores with no mitigating factor creates perverse incentives for teachers to seek out high performing schools with high performing students.
... Jumps in test scores are not solely a product of quality instruction. Tutoring, remedial course work, individual attention from support specialists, and a variety of other factors may all help increase a student’s test score.
Linking job performance to pay is a logical step. Measuring teacher performance is not as easy as test scores though. Along with student improvement, peer evaluations, principle evaluations, and conforming to industry recognized best practices should all be taken into consideration when determining a teacher’s salary.
Further, teachers cannot be thrown to the wolves. Adequate support and training for those teachers who are underperforming, but have the desire to succeed must be provided. As it stands now, too many conferences are a waste, and teacher work days are for administrative tasks, not improving instruction.

aed3: Linking pay to performance is a good idea in theory. The real problem is how to evaluate the performance? There are many issues at stake, but one issue to think about is that the results of good teaching are often not evident for many years. Sometimes it’s the quiet, steady, unobtrusive kindergarten or first grade teacher who sets a student up for a lifetime of success.

But it is difficult to put great stock in standardized test scores when you know about the inherent limitations and the conditions under which they are given.

Read here a view from a D.C. teacher about giving the CAS to a special education student, and some responses.

And following are excerpts of a letter received by The Answer Sheet from Ken and Lisa McCallister of in Colorado, where standardized testing fever is running high as well:

Lisa and I just got home from curriculum night at our youngest daughter’s elementary school here in Littleton. The entire night from the principals "welcome" to the "grade specific orientation" for parents contained nothing but information related to CSAP [Colorado State Assessment Program test] The principal touched very lightly on safety issues related to hug and drop zones and then spent the remainder of his half hour on CSAP and the "growth movement" and ability to click on any student and track their progress over the years. There was no time allowed for parent questions to the principal. This "growth movement" tracking information is used to help "correct deficiencies" in the students’ learning progress.
The principal touted the large improvements in the CSAP scores for our school and pressed the parents to encourage their students to continue to do better each year in the "competition against" other area schools. Two of the other area schools close by that he wants to compete with are in the Ken Caryl Valley which is a very affluent area where money is not any sort of problem or second thought and "status" is very important to these parents. The other is another elementary school that pretty much "mirrors" our school, but, received better overall scores.

Apparently, this year will contain more "practice" for CSAP coming in many different forms to "coach" the students in order to be prepared for the actual CSAP testing. The principal as well as the teachers all acknowledged that there is "considerable anxiety" to the students involved with CSAP practice and testing. They encouraged all of us parents to help ease the anxiety and encourage them to do their very best.....

Our teacher knows that we opted our children out of CSAP and will continue to do so.

By Valerie Strauss  | September 16, 2009; 11:46 AM ET
Categories:  D.C. Schools, Standardized Tests  | Tags:  D.C. Schools, Diane Ravitch, Michelle Rhee, Standardized Tests  
Save & Share:  Send E-mail   Facebook   Twitter   Digg   Yahoo Buzz   Del.icio.us   StumbleUpon   Technorati   Google Buzz   Previous: SPOTLIGHT: How Unfair Are Grades?
Next: Sniper's Execution Set;Talking to Your Kids About Capital Punishment

Comments

Will the Post education page print the names of area National Merit semifinalists announced today?

Posted by: fmjk | September 16, 2009 2:40 PM | Report abuse

Another reason it would be useless to retest the students is that, sadly, they often don't remember a lot of the material on the test. They are prepped to respond to the multiple choice format at a given point in time using certain strategies to calculate what the best answer is, but they haven't necessarily learned it well enough to take the same test months or years later. They often cannot relate the material to anything outside the multiple choice prep work.

Posted by: aed3 | September 16, 2009 3:35 PM | Report abuse

Evaluation can be part of weekly, even daily, classroom protocol. It certainly is in selective private schools. From the simple, "What did we learn today?" at the end of a class to correcting & returning homework, from quizzes to a day reserved as "test day" each week, the best teachers I've known have always built in assessments. They were also fearless in the face of standardized exams.
There are some standardized exam issues I do challenge. Filling in bubble sheets is a skill that doesn't transfer to anything else in life. Copying from a chalkboard to a marble copybook was more valuable. No school system should buy the cheapest, most easily corrected exam--unless that test can be shown to favor critical thinkers and demonstrate that it does not place an undue burden on special education students.

Posted by: daisyallen | September 16, 2009 4:03 PM | Report abuse

Good article, and it helps for to hear a different perspective on the issue. However, I have two main issues:

1. How do you know it was teachers? I agree that it wasn't kids, but it could have been other school adults (administrators, etc.) I don't think it is fair to single out teachers unless there is some evidence you have that we don't. It would have been helpful to say "school adult personnel" or something related to that. NOT just teachers.

2. Why can't you or Jay Matthew's link to Bill Turque's last article about the erasures???? Why can't someone at the Washington post comment on the fact that no study or report was done??? For me, that is a far larger issue. Why did Rhee and DCPS refuse to have a CTB study done and a followup report by a second part (AIR) done when, in Bill Turque's first two articles they suggested they did?

Also, your very carefully worded fifth paragraph with "no conclusions should be drawn about cheating" is misleading. There really wasn't a report or study done. A more straightforward Bill Turque reported "He [Nickles] says the actual erasure analysis was more like an erasure conversation," in the phantom article that everyone refuses to reference. http://voices.washingtonpost.com/dc/2009/09/was_erasure_analysis_erased.html

Further, your 3rd paragraph is also misleading. Yes, Bill Turque did report that DC was slow to look into it. However, in his 3rd article on his blog (which I linked to above) Mr. Turque discovered that it really wasn't ever looked into despite the repeated requests from Deborah Gist. Instead, it seems that there was a brief "erasure conversation" with DCPS and a person at CTB, and then that was the end of it. So, again, did DCPS look into it, did they investigate possible cheating at the schools at all? No, Mr. Turque reported "In an interview Thursday, Rhee said: "Given that the people who actually developed the test said that it was inconclusive, we just didn't think it was necessary" to investigate possible cheating."

How about some clarity and some justified finger pointing? A "conversation" with a person at CTB suggests that it might be inconclusive (although no study or report was done), AND since when does "inconclusive" mean that nothing happened. Doesn’t inconclusive mean we don't know for sure if they were innocent or guilty? I would think that inconclusive should warrant some investigation at the local schools as opposed to sweeping in under the carpet.

Posted by: mfalcon | September 16, 2009 9:57 PM | Report abuse

mfalcon--You are right on target in questioning the assumption that teachers should be suspected of cheating. In most districts, security measures make it virtually impossible for teachers to cheat. But once the tests have been turned in to the school administration, there is nothing to keep and principals from tampering, and they have more at stake than teachers, both professionally and monetarily. Most of the cheating scandals in Texas resulted from tampering by school administration, and erasures played a role in identifying the problem.

Posted by: aed3 | September 16, 2009 11:04 PM | Report abuse

The comments to this entry are closed.

 
 
RSS Feed
Subscribe to The Post

© 2010 The Washington Post Company