How is the DC cheating scandal like a 1990s Intel chip?

On Friday, DC's Office of the State Superintendent of Education confirmed allegations of cheating on the DC CAS test at close to a dozen schools. Officials are downplaying the significance, saying that only a small percentage of classrooms had cheating. But this misses the point; a problem that affects few can shake trust for many, as Intel Corp. found out with its Pentium processor.

Photo by mark.sze on Flickr.

Intel launched the Pentium chip in 1993. This tiny microprocessor is the CPU (central processing unit) for what were some of the widest-selling personal computers in the mid to late 1990s. Intel spent approximately $150 million in an advertising campaign around the Pentium's highly-anticipated launch, but a flaw soon came to light.

In late 1994, Thomas Nicely, a math professor at Lychburg College in Virginia, was performing complex computations on his Pentium-powered computer. Checking his computations, he found his sums vastly differed from the theoretical values he expected.

After a different CPU gave him the correct results, Nicely tracked the flaw to the new Pentium chip. Ultimately, the difference in lay with the digits in the one billionth place value, or 0.000000001, but it was a flaw and the Pentabug Problem was born.

Dr. Nicely posted his findings on the Internet with some colleagues, which quickly resulted in a wave of press coverage including interviews with CNN and other world news media outlets. When contacted, Intel responded they were aware of the problem, but it would only occur once every 27,000 years for most users.

The company offered to consider replacing chips on case-by-case basis, insisting the customer held the burden to prove that they needed a replacement. Instead of accepting accountability for the flawed product, they downplayed its magnitude, deflected blame, and hoped the media storm would blow over.

This ignited the public relations nightmare. On December 12, 1994, IBM halted all shipments of the Pentium processor causing Intel's stock to drop 5% almost immediately.

This was an inflection point for former Intel CEO Andrew Grove. He issued a public apology and bore the brunt of the replacement cost, totaling $475 million. In his apology, he offered, "We are today announcing a no-questions-asked return policy on the current version of the Pentium processor. Our previous policy was to talk with users to determine whether their needs required replacement of the processor. To some people, this policy seemed arrogant and uncaring. We apologize."

Putting in place hundreds of customer service employees to field customer requests, Grove was able to turn the corner on the Pentabug Problem, and by the end of the year, the company had locked up orders for its next generation Pentium chips for 1995. It has been able to remain the world's number one semiconductor manufacturer by revenue every single year since.

What can and should DCPS learn from the Pentabug Problem?

Chancellor Henderson said in her statement, "The results released today by the OSSE confirm what we at DCPS already know: nearly all of our teachers, more than 99 percent, are following the very strict procedures and protocols we have in place to prevent testing impropriety." The 18 confirmed violations amount to 0.6% of all the system's classrooms, which some may consider a very small amount.

But while the OSSE investigation solely covers the 2012 DC CAS test, concerns about cheating first cropped up in 2008 when Chancellor Michelle Rhee was at the helm. Last week, a memo surfaced where consultant Fay G. "Sandy" Sanford warned former Chief of Data and Accountability Erin McGoldrick that erasure data might implicate cheating in 70 schools.

Officials took action against a few principals, but kept the findings quiet. Throughout allegations that surfaced in 2008 and 2011, and even now that OSSE has confirmed there was cheating, DCPS continues to say that test integrity is of the utmost importance. However, it is unclear what has been done to confront and prevent systemic abuse. In the interim, though, hundreds of teachers have been fired due to poor results on tests.

The longer an organization avoids taking blame, the more difficult to move forward. Grove took responsibility and initiated the necessary action to overcome the Pentium's flaw. The Pentabug became the Repentium, and Intel regained the public's trust.

The DCPS cheating scandal has already elicited strong responses from reporters, elected officials, parents, and other residents. Some say the cheating shows that using tests in teacher evaluations is a failed strategy, while others say it just shows we need to crack down on cheating. Either way, DCPS needs to take responsibility for their actions and show a strong response.

Eboni-Rose Thompson works as a program specialist with Save the Children, overseeing their early childhood and school age programs in the District. She serves as the Chair of the Ward 7 Education Council and is an Advisory Neighborhood Commissioner. 


Add a comment »

The USA Today coverage from March 2011 is a good place to start.

It seems no one really wanted to do this investigation. Even when a newspaper found 2008-2010 results showed recurring, statistically undeniable cheating at JO Wilson, Noyes, and a number of other schools, they just had an outside firm investigate Noyes. Maybe it was politically difficult. Maybe no one saw the upside. The downside you mention seems entirely accurate - people don't trust the results.

As a parent choosing schools, I was pleasantly surprised to see the good that Meridian PCS was doing with its students - getting good DC CAS results, particularly, relatively good outcomes for students from poor backgrounds. And now, we see that a huge proportion of Meridian's success in 2012, four years after cheating was first clearly identified, was due to cheating.

It seems clear that the lackluster investigation in previous years failed to deter cheating like this.

by andy on Apr 17, 2013 10:36 am • linkreport

The Intel analogy here is not particularly apt.

It took me a while to find it, but the point you are trying to make is "DCPS needs to take responsibility for their actions and show a strong response." Clearly. Corruption should always be met with a response from government that they intend to clean it up.

However, it isn't necessarily true that "The longer an organization avoids taking blame, the more difficult [it is] to move forward."

That's not an axiom. Sometimes it is easier to move forward later because the people who want to hide the problem are gone. OR, sometimes the people who want to hide the problem can be removed because it is found out. ...

Ease of moving forward also depends on the nature of the problem. An an honest though costly mistake (like Intel) versus commission of a an intentional bad act that is then covered up (DC cheating) are very different. Thus, the public's willingness to forgive and forget may be very different.

Perhaps the better thought would be to compare the DCPS cheating scandal to the Atlanta public school cheating scandal... is DC better off, worse off, or in the same boat? Atlanta is seeing indictments of teachers on the evening news.

by Finding Focus on Apr 17, 2013 11:59 am • linkreport

@Finding Focus

I think this is a great analogy about taking responsibility for mistakes made. The longer the scandal lingers, the more difficult it will be for DCPS to retain students just as with IBM taking responsibility for the chip problems allowed them to retain customers.

by Jessica Christy on Apr 17, 2013 1:52 pm • linkreport

@Jessica I very much appreciate that you want to be supportive of the post, and the overall point that DCPS needs to get out in front of this is well taken. However, for analogies to be apt situations need to be similar in more than one respect.

Still, the only reason a poor analogy in a blog post is relevant is because in not realizing that it doesn't fit, I think that you are overlooking key concerns about the scandal. You are calling this a "mistake made" and that does not due justice to the possible allegations that test answers were erased and corrected by teachers. This is an intensely difficult and delicate situation for DCPS, and how they handle it will say as much as how quickly.

Intel doesn't work as analogy because, as I said, what was at stake in that situation was fundamentally different and the behavior that led to the Intel mistake was fundamentally different. In Intel no one was under threat of subpoenas or possible jail time for the not-very-often relevant flaw that Mr. Nicely discovered. In Intel, the integrity of public servants was not questioned and nothing akin to government corruption was alleged. In Intel we did not face a public backlash and political fallout that could threaten to deepen the socio-political and racial divides that a community has trying so hard to overcome. In Intel, the situation did not overshadow the real work of critical public policy reform that could be delayed or derailed impacting the future of thousands of kids.

Communities are not companies, and they shouldn’t be run like one. Community institutions have a different relationship with public trust, and in this situation that matters.

Think about this. The USA today article I linked to noted that 35 Atlanta school teachers have been indicted, and that the Atlanta Journal Constitution's investigation suggested that 196 school districts nationwide might have been partaking in similar cheating to the Atlanta cheating scandal.

Think about this. The USA today article I linked to described the entire Atlanta scandal along racial lines (accurately or inaccurately I have no idea). I hope that we can avoid such an ugly article being written about DC in the near future, and I fear what it will mean if such an article is written.

Think about this. The question that will start to percolate is whether Rhee was trying to protect the cheating (piling on to the negative view of her) or throw those teachers out (vindicating her "good teachers" crusade). We may never know the truth of that one, but likely both pro and anti Rhee partisans will decide it was one or the other. Think about the mess it will make if we battle that one out along those partisan lines.

There are calls to subpoena Rhee, which very well may be appropriate but which also threatens to be a political spectacle that could overshadow the real business of school reform. If we rehash Rhee. If anything starts to describe our situation as being remotely like Atlanta, this threatens to tear open wounds that have barely had the bandages off in this city. IF that happens, who gets hurt the most?

I certainly support the idea that "DCPS needs to take responsibility for their actions and show a strong response." But it is going to get ugly before it gets better, and we need to focus on how to do this the right way with the least damage to the mission of DCPS. It's a balance between justice and compassion, and that line is always hard toe.

by Finding Focus on Apr 17, 2013 5:06 pm • linkreport

Add a Comment

Name: (will be displayed on the comments page)

Email: (must be your real address, but will be kept private)

URL: (optional, will be displayed)

You can use some HTML, like <blockquote>quoting another comment</blockquote>, <i>italics</i>, and <a href="http://url_here">hyperlinks</a>. More here.

Your comment:

By submitting a comment, you agree to abide by our comment policy.
Notify me of followup comments via email. (You can also subscribe without commenting.)
Save my name and email address on this computer so I don't have to enter it next time, and so I don't have to answer the anti-spam map challenge question in the future.