1 in 4 Statisticians Say They Were Asked to Commit Scientific Fraud – By Alex Berezow — October 30, 2018
This article definitely points toward a sad truth, but the sample of 390/522 statisticians from whom they “received sufficient responses” doesn’t look like a representative sample at all.
Only someone who’s been in this situation themselves would answer a survey about “inappropriate requests”. For those who haven’t, they would only check some box saying “it hasn’t happened to me” and then the rest of the survey would be pointless to fill out because it wouldn’t apply to them.
Without access to a full explanation of how they picked their sample, I wouldn’t quote these results. However…
In all the scientific articles I’ve read over the last years, I’ve noticed this trend as well. The article will be full of convoluted sentences, double negatives, and irrelevant detail, giving the impression that the study’s authors put the data through some torturous gyrations to ensure the results align with their employer’s desires.
And how can they not?
If they expose results damaging to their employer,’s bottom line, not only will their research never leave the lab, but they will lose their job and be blacklisted as a “troublemaker”, eventually losing their whole career.
When everything is valued only in dollars, I see no solution for this. Not many can afford huge financial and social consequences of “biting the hand that feeds you”.
As the saying goes, “There are three kinds of lies: lies, damned lies, and statistics.” We know that’s true because statisticians themselves just said so.
A stunning report published in the Annals of Internal Medicine concludes that researchers often make “inappropriate requests” to statisticians.
And by “inappropriate,” the authors aren’t referring to accidental requests for incorrect statistical analyses; instead, they’re referring to requests for unscrupulous data manipulation or even fraud.
The authors surveyed 522 consulting biostatisticians and received sufficient responses from 390.
Then, they constructed a table (shown below) that ranks requests by level of inappropriateness. For instance,
- at the very top is “falsify the statistical significance to support a desired result,” which is outright fraud.
- At the bottom is “do not show plot because it did not show as strong an effect as you had hoped,” which is only slightly naughty.
On the right, the authors report how often the biostatisticians estimated that they received such a request over the past five years. The results are jaw-dropping.
The absolute worst offense (i.e., being asked to fake statistical significance) occurred to 3% of the survey respondents.
Another 7% reported being asked to change data, and a whopping 24% — nearly 1 in 4 — said they were asked to remove or alter data.
Unequivocally, that is a request to commit scientific fraud.
As if this weren’t bad enough, the comments on this article were even more shocking. They revealed that these kinds of deception are well-known and common.
In light of the current political and economic situation, this is just a logical outcome when a researcher’s job depends on achieving results favorable to his employer.
When I took my first post-graduate course in evaluation at a major state university, we were told by our professor that, if we did not take into consideration our client’s wishes, “we might find ourselves teaching school in Peoria.”
It seems that this is what had happened to him, after he spoke the truth to the client who supplied his income.
Although we did not question our professor’s integrity, we realized that our future sustenance clearly depended on satisfying the wishes of whoever paid our salaries.
The saddest impact of this situation is in the pursuit of government funding for research.
Because politics creates strange bedfellows, in any sensitive area in which certain outcomes are preferred by the political classes, evaluation is highly dangerous if objectivity is practiced by the evaluators.
This explains exactly why countless studies devise creative ways to demonstrate that “opioids are bad”, yet none dare challenge this viewpoint in this virulent anti-opioid political culture. I demonstrated this in Opioids Blamed for Side-Effects of Chronic Pain.
Serious statisticians find it difficult to walk the minefield of grant writing and final reporting when their livelihood is challenged by political correctness.