NSF confirms results of Penn State investigation, exonerates Michael Mann of research misconduct

First in a series

As a result of the illegal publication of CRU climate emails, the Pennsylvania State University (PSU) conducted an inquiry and investigation into allegations of research misconduct by Professor Michael Mann. The University exonerated Mann of all four allegations in July 2010, but the National Science Foundation Office of Inspector General (OIG) reviews such investigations for completeness and correctness. On August 15, 2011, the OIG released the results of their own review, agreeing with all of the conclusions of the PSU investigation and subsequently acquitting Mann of all the allegations of research misconduct made against him.

PSU published the results of an their internal investigation into alleged research misconduct by climatologist Michael Mann on July 1, 2010. As S&R reported, the university’s conclusions were that Mann did not falsify data over the course of his research, that he did not destroy any emails in possible breach of the Freedom of Information Act, that he did not misuse his position or abuse confidentiality agreements, and that he did not deviate from accepted practices of conduct for his field.

As required by law, PSU reported their results to the OIG for independent review. The OIG’s review was completed and closed on August 15, 2011, with the OIG writing:

Finding no research misconduct or other matter raised by the various regulations and laws discussed above, this case is closed.

The conclusion – that Mann is acquitted of research misconduct – is obviously significant. But the details in the OIG closeout memo are important because of what they show about the original PSU investigation. Specifically, the OIG closeout memo shows that the critics who labeled the PSU investigation a “whitewash” were wrong.

When the OIG received the inquiry and investigation reports from PSU, the reviewed the reports and a significant amount of additional documentation that PSU provided upon request. Based on the OIG’s review, they “were satisfied that the University adequately addressed its Allegations 3 and 4 (misusing privileged information and serious deviation from accepted practices).” The OIG also concluded that neither of these issues rose to the level of research misconduct as defined by the NSF Research Misconduct Regulation, 45 CFR §689.

The OIG also independently reviewed Mann’s emails and PSU’s inquiry into whether or not Mann deleted emails as requested by Phil Jones in the “Climategate” emails (aka Allegation 2). The OIG concluded after reviewing the the published CRU emails and the additional information provided by PSU that “nothing in [the emails] evidenced research misconduct within the definition of the NSF Research Misconduct Regulation.” Furthermore, the OIG accepted the conclusions of the PSU inquiry regarding whether Mann deleted emails and agreed with PSU’s conclusion that Mann had not.

The OIG did conclude that PSU didn’t meet the NSF’s standard for investigating the charge of data falsification because PSU “didn’t interview any of the experts critical of [Mann’s] research to determine if they had any information that might support the allegation.” As a result, the OIG conducted their own independent investigation, reviewing both PSU’s documentation, publicly available documents written about Mann and his co-researchers, and “interviewed the subject, critics, and disciplinary experts” in reaching their conclusions. The details of what publicly available documents were reviewed and whom among Mann’s critics were interviewed is not mentioned in the closeout memo.

The OIG concluded as a result of their additional investigation that:

  1. [Mann] did not directly receive NSF research funding as a Principal Investigator until late 2001 or 2002.
  2. [Mann’s] data is documented and available to researchers.
  3. There are several concerns raised about the quality of the statistical analysis techniques that were used in [Mann’s] research.
  4. There is no specific evidence that [Mann] falsified or fabricated any data and no evidence that his actions amounted to research misconduct.
  5. There was concern about how extensively [Mann’s] research had influenced the debate in the overall research field.

Point 1 essentially means that Mann’s work prior to 2001 or 2002 was not subject to NSF review, but that the NSF appears to have reviewed it regardless. Point 2 is significant because one of the allegations of Mann’s critics is that he refused to make his data available – even though the illegally published CRU emails make it clear that Mann had made his data publicly available. Point 5 is an observation on which the OIG offered no additional comment or analysis and is a subject of additional research by S&R.

Point 3 is significant because Mann has been criticized for truncating tree ring data where it diverges from the historical temperature data in his original papers of [This issue has nothing to do with Mann, but rather Keith Briffa who generated the data in question. We apologize for the confusion] using sub-standard statistical techniques. However, the OIG addresses this point specifically, writing that there is a lot of debate about “the viability of the statistical procedures [Mann] employed, the statistics used to confirm the accuracy of the results, and the degree to which one specific set of data impacts the statistical results.” But, the OIG says, “these concerns are all appropriate for scientific debate” and that “such scientific debate… does not, in itself, constitute evidence of research misconduct.”

Point 4 is the key conclusion – there is “no specific evidence that [Mann] falsified or fabricated any data” as some of his more vocal critics have contended. The OIG reached this conclusion after interviewing Mann’s critics, after reviewing the CRU emails, and after reviewing other “publically available documentation concerning both [Mann’s] research and parallel research conducted by his collaborators and other scientists….” Furthermore, the OIG didn’t just limit their investigation to data fabrication as the PSU investigation did – the OIG did a full research misconduct investigation according to the NSF Research Misconduct Regulation. According to this regulation, research misconduct is defined as

fabrication, falsification, or plagiarism in proposing or performing research funded by NSF, reviewing research proposals submitted to NSF, or in reporting research results funded by NSF.

The regulation further define fabrication as “making up data or results” and falsification as “manipulating… or changing or omitting data or results” to lead to false conclusions. Mann’s critics have claimed that Mann manipulated the data he used in his papers, but the OIG specifically ruled that this was not the case. After all, the regulation states that “research misconduct does not include honest error or differences of opinion.”

Ultimately the OIG’s review and supplemental investigation agreed on all counts with the PSU inquiry and investigation – Mann did not falsify data, he did not destroy any emails, he did not misuse any privileged information, and he did not deviate from accepted scientific processes.

Other sites reporting on the OIG’s exoneration of Mann:

Joe Romm of ClimateProgress broke the story
Climate Science Watch
Douglas Fischer at The Daily Climate
Richard Littlemore at DeSmogBlog
James Fallows of The Atlantic
Hank Campbell at Science 2.0
The Policy Lass
Eli at Rabbett Run
Centre Daily Times
Greg Laden at Science Blogs
Scott Mandia
the Unitarian-Universalist United Nations Office
Barry Bickmore
Andy Revkin at dotEarth
TPM Muckraker
The Summit County Voice
Bad Astronomy

6 replies »