Research Misconduct: Sources and Experiences

Section Overview

In this section, we asked scientists their opinions regarding activities that may or may not be considered research misconduct and possible causes.

Question

Finding:

The top three forms of research misconduct scientists report as being extremely serious are fabricating data (90%), willfully suppressing or distorting data (76%), and violating laws related to human and animal research or biosafety (68%).

Other forms of misconduct that more than half of respondents view as “extremely serious” include making false accusations of misconduct (66%), plagiarism (64%), manipulating experiments through statistical techniques (62%), deceptively and selectively reporting findings (59%), and misappropriating funds (54%).

---

Question

Finding:

The three types of misconduct that approximately half of the scientists in our sample believe their colleagues have never engaged in during the past five years include: making false accusations of scientific misconduct (50%), fabricating data (50%), and violating regulations and laws about human subjects and animal research, biosafety, and radiation safety (46%). Also, during the past five years, more than one-third (37%) believe colleagues within their institution have not submitted manuscripts to more than one journal at a time, and nearly one-third (31%) believe their colleagues have never misappropriated funds.

Other activities scientists believe their colleagues have never engaged in (during the past five years) include suppressing or distorting data (29%), failing to report misconduct (26%), failing to declare conflicts of interest (24%), deceptively manipulating experiments (24%), and deceptively and selectively reporting findings (24%).

---

Question

Finding:

Pressure to publish is most widely recognized as a major cause of research misconduct (78%). Nearly half (44%) of scientists in our sample identified time pressures as a major cause of misconduct, while one-quarter identified lack of funding (26%) and political interference (25%).

---

Question

Finding:

Among respondents, 47% believe that expanding political control over the U.S. science system will increase research misconduct “a lot,” while one-third (33%) believe that this will increase research misconduct “a little.” Only 15% felt expansion of political control would have “no impact” on research misconduct, and 5% felt it would decrease research misconduct either “a little” or “a lot.”

---

Question

Finding:

Among all respondents, over one-quarter (28%) report ever having experienced pressure to misrepresent authorship in their research. This is followed by ever having been pressured to misrepresent works cited (18%), misrepresent findings in conclusions (6%), and misrepresent data analysis (6%).

---

Question

Finding:

One-fifth of all respondents report having been pressured by a collaborator to misrepresent authorship (22%), and 5% identify other researchers in their institution as having pressured them regarding authorship. [TC1] [TJ2] A few scientists also identify their collaborators as a source of pressure to misrepresent other aspects of their research, including conclusions of research findings in scientific products (4%), data analysis (3%), works cited (3%), and oral/written presentation of research findings (3%).

9% of scientists also indicated having ever been pressured by reviewers to misrepresent the works cited in their research. 2% identified funders as ever having been a source of pressure to misrepresent the conclusions of research findings in either policy documents or scientific products.