 |
|
You are currently at :: Resources > Articles on Research Publishing |
| |
 |
 |
| |
Honesty in Research
Honesty is defined by Merriam-Webster's online dictionary as "fairness and straightforwardness of conduct" and as "adherence to the facts" (http://www.m-w.com/dictionary/honesty). These definitions of honesty sum up what is required for scientific integrity in research as well.
Honesty in research, and in the communication of that research, is crucial. When a research paper is read, the reader assumes that the experiments described were actually performed and the results mentioned were actually obtained. Research that has had its experiments fabricated, its results falsified or created, and research based on ideas stolen from another researcher damages the future credibility of the concerned researcher.
Examples of dishonest research
The scientific community has numerous anecdotal examples of scientists who were not honest in reporting their research. One of the well-known early cases is that of Piltdown Man in 1912. It was reported by Charles Dawson and others that fragments of a skull and jawbone from an early human had been found in a place called Piltdown in England. This discovery was proclaimed as the fossilized remains of an unknown, early form of human and it caused much excitement in archeological circles. It was only some 40 years later that it was revealed to be a complete hoax, consisting of a modern human skull and modified jawbone from an orangutan.
In the physical sciences, there is the well-known Ninov case. In 1999, Victor Ninov and his colleagues working at the Lawrence Berkeley National Laboratory in USA published a report that they had found Element 118. Later on, Ninov's colleagues retracted the paper when an investigation revealed that the data had been altered, and it was found that the described experiment could not be replicated.
In 2002, Lucent Technologies' Bell Labs fired Jan Hendrik Schön after an investigation committee discovered that his research claims were fraudulent. Schön was an award winning scientist who, in 2001, published a paper in Nature, claiming to have produced a transistor by using organic dye molecules. This was path-breaking research, but discrepancies in the published graphs, etc., drew the attention of other scientists. An investigation later pointed out that Schön alone was responsible for the fraud.
A very recent case from the field of biomedical research is that of Hwang Woo-Suk, a biomedical scientist from South Korea. He was considered a brilliant scientist doing pioneering stem cell research. However, in articles published in Science in 2004 and 2005 he described the first-ever cloning of a human embryonic cell, using the somatic cell nuclear transfer method. This claim was later shown to be fraudulent and he was dismissed from Seoul National University in March 2006 after the university probe confirmed the allegations against him.
The examples above are just a few from the large number of known instances where researchers have not been honest in their research. The question that naturally arises is "Why do they do this?"
Why are many researchers not honest?
It is often difficult for a researcher to remain honest while conducting or reporting research. Like human beings in other settings, sometimes researchers perform actions that are less than noble. Here are some of the common reasons why
"Publish or perish": Competition for funds, tenured positions, and institutional support is very intense. As a result, there is tremendous pressure on researchers to publish. Funding agencies, universities, and review boards like to know that the researcher has a steady output of research that has been published, preferably in high impact journals. To show, at least in terms of numbers, that this high output of papers is occurring, a researcher may be tempted to publish one large piece of work as a number of smaller papers (colloquially termed salami-slicing). Even worse, the researcher may fabricate or alter data obtained in an experiment to make the paper more attractive to a journal for publication.
Pride and personal ambition: An ambitious researcher who is eager to reach the top of the ladder quickly, a researcher who has published one high-impact paper and is keen to show the world that (s)he can do it again, a senior researcher who would like the world to think (s)he is still doing cutting edge research, or a researcher who is willing to bend the rules in order to get the funding needed for research might present dishonest research. All these examples show you that researchers are, after all, only human. They suffer from the emotions of pride and personal ambition just like human beings in other situations. However, it is important for researchers to remember that the results they publish serve as a basis for future research on the topic and, in the biomedical sciences, for treatment modalities. It is not just the researcher's career and reputation that will suffer, but also the well-being of others.
Too confident of knowing the right answer: It may happen that researchers have a preconceived notion about the results of their research. They are sure that what they think or feel should be the result, actually is the correct result. So, when their data does not deliver the preconceived result, they see nothing wrong in manipulating their data a little bit or in falsifying just a few numbers to achieve the result they are sure is the right one.
It is important for researchers to remember that the results of a research project may throw up surprises. The results may prove the initial hypothesis wrong and may point to an entirely new direction for future research. Researchers must remember that there are no right and wrong results—there are only results. These have to be intelligently interpreted to advance our understanding of a particular topic. Manipulating the data to achieve desired results may actually misguide future workers on the topic, sometimes with serious consequences in fields like biomedical research.
Assuming that no one will ever find out: Some researchers who fabricate or alter data often feel that if the statistical analysis or math looks logical or if the paper is very technically complicated and obscure, the average reader will accept the result at face value and not try to verify it. They will then be able to get away with their fraud. This is another human failing—of thinking that Abraham Lincoln was wrong and that it is possible to fool "all of the people, all of the time"!
Types of scientific misconduct
When a researcher is not honest in performing or communicating research, it is termed scientific misconduct. There are 6 main types of scientific misconduct.
|
| 1. |
Fabrication |
| 2. |
Falsification |
| 3. |
Plagiarism |
| 4. |
Conflict of interest |
| 5. |
Unethical treatment of research subjects |
| 6. |
Gift authorship
|
|
Fabrication
Fabrication is the deliberate inclusion of false or misleading data in a research paper. Read more about one such instance where a researcher named Jon Sudbø based his submission to the Lancet on fabricated data.
Falsification
Falsification means altering research data in a research paper or omitting relevant data from the paper or manipulating experimental procedures to achieve a desired result. Read more about Sir Cyril Burt, who is accused of fabricating data in his studies on behavior genetics using twins as subjects for the study.
Plagiarism
Plagiarism has been defined by Wikipedia as "the practice of dishonestly claiming or implying original authorship of material which one has not actually created." The term plagiarism includes self-plagiarism, i.e., publishing the same material in different journals or with different titles, and citation plagiarism, i.e., not giving due credit to earlier work on the subject.
Read more about plagiarism in the Center for Research Writing Resources' article on this subject.
Conflict of interest
A conflict of interest is a situation where a researcher has competing personal and professional interests. For example, a researcher working on a particular drug may be provided an all expenses paid vacation by the drug company. The correct thing to do would be to refuse the vacation, but unfortunately there are many researchers who do not feel that by accepting a gift of stock options or paid vacations they are compromising on the integrity of their research in any way. For a well-written article on this topic, you may wish to read Shannon Brownlee's article Doctors Without Borders in the Washington Monthly.
Unethical treatment of research subjects
This is a very serious form of scientific misconduct. It can be considered to occur when researchers
|
 |
do not provide adequate information about the experiments and any side effects to the human subjects or to people who will be indirectly affected by the research, |
|
do not provide for humane care of animal subjects, |
|
knowingly undertake studies that will cause problems for the subjects (read about the Tuskegee study—an unethical study that was conducted in USA and led to many of the rules now framed for patient protection), and |
|
do not disclose all the details of the experiments that will be conducted along with their potential side effects to the human subjects participating in the study.
|
|
Gift authorship
Gift authorship is the term used to describe the practice of making the laboratory director/senior professor one of the coauthors of a paper, even if the director/ professor did no actual work on the project. The project, however, was supported by funds obtained by the director/professor from funding agencies as the principal investigator of the laboratory. A classic case is that of Professor Eugene Braunwald and John Darsee at Harvard. Read more about the Darsee case and learn how this practice of gift authorship must have embarrassed a senior and respected Professor like Eugene Braunwald, because John Darsee had fabricated data in papers in which Eugene Braunwald was also a coauthor based on the practice of gift authorship.
Conclusion
Honesty in research is vital across all aspects of research—from designing a study to collecting and analyzing data and publishing the results in a paper. It is equally important to give credit where it is due and also to not take credit if no contribution was made to a study (i.e., gift authorship). It is very important to maintain proper records of the work done and the data collected. All research proposals that involve human or animal subjects must be cleared by an Ethics committee.
Honesty in performing and reporting research is crucial both for maintenance of good scientific practice and to avoid adverse effects on health or environment for others. Honesty in research is the foundation on which academic knowledge in all subjects stands. |
|
|
|
