http://ori.hhs.gov/THELAB
Just in case you want to play before we look at it in class, or follow-up after class.
Q: What should happen if a team can't reproduce another researcher's results?
Just in case you want to play before we look at it in class, or follow-up after class.
The Rules of Replication By Kerry Grens
http://www.the-scientist.com//?articles.view/articleNo/41265/title/The-Rules-of-Replication/Q: What should happen if a team can't reproduce another researcher's results?
The debate about replication is slippery slope. If an experiment was well designed and very detail oriented then one would think that it would be easy to replicate the study and have the same findings but as the article explains, this isn't always the case. If a team can't reproduce the same or similar results of a study that was previous done, then steps need to be taken to see who made the error. Was the error made by the researcher trying to replicate the study or is there an error from the original author? But one must go through and analyze what is wrong and how to modify the problem before you go accusing someone. The original researcher might have poured their hearts into this experiment and yielded a certain product but it isn't fair or right to call them out, until you know why or how things went wrong.
ReplyDeleteI agree with you that contradictory scientific findings should only be reported once the scientist is sure that they have correctly handed the experiment in an objective detail oriented way. I also think that this could be tied into the "treating other's viewpoint with respect" aspect of the critical thinking analysis that we did for the worksheets. By putting yourself in the original scientist's shoes when going through the steps of their experiment, you can gain a better insight as to how they reached their conclusions, and in turn make a more informed opinion as to what you personally believe on the subject. This form of non-biased objectivity can lead to a better interpretation of the data as well.
DeleteI agree with your assessment of the replication of data. Before any serious implications are taken you should make sure that the attempt at replication was done in an objective manner, eliminating to the best of your abilities to the pressures of subjective forces such as sources on funding and personal motivation and implications. It is this role of strong objectivity that the feminist emphasize and I am inclined to agree with them. Also Gould, comments on the role of subjectivity in science in the essay that we read "In the Bind of the Beholder". IN that essay he emphasized the difference between the context of discovery and the context of justification where the issue of replication comes into play.
DeleteIf there is an experiment done that is giving you all the right answers then as a person trying to replicate that experiment then you would expect it to be perfect. But sometimes that doesn't always happen with experiments. For example Marshall and his team wanted to replicate the experiment done by Chen-yu Zhang and found that there were some errors. If this is the casen then like Lauren said someone needs to go and figure out the problem. You go back and do what you have to do to find the error. As a scientist you don't go calling out other scientists about the work they have done, you learn from their mistakes and maybe come to a conclusion together.
ReplyDeleteI think your last point is the most important especially when it comes to Science and experiments where if there is error then lots of things could be effected. As a scientist you need to use history as a blueprint to help you create, and if something is in error along the way then you need to figure out and fix the error instead of continuing with wrong information/experiments
DeleteI agree with Sawyer. The process of replication of an experiment isn't for the scientist to try and find where there could be mistakes, but sometimes that is the case. When that occurs it is important to publish the findings of the replication study so that scientist can then build off of that. Scientist have build off one another's findings and produced new findings all throughout history. So although in some cases there are new findings there are also cases where the replication experiment comes out with the same results.
DeleteI agree with Sawyer, Ian and Chrissy, If replicating a persons experiment does not give u the same result then something must have changed or been done differently. Instead of pointing blame or saying it the experiment was wrong go back and see what could have possibly went wrong or have been different, move forward and continue to make advancements in the research. I like what Chrissy said that scientists build off one anothers findings, i really agree with that, i think that sums up science and how research is done really well. Scientists need other scientists to help explain the unknown and also to help check each others work.
DeleteIn science, replication is something that is very hard to do. Because in order to get the same exact results from two separate experiments by two different groups of scientists, both group have to be exact with measures and can not have any error. Also you do not know what the else is taking place at the site of the experiment. Are there different air qualities? Different room temperatures etc. But if the results are completely different then steps are needed to take place in order to find where the error is and who made the error. When Marshall did his separate experiment of Zhang's study, he indeed did find error that Zhang committed. And that is what is good I think about science. Other scientists are always out to get other scientists. ALways trying to prove them wrong and trying to become the one who is right.
ReplyDeleteI don't think scientists are out to get other scientists necessarily. I agree with what you said about replication and how it is needed to recognize and scrutinize error but the soul purpose is not to prove other scientists wrong. If the experiment was successfully replicated without error, it would still be successful. If Zhang's experiment was recorded correctly, even if it has an immense amount of errors, it still has scientific value. Though scientists do enjoy their hypothesis being supported, a scientist's main goal is to record data and gain knowledge which Marshall did.
DeleteI agree with you Brandon, I dont think that they are out to get them but I think that what we learned about how science is falsifiable is what makes it seem like this. Scientists, I believe are out for the betterment of the people who the experiment it is helping and for the betterment of science. If people weren't out redoing experiments then after one or two experiments and getting the same answer a few times then it wouldnt be talked about again and wouldnt be able to improve on that experiment or study. Science being falsifiable and the "competitive nature" of scientists is what helps it to continually grow and i think that replication is a good way to do that.
DeleteReplication in the scientific field can be a difficult thing. For example if scientists are attempting to recreate the results of an experiment and they come out with a different answer, there must be a form of error in either the recreated experiment or the first initial experiment they conducted. In order to figure this out, scientists must find that source of error by conducting the same experiment multiple times under perfecting it. This could take years to accomplish and the original researcher may not even be the scientists to perfect the experiment like when Marshall conducted his experiment and found an error that Dr. Zhang had done prior to him.
ReplyDeleteI think replication in science is difficult but should be necessary for research purposes. If another scientist replicates something doesn't it prove their hypothesis even more? However, if there is competition amongst the scientists, and the replicators use it for a profit, then the law should be involved. In art terms: It is good to study copy the way the masters had painted but in the title of the artwork you must credit them. Same goes for replication in science. If someone creates something and someone else tries to replicate it, they must give credit where credit is due.
ReplyDeleteIn classical music, it is said that the great composers steal from those that came before them. I agree with you that the process of replication is very important in science because it serves as a means of solidifying a body of knowledge to a further degree while minimizing the levels of reasonable doubt. This is why science must be falsifiable. If someone else cannot recreate your findings, you are asking them to have blind faith in you. As you can imagine, if your findings happened to be wrong, and other scientists built theories and models on top of yours, there could be very drastic consequences.
DeleteReplication of a lab can be difficult. With a mis-communication in the procedure, even a replication expert like Bill Marshal can mis-replicate a lab. A good experiment is set up for replication, in fact, that is its part of its purpose. Bill found error in Zhang's study and Zhang said if he could just talk to him before the experiment he could have told him about the error. That would have negated the replication process because Marshall would have been making the same mistakes.Replication is essential to experiments because it finds errors in data and procedure or confirms the data found in the previous lab.
ReplyDeleteGold star for Brandon for this answer. If replication of an experiment can not be done or is so hard to conduct. It would beg into question the data received and outcome to be accurate
DeleteIn terms of producing another researcher's results, I think that it is important to first do some background research on the original project so that you can effectively replicate the process step by step. This issue became a topic of debate in the study of the cross-kingdom transfer of microRNA from plants to mammals where Bill Marshal and scientists from the Monsanto group attempted to replicate a study by Chen-Yu Zhang of Nanjing University. The basis for the argument was that Zhang claimed that the replicated experiment (That produced contradictory results to the findings of Zhang that claimed that a microRNA in rice could regulate genes in the liver of mice that had eaten the rice.) by Marshal had fundamental flaws that caused inconsistency in the findings.
ReplyDeleteI think that the proper channels that should be gone through in producing another scientist's work should be first to contact the scientist, (To get background information on their processes in the conduction of the experiment.) second to conduct your experiments, third take your findings back to the original scientist to confirm or deny his or her findings, (To give them a chance to refute your process, and to allow themselves time to alter their viewpoints if something was found to be wrong in their process.) and lastly to publish your findings. Replication is a very important part of the scientific discipline because it allows for a community of thinkers to test theories again and again in order to find fundamental truths about the world that we live in. Additionally, once contradictory evidence for a claim is published, the owner of the original study should not be labeled a "sore loser" for trying to refute the findings of the replicated study. Instead they should be viewed as further replicators in their pursuits to find the errors in the original replication. Producing falsifiable theories is imperative to this process.
I like how you said repetition solidifies fundamental proofs in the world that we live in. It just gives confidence in theories and experiments.
DeleteTo me replication in a scientific laboratory depends upon the type of experiment. In physics we have been taught that it is nearly impossible to replicate perfectly anything but the simplest of experiments. This becomes especially true if one begins to work in the quantum would because particles in the quantum view do not have a set position or velocity until they are observed. So unless one is working in the simplest terms perfect replication is just an achievable goal. It is much more reasonable to come close to the same result as previous attempt at an experiment.
ReplyDeleteI am in no way a scientist but I believe that if one cannot replicate the results of another, there is no reason for worry, just re-trying. It is my understanding that even the smallest of errors can produce huge differences in results. That is why it is so important in my opinion that scientists should do the experiment multiple times and then look at the overall results.
ReplyDeleteReplicating experiments i believe is something that must be done by the original scientist to prove if the data and results are conclusive as well as by other scientists to somewhat proof read or make sure the results are true. science is great at checking each others work and is the basic foundation of what holds the science community together. although i do believe in attempting to replicate an experiment to the best of ones ability, there are to many variables in an experiment that could cause a slight difference and ultimately affect the outcome of the experiment.
ReplyDeleteI agree with Ford here, the people who should be replicating and reproducing the results are the original scientists of the experiment. If another team cant reproduce the results, then someone did something wrong. Im not sure if only one other team trying to replicate the process is sufficient enough to question the data. To summarize, the original team should be able to reproduce every time before someone else gives it a try.
DeleteI agree with ford as well. I think if someone is going to replicate an experiment it should be the scientist who did the experiment in the first place. I think it could possibly cause conflict between scientist replicating others work.
DeleteI do agree with Jensen and Ford in most aspects here, but I think replicating experiments is an okay thing to do. Sometimes experiments and results should be found by many others than just the original scientist. I think if more then one scientist put their brain power into a certain experiment, they would come up with more precise data. I've always been taught that two minds put together is better then one.
ReplyDeleteI enjoyed the simulation game of the Lab because it showed some of the previously unthought of question involving integrity and honesty in the workplace. Inability to replicate the results of a published study can be a result of simple miscalculation or an intentional falsification of data to due personal, professional, or financial needs. One from the issues I found interesting while I was playing the character of the post doc was the conflict between the requirements of being a scientist ( in some fields) and the requirements of being a family man. Also the risk associated with being a complainant including isolation from coworkers, lack of support for you experiments and, a setback of academic progress. The allegation of research falsification of data should be taken serious a through investigation should be done and all concession should be made to ensure the confidentiality for both parties. In a perfectly objective world there would be no need to fabricate date because there would be no subjective bias in the field of science,
ReplyDeleteI found the lab stimulation very intriguing. I picked to be the lab director. It was a learning experience because I've never known what really goes on in a science lab so even though it was a stimulation it was interesting to see what kind of decisions a director of a lab has to make on daily basis.
ReplyDeleteI too found the lab interesting. Playing the married post doc visa researcher, it gave a good realistic stresses that can play out in people lives
ReplyDelete