Several institutions from developing countries like Nepal don't have a single statistician or any personel from community medicine in the research and ethical committee of the medical institution. Finally there will be a poorly conducted trial. Is it correct?
Wasn't it Ronald Fisher who said: "To consult the statistician after an experiment is finished is often merely to ask him to conduct a post mortem examination. He can perhaps say what the experiment died of. "?
Several medical journals in developing countries are found to have methodological rigor when compared to those from developed European countries, which may contribute to the reasoning that research from Asian region is ‘fraudulent research’. It is very easy to find out whether the research is fraudulent or not by the critical review of the research article by the expert editors. It is sincerely recommended and encouraged by the Managing editor and Author that the contributing researchers follow a diligent and systematic pattern in conducting and presenting their studies. This will not only lead to improved quality of research but will also enhance and augment the quality of Asian journals and thus, contribute meaningfully to the progress of research and improvement of medical care in developing countries.
Whether the medical research should be a group work of Medical Personal and statistician?
Typically, research type institutional committees will have a statistician representative. It's good to have all types of experts on these types of committees, because they each bring something different to the table. To answer your question, no, it is probably not necessary. But it does add to the validity of the research, especially when it involves some type of study design.
And what about research where Biology is not involved. I think teaching Biological sampling to statistician is simpler than teaching statistics to a biologist
One seems both necesary and sufficiient. What an odd question. What's the dfference with asking if a geologist should be in a committee judging a research proposal in which geology is used? (Obviously yes.)
If you are perhaps looking for arguments NOT to include somene you don't think is compentent into a concrete body, this forum won't help, obviously.
The purpose of research is to test a hypothesis, in order to make sure that the test is both fair and accurate we use statistics. Once the hypothesis becomes complex so does the statistics needed to test it. Studies that seek funding without having the means (both in terms of design and personnel) to produce meaningful results are unlikely to be funded, or worse may be funded but will not add to the body of knowledge.
Ethical considerations are also important in the construction and conduct of research studies, both for the well being of participants and to meet the requirements of many scientific publishers (who may now compel authors to demonstrate that their work received proper ethical review).
Therefore it is of the utmost importance that both Statisticians and Ethicists are involved in the planning and review of studies before they are presented to funders. I do not think it should be a requirement that either are named as PIs however I do think that proposals should indicate how and what contribution such experts will make in the overall process. You may wish to consider collaboration with other organizations who do have such expertise, as this is a useful way of taking research forward.
Statisticians always a part of the research committee: whether reviewing a manuscript or statistical part.....
they help to optimize design, analysis and interpretation of results and conclusions.
Haklısınız. Malprakits alanında düzeltme ve önleme yaparken akredite sistem oluştururken gönüllülük esasına göre disiplinlerarası tıp kurallarına göre çalışmak gerekir. Saygı ile.
The purpose of an IRB is to protect the rights of research subjects. The scientific rigor of the study is relevant only because it affects the balance of risks and benefits. If a study can't contribute any useful knowledge, it probably provides no benefits and therefore no risks can be justified. If there are no meaningful risks to subjects, and if there are potential benefits even in the absence of statistically significant results, even an underpowered study could be approved.
An IRB should have members who can evaluate the likely risks and benefits of the research proposed. That may or may not require a statistician, epidemiologist, or other research scientist.
I agree with Ghaiath. A study's power to show what it aims to test should almost always be assessed if only to ensure that resources are not being wasted on an underpowered study that can never deliver an answer to its question. The opportunity cost of such studies needs to be accounted for. Even if an observational study in front of an IRB poses no to physical risk to a patient, it is still the IRB's task to protect them from unjustified inconvenience and to act as a bridge between researcher and lay person. A lay person would not be able to judge if a study is adequately powered, but their decision to participate may well be influenced if a statistician were to conclude that the study is very unlikely to be able to answer the question that it set out to.
According to FHI360 a scientific review committee is a totally separate animal from the ethical review committee. The proposal should be reviewed by the scientific review committee (some universities call it the university research review committee) as a first step before submission to the ethics committee. The scientific review committee should have the required scientific expertise more so than the ethics review committee whose sole responsibility is to ensure the protection of all research participants. A complete description is available on the FHI360 site.
http://www.fhi360.org/training/en/RETC2/index.html
I agree with the colleagues statisticians. But agree with us non-statisticians, clinicians, too?
Statistician, an important member of research activities. He/She guide researcher as well as an organizaiton by number of ways and avoid complications in furture ,So I strongly fell that statistician must be included in ethical and review committee.
I beleive that involvement of a statistician in a research and ethical committee for reviewing a manuscript and research proposal is a must.
Statistician is an important member in the research institution and ethical board. if not present in site can be communicated.
It is very important and necessary for proper review of papers and publication of genuine and logically done researches.
That makes sense@Ghaiath. the fact remains that not all scholars are proficient enough in the use of statistical packages in the current research would. Lets assume when you are using more complex statistical methods which are not commonly used by regular researchers, reviewing such work required not only the background knowledge of the subject area but also the statistical methods used in analyzing data from the study. i still maintain that "It is very important and necessary for proper review of papers and publication of genuine and logically done researches".
Yes ! Statistics are essential in acquiring knowledge. Let me quote what C.R. Rao , one of the top most statisticians " All methods of acquiring knowledge are essentially through statistics".
Really, no statisticians in Nepal? I'll come. Just pay my way. You guys have some beautiful mountains I would love to climb. Just pay part of my way - I'll pay the rest from my personal income.
But seriously, statistics is a very new science, and it is at the foundation of all science. I believe statistics is a revolution in how we think about the world. We wish we can learn a few things and predict how everything will turn out, but it doesn't always work. Aristotle invented logic - which I totally understand - but logic requires true statements to produce other true statements, and nothing seems really true in this world other than the certainty of our individual existence (as Descartes argued) and our eventual demise (who can I ascribe that to - Schopenhauer?).
All science should be based on data. Statistics not only provides methods for analyzing data, but also for collecting or generating the data. And, admittedly, statisticians have shown that the most valid way to get data is to consult a good statistician in the first place.
OK, hopefully 'nuff said.
The primary reason, that a statistician should be part of an Ethics Comite, is that a study might be worthless due to statistic flaws. Patients might suffer from the study (placebo, risks etc) and no advantage would be for these patients nor future patients.
This is reason, that in GErmany ethic-comites have always a statistician involved - and my experience is, that statistics is that point, which is most often discussed - often with an improvement of the study protocol.
In other regions, there might be a scientific comite focussing to the statistics. In that case an ethic comite might trust their assessment and would not need a statistician.
Statistical packages? ... gag me with a spoon - oh, you don't need to.
Don't get me wrong - statistical packages in the right hands are totally cool. A lot of folks think statistics is just an exercise in computing - find the right formula, plug the numbers in the right place, get your answer - you are done, if you get paid I suppose, and don't care how much harm you may have wrought.
It's actually not so easy to "find the right formula." And plugging in the numbers - heck, I can type a lot of numbers in a minute. I try to convince my non-statistician friends (and really, all my friends are nerd-scientist type folks) that they need to do good study design to get their numbers in the first place.
Hey, but I'm not Jesus of Statistics. I'm not going to be crucified for y'all doing stupid things with worthless data. You would just look at me on the cross and say, "What a nut-case. I never asked him to be crucified for my silly analysis of my ridiculous data." So far, I've exceeded Jesus' alleged age at death by two-fold.
Hello Dr Sathian, Yes this is very important as otherwise even if ethically it is OK if poorly designed it will not show results that are meaningful or valid and then you have wasted valuable resources. Can you not send proposals via email to someone in a country a colleague with knowledge who can input as a volunteer even if your country does not employ a statistician ? Regards, Deborah
I am glad to see that my answer above was up-voted even though it was really really vague. To give a more specific example of why statistics is important you can consider; power analysis and there is some detail on this on the favourite links page of my website in the online statistical book. http://sites.google.com/site/deborahhilton/
How if you design a trial do you know that you have an adequate sample size to detect differences that are sound, real and not as a result of chance or random deviation. ? Are you familiar with terms such as type 1 which is the false positive rate ? do you know what this should be kept at ? Are you familiar with the term power and type two error ? do you know the terms alpha and beta ? how do you calculate sample size ? Is this too high or low and what are the effects of either too large a sample or tooo small a sample. How much do developing countries deviate from calculations as a result of cost limitations imposed by poor finance and what effect will this have on the result ? If you have a chance to read this chapter these are some of the important details that statisticians will consider when designing a trial and while the hypothesis and background is equally important in terms of the design or what you are trying to investigate, a poor design is going to mean your conclusions are clouded as like looking through a stained glass window to see if there are birds in the sky - you can't notice the obvious which can even happen in well designed trials that are analysed by experts so sometimes detail is hard to discover even with mammoth amounts of input and detail and investigation - people are only human and that includes Professors. Sometimes the opposite happens in well off countries - too much effort and time is put into design that is not necessary and as a example we have so many trials on exercise, blood pressure reduction, diet for weight loss and CVD risk reduction in that it is well known that exercise and diet are useful for health so rather than copious more investigations costing millions in terms of the design of studies, why not channel money directly into programs and even put some treadmills in the offices and get the researchers to exercise as in fact at times it seems odd to me that they are designing studies while at the same time getting more sedentary and overweight themselves as they are not exercising - a double edged sword.
Biomedical and medical research activities without statistics is no real science. What's make science is science is the statistics. Therefore, choosing the right method of statistics and type and number of data collection are very importnat in such cases to perform good project and save time. involving statisticians in the project activity and committee of research and evaluation of manscript is like ISO or standardizations scale.
Hi, all!
I am a member of different Committees, both ethical and research committees, and we should differentiate the necessity or the convenience.
From my point of view it is not necessary but is recommendable. From the ethical point of view It is important that the sample size and the statistical analysis is adequate for the objective of the study.
In many occasions the sample size calculation in CT protocols are impossible to undersand (and follow) for clinical researchers. If the number is insufficient to answer the objectives, the ethics are at risk: people set at risk with no scientific benefit. And obiously the other way round, an excess of subject pose some people at risk also without scientific benefit.
Therefore I would recommend either an statistician or a epidemiologist to be part of Ethics Committee.
As a statistician, I think that more than a statistician need a methodologist and statistician when required, however, I think the statistic is also necessary in investigations (right). Many professionals in other subject areas, just to have experience in your field believe fully meet all statistical procedures, therefore, I think this is important.
Using meta-analysis of various investigations, draws attention being varied procedures, noting the team, do not incorporate statisticians.
although I do not deny that there are experienced people who have managed to fully meet statistical processes.
Yes for a research committee It would indeed be very appropriate to have on board an statistical expert (in biomedical research committee, an epidemiologist with sound practical knowledge of biostatistics would perhaps suffice). This would enable critical evaluation of study designs, appropriateness of sampling strategy and sample size estimations, data collection instruments, data analysis plan etc. all of which have a bearing on the quality of data that will emerge. In terms of an ethics committee, the proposals in normal course would have first been presented to an research committee and only after getting its approval. So it may not be mandatory to have a statistician there since by and large the focus is on the ethical aspects (though some ethics committees also review the scientific aspects)
Yes, it's necessary to have good, qualified statistician in your team.
A statistician would be benificial as most of medical personnel have minimum knowledge on how to design study : the study population and later on analysis of data
Great and impressive views from all the members. In my opinion, it is recommended that any resaerch activity having solution based on the application of any of the statistical methods needs statistician suggestion. In such a case the suggestion of statistician would be benificial for recommending an appropriate method of data analysis.
Yes, all those having some sought of statistical skills and running of statistical packages are also required to take the opinion of statistician because for application of any of the statistical methods we need the fulfilment of basic assumptions behind that statistical method. In case of blind use of statistical method, without checking the basic assumptions, the results will be misleading.
It is essential biologists and medical doctors learn statistics and research methodology, otherwise they are simply wasting time, it is much easier to learn the statistics you need by your content viewpoint than try to teacha pure statistician medical and biological problems.
That sounds very strange to me, data analysis is so strictly embedded with the scientific way of doing in terms of questions, answers, mental model generating that cannot be separated by biology or medicine. You learn statistics by doing it and reasoning on the sense of what you are doing, courses are only a little part of the game.
I'm reading correctly???, then do a course on dermatology and so I avoid going to the doctor because I recognize that it is and I self-medicate. I apologize end of example, but always the same at a different level.
I sincerely believe that training, are involved in the person understands the final results, not necessarily to understand the calculation or at least understand why a procedure or another.
Many researchers whom I have helped, I wonder why a procedure and the response has been: "do the same thing everywhere," or "the trend is" or "is the same that appears in the articles"
I've seen over the 22 years as a statistician, that consultations are always the same and the main one is: What can we say to these data???
Dermatology is only an aspect of medicine, and clearly you must know all the context information that lie around the simple 'observe the skin' like the presence of angiomatic lesions, fibrous vs. lipidic aggregation and so forth.
To deeply understand what a correlation is and what is the relevant scale for performing a study, what are the statistical units and what are the variables, what does it mean to prepare a balanced sample was, until 30 years ago the normal background of any experimental scientist, I agree that now biologists ask 'what I can do with these data' but this is simply pathological and, even if this is the rule or, better, the majority vote, does not mean that is OK, it remains pathological, even during the epidemy of the mid XIV century more than half of Europenas got sick but they did not consider themselves as the 'norm'. Especially now that we have computers at zero price, people can concentrate on the meaning of the statistical methods so they do not need to know calculus and they can concentrate on the adherence of the methods to their concepts and thus recognizing in a first very relevant and all positive loading component a 'size' index, while minor point to 'shape' (it is relevant these names come directly from biological sciences), the same holds true for the concept of cluster as a colelction of very similar elements and the difference between hierarchical and not hierarchical methods (again a concept that was born in biology). Clearly a course is not sufficient to aquire this wisdom, it must be acquired 'on the run' while doing research, in many cases it is sufficient to think about the general sense of what one is doing.
Thank you so much for the comments from the colleagues and seniors. Your comments will help several people like subject experts and statisticians to be together for the better outcome. I want better researches should be carried out in developing countries nothing else. I feel still lots of bio statisticians doesn't know their importance and the responsibilities in their institutions. I have attached one interesting article from international journal of epidemiology
Of course, is very important for the clinical trial designe, is an important member of research activities staff. This proffesional is able to see a clinical problem a different way and give to us an important interpetation and help with the organizaiton by number of ways and avoid complications in furture , I ' m agree to with the idea that a statistician must be included in ethical and review committee.
The initial question should be modified to clearly explain the actual responsibilities of the committee. It's clear it deals with ethics. It's not clear whether this, or another committee has the lead on design and statistics review. The responses seem to have been muddied by this lack of clarity. It's launched a defence of the statistics per se, which was not the intent.
I don't have a problem with statistician being on an ethics committee. If that is what the question is. My response is qualified by my comments below. Also, if a committee has a limited number of seats, you might want to ask what the best suite of skills to bring to the table is. Maybe a researcher, who has pragmatic experience in balancing ethics, stats, budget etc.
My concerns around the review process are broader. They also partly apply to journal reviewers. They need to help maintain standards such as those around design and ethics, with impartiality, without conflicts of interest, with the knowledge, and finally the people skills to work within a committee, and to provide constructive advice. The institution should have programs to ensure such people exist within these committees.
I think one should consider the rhetorical question....Would there be an ethical issue if a study is designed and implemented with little or no potential to detect effects?
If this is an ethical issue, statisticians should be involved.
I am a member of Research committee( Technical advisory committee) and also Ethical review committee of our institution. In my view it is necessary to include a statistician in ethical review board also because any significant statistical error renders the study results invalid and distorts the scientific quality of the study. Statistically wrong studies are definitely ethically wrong for one simple reason that we need not waste limited research resources because society has invested in it and if the study is an interventional trial like a drug trial what right we do have to place patients or human subjects on risk of research when our research in the first place has wrong statistics in it.
we hardly see in INDIA that an ethical clearance in medicine or medical research seriously: except in nationally funded and reputed institutes: i have seen that it is merely considered as a required document: we should slowly make it a good habit to have an organized ethical board
Statistician should be the core member of an ethics committee because unless a research proposal is methodically sound and with appropriately selected adequate sample size, using resource whether human or financial is clearly unethical and this type of research are not eligible for fund allottment.
I believe a statistician should be included in the ethics committe if possible.
In ethical reviews, an issue that crop up regularly is whether what is gained from the study is in proportion to the damage (of life, privacy, health etc) and cost of running the study. Conducting a much larger study than is necessary for reaching the research objectives is unethical because more damage is caused than needed. In animal research, this is expressed in the three R's: Refinement, Reduction and Replacement.
Conversely, conducting a study that is too small, with too little power to reach the research objectives is unethical because damage and cost has been caused in vain.
Because of these reasons, a study size investigation is crucial for any research plan, and study size investigations tend to be tricky from a statistical standpoint, both to carry out, and to understand. Further, using the optimal statistical methodology can reduce the required study size, and so for the same reasons as above the quality of statistical methodology is an integral part of an ethics evaluation.
Regular MDs cannot be expected to be able to judge if a specific study size calculation is sound or if the methodology suggested is suitable. A statistician provides much needed competence for these tasks.
Dear senior scientist's and colleagues thanks for your valuable comments.... Now we are getting more positive comments towards to remove the methodological rigors in medical research. I need to answer the reply for few comments. Institutional ethical and research committee anyways have subject experts according to the institution to check the quality of research. eg. from a medical institution each specialty one person. But they don't keep the bio-statistician. Because of this reason researchers go to the bio statistician after collecting the data or don't go. Once study finished required sample size is not there then researcher has to manipulate data. We, Journal editors are wasting a lots of time for reading and rejecting this types of research manuscript submitted by the authors.
An extremely important fact is put forward by dr brijesh: yes we don't bother to consult a bio statistician before going ahead with proposed project: we need to get away with such habits: As research by it self means any project that is conducted in a very organized manner
Wasn't it Ronald Fisher who said: "To consult the statistician after an experiment is finished is often merely to ask him to conduct a post mortem examination. He can perhaps say what the experiment died of. "?
Rense Sir, Your answer is absolutely correct.... Please give the reference of these lines...
Brijesh: the quote is easily found on the internet, although some (slightly) different versions are around. The (published) source is:
R. A. Fisher
Sankhyā: The Indian Journal of Statistics (1933-1960) , Vol. 4, No. 1 (1938), pp. 14-17
It can be found here:
http://www.jstor.org/stable/40383882
The increasing number of underpowered research and research presenting wrong statistical methods says that it is absolutely necessary to have statisticians in the institucional research and ethical committee.
I come from a tertiary health care center, B.P. Koirala Institute of Health Sciences, in Eastern Nepal. Our institute has a community medicine and school of public health with three statisticians as faculty members, with MD Community Medicine and MPH programs.Medical institutions in Nepal, do have community medicine departments and also statistical advisors as present as faculty members.
Ethical committee (IERB) must have an statistical advisor. According to Nepal Health Research Council guidelines for the formation of ethical committee for heath research in Nepal in point 5.2 it is mentioned that IRC should Include persons who are able to ascertain the acceptability of research applications in terms of institutional commitments, applicable law and professional standards.
http://nhrc.org.np/index.php?option=com_content&view=article&id=51&Itemid=38
http://nhrc.org.np/index.php?option=com_content&view=article&id=51&Itemid=38
Thank you so much sirs Dr. Rense and Dr. Ashwini for very good answers and references.
Good and professional scientists/researchers like you peoples 64 comments from world wide is leading to a good conclusion related to this question. Anybody who reads the question and comments will get the importance of this topic. We will wait for the comments of some more good researchers/ scientists who knows actual research like you people to conclude this topic.
I would say that not only you should include a statistician in review of manuscripts, a statistician should be a co-author of manuscripts. I cannot imagine how including a statistician would hurt a study, but i can think of many ways in how it would improve (well, given that a statistician has good environment)
I agree with your opinion about statisticians being co-authors of manuscripts @ Iryna Lobach but in Africa and my country in particular just like the author of the question stated, there are hardly a statistician in most institution let alone being members of ethical committees. Some thing need to be done.....................encouraging specialization will enhance bringing in more of them in the system.
Basically, it would be very helpful to have statisticains in the university or institute or at the ethical committe to review the research proposal and may he gives an advice in sampling number and type to reduce any unnecessary samples. However, when it comes to the manuscript (reviewing), it can be nice to have his advice in statistical methods (section), but including him as a coauthor is not fair in a way. The manuscript is usually a novel and is not something exist to repeat (he will not give you something new). The authors of a manuscript are considered according to the contribution and usually limited in number. Acknowledement section can be used to thank statistical advice.
You will have to accept that in terms of published articles: what i have experienced is that after including a statistician i could increase the number of papers from the present data that otherwise i would have published one big paper
Statistician are part of research committe but are not a part of ethical resarch committee in our set up
in medical research: A statisticians role is often ignored: they are rather just used as persons who do your data analysis: Personally though i feel if properly handled the medical research can benefit a lot with a good statistitian
I really do not see the pregidous of having a statistician sitting on a review pannel, unless the paper or the proposal being reviewed is based on statistical conclusion. I think Iryna is correct that if a ppaer is heavily based on a statistical inference, then a statistician should be a co-author. However, this condition cannot be generalized for all papser reviews.
If you are doing any sort of statsitical analysis then yes - it is a common requirement for most university ethics committees. If you don't have any available grab an introductory statistics text and start reading or use the same design from a published study in your area.
If a proposed project is approved by an ethics committee of a hospital or research institute due to its impact on patient care and danger to the subject, then there is no need for a statistician since the statistical analyses required for publication are the responsibility of the author(s) of the submitted study for publication.
It is very important to ask what is the purpose of a IREC.
I believe one of the main goals is to prevent research subjects undergo unnecessary procedures. Any risk that an individual will undergo, no matter how small, makes sense only in the face of possible advances from the research results.
A research conducted without a good statistical design or wrong statistical analysis will not bring any contribution to science. Thus, this research can not be approved by the committee. And this kind of research, as hard to believe as it is, is extremely common.
It's amazing how many mice are euthanized annually in biological research totally unnecessary, because its statistical design completely mistaken indicates, in principle, that no results will be obtained because the very low statistical power.
Evaluate research, which is the function of an ethics committee, without a serious statistical analysis is inconceivable!
I totally agree with Sandro Sperandei. It is fundamental to include statisticians in those Committees.
Yes i too cast my vote for a statistician be included to give appropriate direction to a research work where a lot of funding is involved and where research is defined to translate in to practicality.
Yes, it is necessary, always, unless you are sufficiently trained in statistics to fill the role yourself.
I believe we need to distinguish the role of a statistician in an ethics committee which evaluates the scientific/medical merits of a proposed project versus the statistician that provides guidance to a publication committee whose role is distinctively different. The role of the former is one of evaluating the statistical component of the proposed project whereas that of the latter is one of providing an opinion to the publication committee as to whether the results described have any significance or merit. As stated earlier, once the project has been approved by the ethics committee, the appropriate statistical measures presented in the study to be published are the responsibility of the authors who, for many American Univ/Med environments, require prior approval for publication, etc..
Optimal design of any IRB should include a statistician or a member of the IRB with advanced statistical expertise. The IRB should include diverse members one of which should be a statistician if possible. The statistician can be very helpful in the research design aspect of the proposed study and optimize the scietnific merit of the study! I have seen over the years pr the years plenty of studies with good research questions with poor research designs resulting in no statistical signficance achieved (probably due to the design). The statistician can definitiely improve the studies' efficiency as well as optimize their outcomes.
In New Zealand, as I understand it, statisticians have been removed from our reorganised health ethics committees - the ethics committees ask instead if the study been assessed by an independent scientific committee eg an accepted funding organisation . If approved by one of those (that do have experienced stats folks on them), then the ethics committee accepts their judgement. I don't know what they do if there has been no independent assessment though. I think that basically the decision was a question of potential for duplication of effort by two different government-funded bodies.
Certainly any study which is crunching numbers needs can benefit enormously by having a methods/stats person - not just in checking out the design, but as an essential member of the team who can consider the best options all stages - clarifying the research question, study design, planning and operations, as well as during close-out, analysis & write-up. Such a person can save $$ too - by eg suggesting a cross-over design rather than a parallel trial, if appropriate. And of course avoid needless re-work if good monitoring is done during the data collection phase. But I am getting off topic now.
Then also in a blinded trial, a stats/methods person is important to have on the Data Safety Monitoring Advisory Committee.
Like Thomas, I agree with Joanna. This argues for the presence of a "stats person" on ethics committees who ought to be tasked to bring his/her particular skills to bear on a study which has been approved by an entity that the committee deems deficient in methodological review. In other cases the opinion of the resident stats person is unnecessary. In short, all studies will be reviewed by some person with the necessary skills to do so, and redundancy of effort will be avoided.
In the initial question coda, Brijesh notes that people with statistical/methodological skills are in short supply in some countries such as Nepal. I'm unsure how to say it gently, but those institutions/organizations going forward without appropriate review in any regard are placing themselves at risk of having those studies or study results challenged by others with more methodological resources at hand. Find someone with the expertise, recruit or hire them even on a consulting basis, and legitimize your science in terms of ethics, conduct, and presentation of results and conclusions.
...to solve this problem from the root, all fields of study should incorporate intermediate level statistics at the undergraduate and Masters degree level, and introductory Advanced level statistics at the post-graduate level.
Its important for every field to think of their field of study side-by side with statistical considerations,as this will aid appropriate project designs.
On the long run, researchers will not be on the disadvantaged pedestal when it comes to the need of statistical help.
A statistician should be a must have for an institute. However, he should be consulted in planning the experiments and for data analyses as well to have his expert opinion about the results. However, it is not the case most of the times. Researchers have to go on with whatever statistical knowledge and background they might have. That often imbalances the equilibrium. Sometimes we have a good statistician than a researcher and otherwise. In short, ther should be a statistician who has some backgorund of the discipline he is working for too.
It is absolutely necessary to include a statistician to throw light on sample size and other details of significance of the study etc.
lalitha kabilan
Yes statistician is necessary to consider sample size and consider the analysis part of the study. Research paper should go through an ethical committee. Otherwise it will be ended up with unethical problems.
It's highly advised. It's like I've always said, "You can lie with statistics but not to a statistician."
OK, so this is a clearly racist comment...but funny saying not enough Indian statisticians is like bemoaning the dearth of Russian Mathematicians....
OK, please don't take this seriously...really just a joke. Please don't excommunicate me from the group!!!
I guess it's OK for a statistician to be a little bit racist given the plethora of lying statistician jokes we have to endure!!
You guys are a hoot. Okay, I've got it now. So to summarize:
Presence of a Statistician in the Ethical Committee will help a lot in
1. deciding the sample size (which is the backbone of the study).
2. deciding the right statistical tool for that particular experiment (as most of us might have seen that in the absence of a statistician a particular is repeatedly used in all kinds of experiments in a lab. as that is the only test that the researchers there know.).
3. better representation and comprehensibility of results (hence, better publications and more visibility).
4. easing out the workload of a scientist (so that he/she may be able to do more of experimental work rather that juggling with figures.).
5. identifying who is lying and who is not.
6. solving the world's race problems.
7. clarifying language through the addition of precise terms such as "contumacious" and "proprioception".
8. Disbanding the Ethical Committee
Is that about right?
Evaluation of proposal surely will gain, if professional statistician will evaluate it before money will be wasted.
Another ethical question: if institution has a good statistician, will he/she co-author all consulted publications? If yes, then how about ethics? will he.she became a most productive author of that institution?
Brijesh,
statistical analysis is usually applicable for quantitative research. For example, if a researcher wants to determine whether there is a difference in mean between two groups, then the use of a T-test is appropriate. However, if a researcher fails to calculate it, then the editor will have the right to reject the paper. The interpretation here is that the researcher does not what he or she is doing..
Your second question about the institutional Review Board (IRB), also known as ethical committee: The purpose of IRB review is to ensure that study participants are treated with respect, justice, and benevolence.
Respect implies safeguarding the study participants' personal identifiable information: Name, profession, area of residence, and many other attributes.
Justice means randomly assigning the study participants for the study. The advantage here is that every participant will have an opportunity to be assigned to the exposure. This is to prevent deliberate assignments to poor or marginalized populations to harmful exposures. For example, treatment of Syphilis was withheld from American blacks even though Penicillin was available. As a results the majority of the Blacks died from Syphilis during a 30-year period of experimentation. The purpose of the study was for Doctors to study the symptoms of the disease, which they were not familiar with. This then leads to the issue of beneficence
Beneficence implies gaining benefits from the results of a study. For example, the application of HIV treatment to everyone who has the disease.
Thanks,
Ben.
Not only should a statistician be involved in the ethical review committee, but as a matter of appropriateness be consulted in the build-up of the research proposal. This will not only help accuracy in sampling design, but prevent incidences of wasted field efforts due to poor designs.
Yep, In my opinion Statistician should include in the institutional research and ethical review committee because he could analyze better as compare to other researcher.