Think I completely bombed one CDA question...your thoughts?

[ryan1580]

At U of T interview, one question:

 

You are a graduate student. You have recentlly conducted an experiment and got back some data; however, the data you get back is unexpected and does not fit your initial question/hypothesis. You meet with your supervisor and he asks you to change the initial question. What do you do.

 

 

 

My answer, in a nutshell:

 

Basically, in research, you are going to often get back results that were not predicted and could have been a result of factors you did not take into account. This often leads to new insight as to what may actually be happening, and thus new questions can be asked/new hypotheses tested. This may cause the research goals/direction of the lab to completely change, which may not be in line with my own personal interests. (I then elaborated a little bit on this).

 

However, in hindsight I think I completely missed the point of the question:

 

Do you guys think the answer they were looking for had to do with the fact that what the supervisor is asking for unethical and dishonest? Should I have instead talked about how I would deal with that aspect?

 

 

This question was kind of stupid and also kind of ambiguous, with different interpretations. What do you guys think.

[eugenius]

yeh it is a bit ambiguous. i think ur answer was pretty good. imo though, the question represents a problem/issue, and u kind of only talked about and justified one side of the problem. u should have maybe mentioned reasons for staying with ur initial hypothesis and perhaps getting permission from ur prof to carry out further experiments if the data that u obtained so far did not fit, instead of just completely changing ur hypothesis. of course, ur prof knows best and if he seems keen on u changing ur hypothesis then u cant argue with him.

[cleanup]

I would've answered like this:

 

I would consider my professor's request (shows respect), but would ask him if there was anything he thought I could've done incorrectly or done differently for my current hypothesis. After discussing this with the professor I would request that I be given some more time to investigate the findings and see if there was anything that I missed, or if I've stumbled upon something new (this shows a semblance of dedication to your professor's initial requests and also shows that you respect the scientific method without being condescending). At no point would I suggest that my supervisor's idea was unethical, unfounded or crazy. That would be morally presumptuous and is not necessarily my place as his/her student. If I again yield obscure, unexplainable results then I would confer with my professor once more and we could come to a consensus as to how we should "change the question" (though I hate that wording; I think "reevaluate our hypothesis" might be better).

 

The question is weird. It hints at a lot of different issues but the fact of the matter is that you should try again, but only after discussing with your supervisor. Don't immediately assume that his decision is unethical, as he is far more experienced in the field than you are; he may simply know something that you don't. Instead, request that you be allowed to try again, and come out of it in a better situation no matter what happens.

[powerpenguin]

To be honest I remember the question as being "asked you to change the data"... so I think _I_ completely bombed the question by focusing too much on the moral aspects of it.

[cleanup]

To be honest I remember the question as being "asked you to change the data"... so I think _I_ completely bombed the question by focusing too much on the moral aspects of it.

 

That's actually a much tougher question at first, though I would answer it similarly to my post above. Imply that you want to try again, but don't be condescending. He is your superior, and although you know he is morally wrong, it is out of bounds to call him out on it. Just insist that you try again with his blessing, and you don't want to resort to anything drastic just yet (but don't call it wrong, per se).

 

I didn't interview at U of T so please share! What did you say?

[ryan1580]

To be honest I remember the question as being "asked you to change the data"... so I think _I_ completely bombed the question by focusing too much on the moral aspects of it.

 

 

 

This is a variation of my question, and I wish I had that one instead because it would been, for me, a lot more straightforward to answer.

 

Thanks for the answers, this kinda sucks it makes sense what I should have added.

[tnals2011]

o whut i had the same Q and the whole unethical/dishonest part didnt even occur to me

i approached it in pretty much the same way

crapzor

o well:(

[blondie87]

I think also potentially contacting other experts within the field of research and asking them their opinion of the results might also be helpful. Maybe there is an important factor you and your professor have forgot to consider (in the experiment protocol or even within the results).

 

Ensuring you are well informed- is there potentially different ways/ methods you can attempt in seeking results (perhaps look to previous/ similar studies- background info as well)

[XZeRO]

At U of T interview, one question:

 

You are a graduate student. You have recentlly conducted an experiment and got back some data; however, the data you get back is unexpected and does not fit your initial question/hypothesis. You meet with your supervisor and he asks you to change the initial question. What do you do.

 

 

 

My answer, in a nutshell:

 

Basically, in research, you are going to often get back results that were not predicted and could have been a result of factors you did not take into account. This often leads to new insight as to what may actually be happening, and thus new questions can be asked/new hypotheses tested. This may cause the research goals/direction of the lab to completely change, which may not be in line with my own personal interests. (I then elaborated a little bit on this).

 

However, in hindsight I think I completely missed the point of the question:

 

Do you guys think the answer they were looking for had to do with the fact that what the supervisor is asking for unethical and dishonest? Should I have instead talked about how I would deal with that aspect?

 

 

This question was kind of stupid and also kind of ambiguous, with different interpretations. What do you guys think.

 

This situation happens regularly in any kind of hypothesis driven research. If you obtain consistent results that disprove your hypothesis, you have to change your hypothesis to fit your newly obtained results but also perform experiments to prove your new hypothesis. Essentially, it's a cycle until you have proven a hypothesis.

 

In this particular case, the professor asks you to modify your hypothesis/initial question. So there are a number of things you should do:

1. Consider your professor's request. Note that he is your mentor and has more experience than you do. Evaluate the reasoning behind his request; is it valid/sound considering your results?

2. Analyze the results; how do these results differ/contradict your initial hypothesis.

3. Analyze your hypothesis, do you need to make significant/drastic alterations in your hypothesis? how can you change it to match your current data?

4. Given your new hypothesis, what are some of the drawbacks of your new hypothesis, does it fit or contradict previously established knowledge? What can you do to strengthen your hypothesis (experiments)? Confer with experts in the field, or your committee members, with your progress and their opinions

 

I think this situation would be much more difficult if the supervisor asks you to falsify data...

[uwowaitlisted]

I think this question was aiming to test your integrity, one of the seven competencies the CDA interview looks for. If it was as cut and dry as reviewing your data and agreeing with the more experienced supervisor, what competency would this examine? (I guess you could argue that it looks at your judgment and analysis of a situation, but i really think that you may have missed the point of the question).

 

I think when you develop a hypothesis and get data back that doesn't support it, it would be academically dishonest to then change your hypothesis. Experiments are designed to prove or disprove your hypothesis. You are suppose to develop a hypothesis first, then get data that either proves or disproves it. It is okay if your data doesn't support your hypothesis and the data is definitely still salvageable as you can publish an article showing how you disproved a hypothesis.

 

At the very least, even if you believed it was okay to change your hypothesis to match your data, as some of you clearly believe, you should have at least brought up the idea that this might be academically dishonest and then explain how you would deal with the situation if it was.

 

I structured my answer based on this. I formed two lines of thinking. 1- if there was nothing wrong with changing the hypothesis and 2- if it was academically dishonest.

 

I think they wanted you to show that you can stand up to an authority figure if they are doing something that is immoral. It shows that you have INTEGRITY.

 

However, i could be completely wrong and could have blown the question myself. That is the beauty and tragedy that is the CDA interview. Just my two cents.

[XZeRO]

I think this question was aiming to test your integrity, one of the seven competencies the CDA interview looks for. If it was as cut and dry as reviewing your data and agreeing with the more experienced supervisor, what competency would this examine? (I guess you could argue that it looks at your judgment and analysis of a situation, but i really think that you may have missed the point of the question).

 

I think when you develop a hypothesis and get data back that doesn't support it, it would be academically dishonest to then change your hypothesis. Experiments are designed to prove or disprove your hypothesis. You are suppose to develop a hypothesis first, then get data that either proves or disproves it. It is okay if your data doesn't support your hypothesis and the data is definitely still salvageable as you can publish an article showing how you disproved a hypothesis.

 

At the very least, even if you believed it was okay to change your hypothesis to match your data, as some of you clearly believe, you should have at least brought up the idea that this might be academically dishonest and then explain how you would deal with the situation if it was.

 

I structured my answer based on this. I formed two lines of thinking. 1- if there was nothing wrong with changing the hypothesis and 2- if it was academically dishonest.

 

I think they wanted you to show that you can stand up to an authority figure if they are doing something that is immoral. It shows that you have INTEGRITY.

 

However, i could be completely wrong and could have blown the question myself. That is the beauty and tragedy that is the CDA interview. Just my two cents.

 

Changing your hypothesis to match your data is not immoral. Changing your results to match your hypothesis is. I think you need to keep in mind what a hypothesis is. If the current hypothesis is not correct, and you have data to disprove it, you have to take the new data into account to generate a new hypothesis. Also, when you generate a new hypothesis, it must be testable.

[uwowaitlisted]

Changing your hypothesis to match your data is not immoral. Changing your results to match your hypothesis is. I think you need to keep in mind what a hypothesis is. If the current hypothesis is not correct, and you have data to disprove it, you have to take the new data into account to generate a new hypothesis. Also, when you generate a new hypothesis, it must be testable.

 

WE could debate this issue if you like, but clearly some people believe that it could be a moral issue. therefore, it was worthy of discussion and analysis in your CDA answer.

 

furthermore, when you apply for grants you detail the hypotheses of the lab and you will get funding based on them. Changing your hypothesis might mean you were given money for something you are no longer examining.

 

Again, there is nothing wrong with publishing data that proves your hypothesis was wrong.....so why would you change it at all???

 

Maybe im reading too much into the question, maybe not. What competency would you say the question tested if there were no moral implications?

[cleanup]

The question certainly does evaluate your integrity and ability to recognise moral dilemmas. However, keep in mind that it is a DILEMMA, since your supervisor is your superior and though, as a future dentist you are a morally sound and directed person, you're also a respectful one and understand that it is not always your place to disagree with figures of authority.

 

Emphasize that you recognise the immorality of changing your hypothesis/results (whatever it was; I think changing results is a more obvious question and panders more to this sort of answer) and talk about the breach of trust that would be involved in falsifying such things, but also ensure the panel knows that you are not going to be condescending to your professor. Insist that you try the experiment again, and that if you still get peculiar data, you would reconvene with the professor. You would say something like "I don't believe we need to resort to such drastic measures." This is not a forceful statement that places you on a higher moral ground than your supervisor. It shows that you have integrity without being a **** (pardon my French). Then suggest alternatives, like requesting help from other students, other labs, performing further literature review, etc. That's the proper way to go about it, IMHO.

[tnals2011]

urg.. i said i wouldnt change the hypothesis and carefully talk about my decision with the professor but i didnt mention moral dilemmas or integrity whatsoever.

i dont know how they grade the interview Qs but do u guys think my mark will suffer alot?

 

man i wanted to transfer my scores to western

i have to let them know by tomorrow but now i dont know!

[XZeRO]

WE could debate this issue if you like, but clearly some people believe that it could be a moral issue. therefore, it was worthy of discussion and analysis in your CDA answer.

 

furthermore, when you apply for grants you detail the hypotheses of the lab and you will get funding based on them. Changing your hypothesis might mean you were given money for something you are no longer examining.

 

Again, there is nothing wrong with publishing data that proves your hypothesis was wrong.....so why would you change it at all???

 

Maybe im reading too much into the question, maybe not. What competency would you say the question tested if there were no moral implications?

 

Not every question they ask is a moral dilemma. If the question asked you to change/falsify the data, this would definitely be a moral dilemma and would definitely test a number of qualities, amongst them tact/diplomacy, integrity, judgement/analysis, self control, etc etc etc.

 

However, this particular situation is asking you to change the question or hypothesis. Therefore, I would think it is solely testing your judgement and analytical skills. Again, I will emphasize that this is just my opinion. But let me explain the basis for this opinion.

 

As I mentioned before, changing your research question is not ethically or morally wrong, and presents no such dilemma. In fact, some of the biggest discoveries are based on accidental observations or unexpected results. These will generate research questions/hypothesis and are an excellent example of a situation where research questions are proposed after an unexpected observation. Sometimes research questions are not proposed until a full set of data is generated. However, changing/introducing your research question is strategic and requires various considerations.

 

One consideration I would look at is the point in your research career you are at... If you are at the end of your research career (eg masters/PhD) and all of a sudden your research question needs to be changed, it may drastically affect your finishing time, or when you hope to publish your research. If you don't want to significantly hinder your finishing time, you should come up with a question that is sufficiently jives with your results, and requires only a few experiments that would address your question. However, if your results open the door to a huge un-touched field, or in other words is high impact, then it may not be in your best interest to further pursue this avenue of research. If this is the case your research question needs to reflect this and have a much narrower depth of investigation.

 

This situation changes completely depending on your own circumstances though. If you are just beginning your program and you have hit a "golden egg" so to say, you have several years to further investigate your finding. So I would propose a broader research question that would allow you the flexibility to really investigate. However, as you investigate further and further into your question, results may come back unexpected even after several repetitions that would require you to modify your hypothesis and explanation. This is just the cycle of hypothesis driven research.

 

Several researchers tweak their research questions after a set of data are generated; no researcher can guess 100% the mechanical basis for a particular observation before any experimental input. So back to whether changing the question is a moral dilemma, ethically wrong or tests your integrity, etc. I hope you considered this when answering this question. If you made this into a moral dilemma or a question about integrity, and a member of the panel is such a researcher who has changed their research question in some point in their career, what do you think that particular researcher is thinking when you question his/her integrity?