Hi,

I'm a new PhD student and got a bit of an alarming wake up call at my last supervisory meeting. I'd planned to do 3 clear cut experiments in my first year, and four for the second. I wanted to treat each experiment like an individual dissertation... thorough 8,000 word write ups... extensive literature searches for each and a real knowledge of what I wanted to investigate and find out until....

My supervisor suggested I submit around 3-5 experiments for ethical approval next month...! How on earth can I set five experiments up thoroughly so that they are well designed, well thought etc etc in that time? How can I complete extensive lit reviews and have a real sense of what I am setting up?

Then my second supervisor added that he had completed 20 experiments throughout his PhD.. but only 9 were included in the thesis. To me, this seemed like a pretty big waste of time - 11 sets of data collection not used.

I think they think I should run as many experiments as possible and abandon the ones that don't work. I do not know whether I can do this. At least, not to the standard I would want my work to be at. And if you as submitting multiple experiments to ethics at once- how can you feed the results of one experiment into the other...?

I guess it comes down to quantity or quality. I was just hoping for some advice really. I'm a psychology student and are aware that different subjects require different numbers of experiments. I'm having difficulty knowing how to move forward..

Any advice appreciated :) As you can tell I'm new to this!

Stephanie, I think this must vary from field to field, but if your studies are linked it is hard to see how submitting all for ethics can be best practice given that results from one may influence the design of others.
I am doing a pilot and main study involving questionnaire sampling of patients at presentation and one month later. If the main study confirms the results of the pilot a follow up qualitative study involving structured interview is planned. The main study will take 12 months to recruit the required population sample so doing lots of these is not credible in the time frame...
It is interesting how so many of us here have quite fundamental questions about the proceses we are engaging in - or should that be frightening?

Hey Stephanie! I am doing a PhD in psychology as well (clinical) and I'm just doing one huge study then breaking the results down into about 5 chapters. I've never heard of anyone doing so many separate experiments before in psych, I'm guessing you must be in cognitive psychology or something? Do you need NHS approval or just departmental approval? Either way, for me it was really important to have a thorough understanding of the literature before designing my study and applying for ethical approval. I know some people who write their lit review right at the end of their PhD but I think it's really important to have a good insight before you get to that stage, so I would advise against rushing too much and making sure you are being careful and thorough. The last thing you want is to get loads of ethics applications in and then realise you've made a mistake and need to make an ethics amendment or something. Maybe have a chat with your supervisor about how you would rather set things up (i.e. maybe just do one or two studies at a time so you can concentrate on what you're doing). Lit reviews themselves are very time-consuming, I don't envy you having to write that many! Good luck with it, KB

Mine is in psychology too, learning/cognition. 20 experiments sounds like a ridiculous amount of experiments, especially if only 9 were included. Did they have a small sample size? Were some of the experiments pilots? If so then that would make their total seem artificially large.

Alternatively they could be counting the groups they ran in each experiment as experiments in their own right. I ran 7/8 experiments over 3 years, with two very small pilots. If I counted the individual groups from each experiment as one then I'd hit the ~20 mark. Regardless its a bizarrely high number of experiment to say you have carried out. From my experience, and those of others in the department the average amount of experiments in a thesis is nowhere near that high.

The way of planning that you've mentioned seems more sensible to me. And closer to what I did. The planning and ethical approval was done for one experiment, and I had inklings as to maybe one or two experiments that could be done afterwards depending on the results. When the results came in the next experiment was approved and ran, and so on. My second supervisor was from outside of psychology and this approach puzzled him a little because in his field it's possible to have a clear idea from almost the very start of where you'll end up and what experiments you should run to get there.

As for the absolute ideal number of experiments you should aim for...well, just as many as it takes. Not very helpful I know :( but it's possible in some cases to run a couple of experiments and have enough data and results to make up a thesis, and in other cases it takes more. I remember my supervisor telling me that a thesis doesn't have to be doorstop size, nor does it have to change the world. It just needs to provide evidence that you can do research, that is publishable and that it is, in some way, original.