How to Plan Experiments: Thinking with the GROW Model
"How should I go about planning an experiment?" I am occasionally asked this challenging question at work.
If I know the questioner's situation and their way of thinking to some extent, I can provide advice tailored to them, but I wondered if I could create a general explanation. I used the coaching GROW model, but I designed it so that even those who don't know about coaching can follow the flow and think through it.
Rather than "how to think," it's about "in what order to think"
I am operating on the premise that the questioner has a certain level of knowledge and experience, and that if they take the time, they will come up with certain ideas. If they have some experimental experience from university or elsewhere, it shouldn't be completely zero.
On the other hand, if that point is overwhelmingly lacking, I think a teaching approach where the mentor thinks through the experimental plan together with them and explains the background knowledge is effective.
The text from the next section onwards assumes that this is cleared, and is intended as support in a form that makes it easier for the person to think on their own by breaking the flow of thought into steps.
Goal: What is the purpose of the experiment?
In previous cases, the understanding of what the purpose of the experiment actually is was often vague, so I think this step is indeed important. If answering the following questions helps clarify your vision, it indicates that there was still room to deepen your thoughts on the purpose.
What are the premises or hypotheses you want to verify with this experiment?
What kind of results would you consider a success for this experiment?
If you get those results, what is your next action?
Regarding the third item in particular, whether you can concretely imagine the "next step after the next" seems to be one key point. If you are in a company, I think you are working according to project-based or organizational goals, so it would be good to ask how it can be linked to those major goals.
Resource: What are the resources for conducting that experiment?
The next most common situations are: "Your will to work hard on this amount is wonderful, but you might burn out," or "Since your approach is good, I want you to increase the number of experiments a little more."
In the case of the former, where they are trying too hard, there are cases where limitations such as experimental equipment, the person's throughput, and reagents are not considered. In the case of the latter, where they need a little more, I think there are many cases where the concept of periodic productivity is thin.
I hope that through these questions, it will be a trigger for the person to advance their thinking.
How many samples can you process in one go when conducting this experiment?
How many reagents and samples do you have? Do you have backups for failures?
About how many days will it take? Can you draw up a schedule?
Is there a good balance between the time spent on preparation/planning and the actual work time?
Option: How can you approach this with some ingenuity?
Next, let's discuss ways to improve how you conduct your experiments. While it is sometimes necessary for a supervisor to be present on-site to provide guidance to improve experimental efficiency, it would be ideal if you could learn to think through things on your own whenever possible.
How do you usually go about this? How do you think others might do it?
If you were to prepare the day before, what could you do?
Can you anticipate any potential problems? If they were to occur, how could you respond?
Since there are many cases where people have a craftsman-like attachment to their own methods, I phrase my questions in a mild way so as not to touch upon that directly. In short, it might be best to gently nudge them toward thinking outside the framework they might not normally consider.
Will: So, how will you proceed from here?
Once you have thought things through to this point, I want to give one final push so you can experience yourself starting to take action in a way that is slightly different from yesterday.
Through questions like the ones below, I minimize the span between our dialogue and the start of action. If possible, I think it is best if you can start doing something immediately after our conversation.
Is there anything else you think you should check?
When do you want to start? What can we both do to make that happen?
I would appreciate it if you could let me know when something comes up next. When would be a good time for that?
I tend to press for preparation status and progress reports here, but I have phrased these questions on the premise that the individual wants to do it themselves. I intended to make the questions such that they can act on their own volition, rather than asking, 'When will you provide your next report?'
Summary
There seem to be various thoughts behind the question, 'How should I plan my experiments?' However, having a guideline on the order in which to think about things within the scope of one's own creativity and thinking ability should lead to confidence.
When I look at the young people around me, I notice a tendency to worry about the 'OW' of the GROW model without thinking much about the 'GR.' I believe that if a supervisor provides some support in that area, it will become easier for the individual to think for themselves in the future.
For those wondering what other articles might be available.