Design an experiment

Summary
You will construct a thought experiment on a topic examining potential sex differences thought to exist between men and women. Your experiment needs to address a specific, narrowly focused question, and describe all the components of the experiment. You are not actually going to do it. But it should be something that COULD be done. This assignment will be revised for a larger written assignment later.
While your experiment can be based on previous research, your experiment should be original.
You should not need to do any outside research.
You should not worry too much on techniques (see below).
Because it is a thought experiment (and not actually done), ethical considerations can be relaxed (for now). You can consider using “model systems” (see below).
Elements of the study and your submission
For this part of the assignment, you will submit a proposal describing the design of a biological experiment that is connected to a topic of potential sex differences that are thought to exist between men and women.
In the proposal, you will describe 1) what you are planning on testing; 2) why you plan on testing it; 3) what you hope your experiment will show; 4) roughly how you are planning on testing it; 5) what your dependent and independent variables are; 6) and all necessary controls.

Considerations
Things to keep in mind: your hypothesis and experiment should be extremely focused and clearly defined; you should have adequate number of replication of the experiment as part of the description; your sample size in your controls and treatment groups should be adequate; make sure your dependent variable actually addresses your hypothesis; make sure you have enough controls to eliminate other alternate explanations for possible results. If you come up with an experiment that, after it were performed, one can’t draw any useful conclusions from it because there were too many unanswered questions, or there were not enough controls, or because it was a too broadly defined hypothesis or topic, then that experiment is not as well crafted as it could be. Often the issue is trying to create an experiment that address too large a question or has too large a scope. Recall that “big” discoveries are really based on anywhere from scores to hundreds of smaller studies and experiments. With respect to controls, it is common for there to be more than one, as a method to address potential variables.

Recommended steps:
• Brainstorm as many ideas as you can come up with on the topic of potential sex differences that are thought to exist between men and women. Use your lecture notes, resources posted on iLearn, or something you’ve read or always been curious about.
• Choose a few topics that most interests you – and picking something of interest to you is key– and brainstorm some experiments that could be done.
• Avoid experiments that are subjective. For example, avoid any experiment that includes data where you might ask the participants to describe how they feel or that has a survey attached.

Asking if someone “feels better” has too many variables. But you could collect data that demonstrates “feeling better” like improved performance of a task.

  •  Of the one you like best, start to craft what the experiment might look like.
  •  Isolate all the conditions you need to control for.
  •  Think of how you can collect measurable data.
  •  Think about what your data might look like, if you were to run the experiment.

About data
Don’t focus on the technical aspects of how to measure something. For example, say you are interested in seeing how a molecule affects the brain, with the hypothesis that the molecule reduces some activity in a region of the brain. In your experiment, you can simply say you introduce a 1x amount of the molecule, a 2x amount, a 3x or 4x amount, etc…. and you then measure activity in the brain using imaging equipment. You can also say your molecule has a tracer (like an isotope, e.g.) so you can follow where it accumulates. Notice you do not have to specify the molecule, the tracer, or the way you are measuring the effect. You can then give the results (e.g. there was no change in the region of the brain, but the molecule also failed to accumulate in that region, so it was either not transported or was metabolized/degraded). Model systems are a good way to do research. Examples include using cells where you might follow a response cells have to a new medication. Or using other organisms and non–human animals – for example, genetic studies often use everything from bacteria and yeast, to worms, fish, mice, etc… Because this is a thought experiment, and because ethics is less a consideration, these are options to think about but aren’t things you need to include.