One of the critical skills required for any scientist is the ability to consistently design and carry out successful experiments. There are of course many variables that can cause an experiment to fail, but there are some basic steps that, if taken routinely, can increase your chances of success every time. After many failed, and successful, experiment attempts, here is my recipe for experimental success:
1. Define your objective. Why are you doing the experiment and what are the expected results? A good experiment will tell you something, even if you get negative data. Make sure to include all necessary controls!
2. Find the best method to get the job done. With so many protocols out there, it can be overwhelming to try and narrow it down to just one. After taking into account the cost of the reagents, whether someone you know has done it before, and how widely the technique has been used, in the end, you just have to pick one and go with it!
3. Write out the protocol. This step will help you identify all the reagents you will need for your experiment, and help you gauge the time that you will need to carry out the procedure. Additionally, it is much easier to follow a protocol that you have written yourself, rather than one that was written by someone else.
4. Obtain all necessary reagents in advance, and make sure they are all in good condition. You don’t want to be in the middle of your experiment when you suddenly run out of reagent X. As in cooking, freshness is the key. Don’t take your chances with old or expired reagents.
5. Prepare a timeline. Try to realistically estimate how long each procedure will take, and if you’re not sure, double the time that you think it will take. Don’t try to cram too much into one day – you are more likely to make errors when you are rushing.
6. Do the experiment. Prepare everything you can in advance, and start EARLY. Make sure you have booked the necessary equipment ahead of time, and avoid interruptions. The protocol you wrote out in Step 3 should make this step a lot easier! If multiple people are involved in the experiment, make sure everyone knows their role ahead of time.
7. Record everything. Write down any deviations from the protocol, and any difficulties you experienced carrying out the experiment. This will help you make better sense of the data and troubleshoot the experiment if necessary.
8. Analyze the results. Hopefully the experiment turned out as you expected, and the data have lead you to the next step of the project. If not, it’s not the end of the world! Go back to your notes and find out where the experiment could have faltered. Talk to people who have experience with the technique to help you troubleshoot, and don’t be afraid to ask around for advice! If things always worked the first time, they wouldn’t call it “research”.
Got any tips of your own that you’d like to share?