World Statistics Day was yesterday, but we’re celebrating all week long! This celebration means acknowledging the impact statistics has on our world. Who is your favorite statistician? Share with us who they are and why they top your favorites list.
Mar 29, 2016 1:39 PM
| Last Modified: Mar 26, 2018 10:06 AM
We learned how to make even better hard-boiled eggs. (Photos by Caroll Co)In my post last week, I discussed our latest eggsperiment with hard-boiled eggs – and now it’s time for the results!
As a reminder of the factors, we had hard-to-change factors:
Cooking Start (Hot/Cold)
Cooking Time (10 minutes, 12 minutes, 14 minutes)
Salt (0 tsp, ½ tsp, 1 tsp)
Vinegar (0 tbsp, ½ tbsp, 1tbsp)
And easy-to-change factors:
Egg (brown, white)
Cooling Method (cold water/ice water)
The response is the number of seconds it took my wife to peel the egg. The peeling order was randomized and had no effect when added to the analysis.
Before I present the model, if a picture is worth a thousand words, which peeled eggs would you prefer?
Which bowl of peeled eggs would you prefer?
Just as in the first eggsperiment, if there’s one thing to take away from the results, putting the eggs into already boiling water (the hot start) instead of starting with cold water produced the nicest (and easiest to peel) eggs.
Looking at the main effects model, only the cooking start was significant. There does seem to be more to the data when we look a bit deeper. For example, examining peel time vs. cooking start, and overlaying the amount of salt suggests that there’s an interaction between cooking start and salt (notice that higher salt has lower peel times with a hot start, but higher peel times for the cold start).
Interactions involving cooking start are also the most natural place to look for significant interactions, since the main effect is so large. Sure enough, the interaction with salt (and cooking start*vinegar) is significant, and my parameter estimates are as follows:
Not so easy to see what’s going on? Try this interactive HTML to see the results in a Profiler. I've also put the data set on the JMP User Community. Other than the obvious effect of cooking start, the biggest thing I noticed is that salt and vinegar have different results depending on the cooking start (due to the interactions). If I’m wanting to concentrate on where I observed the best results, I want to focus on the hot cooking start. It appears as if the addition of salt helps reduce the peel time for the hot cooking start, although the effect is only marginally significant analyzing the hot start subset of the data. Although it wasn't significant, I also found it interesting that cooling the eggs by running cold water over them reduces the peel time. This was also observed in the first eggsperiment – if there is a difference, I would have thought that the ice bath would be better. Maybe next time, there will be a few more eggs to get a better sense.
This eggsperiment reaffirmed that using the boiling start is the way to go for our household. I’m willing to concede that I’m missing something in my methodology with the cold start, but the boiling start has been consistent in delivering easy-peel and well-cooked eggs. In the past, I’ve also been known to forget to remove the eggs from heat with the cold-start method until long after the pot has been boiling, so the boiling start is more forgiving in that respect.
We will also be adding salt with the boiling start method since it appears to be helpful with peeling cooked eggs. And, for the same reason, we will continue with our practice of running cold water over the cooked eggs.
Any suggestions for another experiment (eggs or otherwise)? The peeler has also told me that if I were to do another eggsperiment, she would much prefer easy-to-peel eggs. So for the eggs option, it’s in my best interest to keep on the boiling start path (I’ve heard steaming is a good way to go too…). Thanks for reading!