Variables and set up

[gpederson]

Hello,

I am an elementary teacher. I would like some advice on behalf of my student. She wants to determine if people can tell the difference between red velvet and chocolate cake. Her hypothesis is, "If I give them red velvet cake they will think it is chocolate cake". Her procedure is to give samples of both red velvet and chocolate cake, and then ask the participant what the cake tastes like, and which they prefer. (Participants are blindfolded.)

The difficulty I am having is figuring out the control, and if this is a proper set up. It is her assertion that people seem to like red velvet cake better just because it has red food coloring in it, so it seems that there should be both a visual sample and a blind sample.

Can you advise please?

Thank you so much,
Gail Pederson
Harbor Montessori School

[Ultra]

Mr. Pederson:

Thank you for using the Science Buddies fora.

It is hard to consider this a true experiment, as your student is not applying any sort of treatment to a test group. By definition, an experiment almost always asks: If I change (independent variable), how will it affect (dependent variable)?

I do agree that the control group should be allowed to see the cake.

If one wanted to do an observational study (not an experiment), one could have two groups of subjects: one that prefers chocolate cake, and one that prefers red velvet. You could then proceed to test how well subjects can differentiate by taste alone in order to correlate appearance with opinion. This is not an experiment, and therefore needs no control group.

As a side note, the experiment seems to assume that red velvet cake IS chocolate cake, but with food coloring added. This is not necessarily the case. A red velvet cake (in theory, anyway) has a different type of chocolate in it, along with other differences (some recipes call for vanilla and coffee, etc.).

I hope that you find my answer satisfactory. If you have any further questions (or if my answer is insufficient), please do not hesitate to continue this thread.


Take care,
- Ultra

[gpederson]

Thank you for such a prompt reply. It was helpful.

We have agreed that she will have each participant taste samples A and B with visual, then taste samples A and B blind. Adding to this she will add a "third blind sample", a Cake C, which will actually just be either Cake A or Cake B repeated.

So the hypothesis will be, "If I change the ability to see the sample, there will be no preference for one sample over another".

Thank you,
Gail