Strawberry Firefly
Posts: 2
Joined: Tue Dec 13, 2005 7:03 pm

Crickets and Colors of Light

Postby Strawberry Firefly » Tue Dec 13, 2005 7:17 pm

Hello, my project is about what light colors attract the most crickets, I used 50 crickets and used 8 different colored lights, white, green, blue, purple, red, yellow, pink, and no light (darkness). I have finished collecting the data but I need to know how to analyze it most accurately. Any help you can give would be greatly appreciated, thank you.

whatarush3244
Posts: 11
Joined: Wed Sep 14, 2005 4:58 pm

Postby whatarush3244 » Fri Dec 16, 2005 7:34 pm

Hello, You can start by researching the sensory receptors of crickets. There will most likely be a lot of jargon (special language used by professionals in a specific field, or work) that will be difficult to understand. Your science book can help you out a lot with that.

here are some key terms that you can use to start researching your topic (don't be scared by big words...you can understand it!):
-mechanosensory receptors in the cricket cercal sensory system
-cricket senses
-ecology

here are some sites:
http://animals.about.com/od/grasshopperscrickets/
http://jeb.biologists.org/cgi/content/full/204/14/2423
http://www.jneurosci.org/cgi/content/full/19/5/1771

also, look up online encyclopedias...these may help you!

good luck!

EDS
Former Expert
Posts: 67
Joined: Thu Nov 18, 2004 4:23 am

Some additoinal comments.

Postby EDS » Sun Dec 18, 2005 9:10 pm

Hope you don't mind me jumping in with a few additional comments.

Strawberry Firefly,

How you analyze your data is going to depend on what data you've collected. We may need a more detailed description of the experiment in order to offer appropriate advice. (Telling us what grade level and math background you have is also a good idea. Data analysis can mean a lot of different things!)

I imagine that you've probably got a list of the number of crickets that were attracted to light of each color in each trial. One thing that you'll almost certainly want is to calculate the average (mean) number of crickets that were captured by each color over all the trials. A quick google search or a high school level statistics book will show you how to do that.

One question your reader is going to ask is, "is the difference between each color real, or could it have just been a coincidence?" There are many different ways to investigate that. (There are people who spend their whole lives thinking about the best way of doing so!) What you decide to do will depend on your own math/science background and what your teachers expect.

The simplest approach is to make a plot that shows the results of all the different trials grouped by color. That way both you and the reader can decide whether the differences between colors seem significant compared to the scatter from one run to the next in the same color. Depending on your grade level and what your teacher expects, that may be all the analysis required.

There are also more quantitative ways to try to approach the question of whether any trends you see are real. I'd be happy to offer some detailed tips to get you started if you're interested in doing so.


On a subject only tangentially related to data analysis - you might want to think a bit about what it means for a light to be colored when looking at your data.

For example, white light is made up from light from a wide spectrum of different colors all mixed together. If you stick a blue filter in front of a white light, then you're cutting off all the non-blue light and letting most of the blue light through.

One could imagine a couple different scenarios. If crickets were only attracted to the blue light, then one might expect both the white and the blue to attract them equally.If instead they actively disliked red light and weren't attracted to any other kind of light, then they might go toward blue and black equally but not toward red or white.
(I've no idea whether either of these are true, but those are the sorts of possibilities you could consider.)

Finally, consider whether anything other than color might be different about the lights. Is the total amount of light coming from each light source the same? If not, then it might be the brightness that affects the cricket response rather than color. (It might be kind of hard to measure that. But, at least mentioning it as a possible alternative explanation might be a good idea, unless you can argue that it isn't important.)


Best,
Erik
--
Erik Shirokoff
Science Buddies
Ask an Expert Program

Strawberry Firefly
Posts: 2
Joined: Tue Dec 13, 2005 7:03 pm

Postby Strawberry Firefly » Wed Jan 04, 2006 8:19 pm

Thanks for the help, I'll try to be a little more specific, I'm in ninth grade and I've already calculated the mean, I need to know what kind of graph would be best to use. Thank you.

thetrans1ent
Posts: 31
Joined: Wed Sep 14, 2005 3:35 pm

Postby thetrans1ent » Sat Jan 07, 2006 3:05 pm

In addition to the mean, you may want to give a measure of spread. If for each color the distribution is relatively symmetrical, consider using standard deviations. If not, consider using IQR.

For the graph, one possibility would be to do a curve: # of crickets vs. color wavelength.

Hope that helps.


Return to “Preparing for the Science Fair”