SPT Shows How Science Works
BY RICK BONNEY
Please cite this Page as:
Bonney, R. 1994. SPT Shows How Science Works. Birdscope, Volume 8,
Number 4: 7.
When we at the Cornell Lab of
Ornithology began planning the Seed Preference Test (SPT), we had a recurring nightmare.
What if no one signed up? What would we do with 10,000 copies of our participant's packet?
As it turned
out, we needn't have worried-over 17,000 of you enrolled.
But it never
occurred to us that the birds might not show. SPT participant Eliza S.
Wolfe of Kalamazoo, Michigan, summed up her experience:
I have to admit, it's some
sort of shame
I threw several parties to which nobody came.
I saw them fly, I heard them sing
But nobody ever took anything!
Indeed,
thousands of you had trouble attracting birds for this experiment. But you
learned an important lesson-that scientific experimentation is fraught with difficulty,
and that the path along the scientific process is often bumpy, and frequently circuitous.
The experiment
itself was simple. Participants placed three identical pieces of
cardboard on the ground in a triangular formation and measured one-half cup of seed onto
each piece, providing a birdseed buffet. Then participants waited for birds to come, so
they could see which types of seeds the birds visited most often.
Unfortunately,
some people waited...and waited...and waited.
Said Kathleen
Ennis of Naples, Florida, "My regulation seed has been royally
ignored by everything in my yard with feathers. Just putting out the seed is enough to
cause them to leave the vicinity!"
Rosalie Dunbar
of Dracut, Massachusetts, came even more to the point. Four
hours of observations produced no birds and the following conclusion:
"Cardboardphobia!"
Honest, folks
– it worked for us here in Ithaca. When birds flocked to the pieces
of cardboard we used in pilot tests at the Lab, we concluded that we'd found a simple,
no-cost feeder setup.
The experiment
also worked for nearly 5,000 of you who sent in completed data
forms.
Why did it
work in some places but not others? We're still sifting through your
letters for clues, but here are some early thoughts.
First, birds
seem to prefer feeding in familiar locations. Therefore, people who
began their experiments by taking down their regular feeders, or allowing them to run dry,
seemed more successful at attracting birds to their experimental seeds.
Second, birds
don't like to venture too far from safety. Our directions called for
placing the pieces of cardboard at least 10 feet from cover, to be sure that birds feeding
on the cardboard were attracted to the seed rather than a shrub or tree growing nearby.
But 10 feet may have been too far.
Third, the
cardboard itself may have been a problem. During our pilot tests, we
used brown cardboard ripped from the backs of notepads. Perhaps other colors, or
different textures of cardboard, frightened birds away.
Several of you
have asked why we called for cardboard instead of hanging
feeders. We didn't want to require participants to purchase three identical feeders just
to conduct our experiment – and the test feeders do have to be identical, otherwise
an apparent preference for one type of seed could really be a preference for a certain
type of feeder.
Some of you
also asked why you couldn't place the test seed directly on the
ground. To be sure that birds were visiting certain types of seeds rather than the
locations of those seeds, we asked you to change the positions of the seeds on each
day you performed the test. We were concerned that if food were placed directly on the
ground, different types of seeds might get all mixed up.
Others have
asked why they couldn't leave the different types of seed on the
ground and see which type disappeared first. Good idea-but how would we know
which kinds of birds (or squirrels, or chipmunks, or dogs) ate which seeds?
Given all
these difficulties, and understanding your frustrations, let us make one
thing clear: No one who attempted this experiment failed. Nobody washed out
as a scientist, even those of you who failed to attract birds. Instead, you learned an
important lesson-that the scientific process is a bumpy road, and that experimental
techniques must undergo constant modifications.
"Failed" approaches to data collection plague professional scientists
constantly.
Seldom does an experiment work right the first time. Most researchers define their
procedures again and again before they work properly, although this aspect of the
scientific process is seldom discussed and infrequently taught.
So, in the
full spirit of the experimental process, we'd like all of you who think you
"failed" (and even those who succeeded) to have another shot. This will
accomplish
three goals: it will give more people the chance to collect data; it will show whether the
food preferences of birds change from year to year; and it will provide more data for
those species, such as the Inca Dove, for which our results are so far inconclusive.
We're going to
change the procedure slightly. First, the cardboard will be optional. Also, you'll
accustom your birds to your experimental setup for several days before you start
your watches. During the experiment we'll suggest that you
remove other sources of food. We'll make some other changes as well, dictated partly by
your comments as we continue to read them. In this experiment you have become the
scientists, and you now have the best ideas for improving the data-collecting process.
If you
participated in this experiment last year, we really appreciate your help. To
those whose experiments worked smoothly-we know how hard you tried. We hope
you'll take our tips for success and try again. And to those who haven't participated at
all-we hope you'll join us this year as the world's largest science experiment continues.
–Rick Bonney
Return |
