Doing Science . . .
Science is a process . . . a way of thinking.
In reality, it is little more than being logical. Science is sometimes
circuitously defined as 'anything that can be studied by the scientific
method'. You have probably been introduced to the scientific method
already many times in your science career, yet surprisingly few people
can properly design and carry out an experiment. But that's what you'll
do in this exercise!
The process of this method is to:
So . . . what hypothesis would you like to test?
Let's not make it too difficult. After all, I don't care as much about
the question as about developing skills in experimental design. Your hypothesis
should be a statement which can be answered, "True or False".
form a testable hypothesis based on these observations;
design a rigorous experiment to test the hypothesis;
collect and statistically analyze the data;
determine whether your hypothesis is supported or not;
publish your results to allow evaluation of your
findings by others in your field.
How about . . .
Men are taller than women.
How would you test this hypothesis? First you
must state your null hypothesis, Ho ('H
sub zero'). What would it be for this example? Write it below.
Whenever testing an hypothesis, you must also
provide at least one viable alternative hypothesis which you will accept
if the null hypothesis is not supported. What will your alternative to
Now, plan the experiment which will test the hypothesis you've chosen.
Here are some of the points that you must consider:
The Importance of Collecting Random Samples
How many subjects should I use?
How will I choose the subjects to be used?
How will I analyze the data once collected?
What kind of results will it take to convince me that my hypothesis is
accepted or rejected?
You are a fishery biologist who has been assigned
the job of determining the average weight of Largemouth Bass in Lake SheaHaley.
You don't have any idea how many fish there are in the lake. How can you
calculate an average?
How many must you collect to get an accurate average
measurement? In statistical-ese, this is phrased, "What sample size do
I need to get statistically significant results?".
How will you collect the fish? If you have
hundreds from which to choose, which one(s) will you measure? Here
are a couple of scenarios. Determine which of these would give you the
a) A bass fishing club holds a tournament and brings you their
catch from the day.
Although they may not look like fish, a bucket
of rocks will substitute for bass. There are 100 rocks of various sizes.
You have neither the time nor manpower to weigh them all. How many will
you do? How will you choose which ones you will weigh? (For instance, if
you chose a sample size of 10, which ten will you chose?
b) Using a shoreline seine, you venture out as deep as you can walk
and catch as many fish as you can on the way into shore.
c) Using an electrofishing boat, you fish for 10 minutes, saving all
bass you collect.
d) Using rotenone, you poison the lake, killing all the fish in it.
Let's see who comes up with the best method
of data collection, i.e., who gets closest to the correct answer!
Instructions for Lab Reports
(Note: This is a summary of materials presented in
McMillan (2001). Consult that fine text for detailed instructions.)
Writing lab reports will provide you with some
experience in analyzing and reporting the results of lab and field studies.
The general format will be like that of a scientific paper for publication
in a journal. For an example, see any journal, e.g. Ecology. Because
we are not doing original research and have only a limited library to back
us up, in reality these write-ups will be at best a hybrid between a lab
report and a scientific publication.
The paper should be organized into major sections:
Title, Abstract, Introduction, Methods and Materials, Results, Discussion,
and Literature Cited.
As concisely as possible, summarize the entire
paper. In only a few sentences, describe your objectives, results, and
important discussion and conclusions. Although it is often published as
the first section of a paper, you may want this to be the last section
This section should be short and to the point,
1) a statement of the subject (what you did),
2) some orientation of the report within the field (relate your work
to other, similar, published accounts), and
3) the purpose of the work (why did you do it?).
Methods and Materials
Included in this section should be a description
of the methods and materials used in sufficient detail that another worker
can repeat your procedures. Statistical procedures used and a description
of the study area should also be included here.
Results should be clearly presented in the
form of tables and figures. In the text of this section, important results
are pointed out and trends in the data are noted. Do not include raw data
here; if they are especially important they may be included in an appendix.
Trends and relationships important to the Discussion are often presented
here first. All tables, charts, figures, and graphs must be titled and
organized in space-saving form. All of these must be specifically
referred to in the text.
This section should contain an interpretation
of the results, including their relevance to related published material,
their significance, and general theories or hypotheses supported or refuted
by the data.
Science has its own writing style and its
method of citing literature may differ from those used in other fields.
When writing scientific literature footnotes are generally not used.
If you want to cite a work in the text, it is normal to use one of the
The full reference appears in the Literature Cited
section at the back of the paper. The style for journal articles, books,
and electronic sources differ. Use the proper form.
Darwin (1859) did not say that man arose from monkeys. Or,
Nowhere in his works does Darwin claim that man came from monkeys (Darwin
Author's name. Publication date. Title
of article. Journal name. Volume #: page numbers.
e.g. Girnish, T. J. 1982. On the adaptive significance of leaf height
in forest herbs. Am. Nat. 120: 353-381.
Author's name. Copyright date. Title.
Publisher, City of publication. # of pages in the book.
e.g. Smith, R. L. 1995. Ecology and Field Biology, 5th Edition.
Harper & Row, NY. 835 p.
Use common sense. As always, provide sufficient
information for your reader to look up the original material. If possible,
include Author's name. Publication date. Title of article. Journal name.
URL (internet address e.g., http://www.keystone.edu) and the date you accessed
William S. Gaud. 1999. Sexual Dimorphism.
September 2, 2002.
Note: Literature Cited includes only publications you have directly
cited. It is not a listing of all the books you examined.
null hypothesis - the condition which would
be expected if there is no effect present.
rotenone - an organic poison extremely
toxic to fish; derived from plants of the chrysanthemum family.
testable hypothesis - a tentative idea
which can be disproved by scientific experimentation.