This study was done in an AP Statistics Course with relatively small sample sizes. The validity of such studies must always be questioned. Please keep this in mind if you use or report the results of this study.
The question this study was trying to answer was which human
reflex was quicker: sight, or sound. The experiment involved a
metal washer (10g), which is tied to a piece of string. The string
had two black marks: one near the washer and another 30cm from
that point. While sitting at a table, the subject attempted to
stop the string with his or her right hand before it had traveled
the 30 cm. There were two trials in my experiment: in the first,
the subject just watched the string, and in the second, the subjects
eyes were closed, and the subject reacted solely to the sound
of the experimenters voice. The proportions of interest were the
proportion of students who stopped the string before the 30 cm
cut-off had been breached, in both the sight and hearing tests.
This study was an experiment because two treatments were imposed
on the subjects in order to observe their response.
The study tried to prove that the proportion of student's sampled
would have a quicker reflex, when reacting to movement (sight
reflex), then they would when reacting to noise (hearing reflex).
The null hypothesis stated that there was no difference between
the proportion of student's who stopped the string before the
cut-off (30 cm), for both the sight and hearing reflex tests.
The alternative hypothesis stated that the proportion's of students
who stopped the string before the cut-off (30 cm) for the sight
test was greater then the proportion of students who stopped the
string before the cut-off during the hearing test.
The population of this study was the Buckingham Browne and
Nichols Schools Upper School student body. The Upper School contained
the grades 9through 12. The school was a coed, private/prep school,
in Cambridge, Massachusetts. It had a student body of approximately
450 student's (p = 450). In the preliminary section, 30 students
were selected, not randomly, to perform the test. Their distances
for each test were recorded, and then used to create a cut off
point of 30 cm (described in Sampling section of the report --
averages) for the actual experiment. Then, using a MiniTab program
(which gave each student in the school a number from 100 to 550,
and then put those numbers in a random, no repeating order) an
SRS of 37 students (n = 37)was selected from the Upper School
student body for the study.
In the sight test, the proportion of subjects that stopped
the string before the 30 cm cut-off was (22/37) . For the hearing
test, the proportion of students that made the cut-off was less,
(15/37). Using a one sided, large sample z-test for two proportions,
the sample turned out to be statistically significant at the 0.10
level (p=.0526, n= 37), which gave some evidence to reject the
null hypothesis in favor of the alternative hypothesis, that the
sight reflex is quicker the sound reflex.
One lurking variable was the reaction between the experimenter's
voice and the releasing of the string (in the hearing test). Perhaps
in a follow up study, one could use a computer program that would
make a beep, while releasing the string at the same time. This
could prevent any bias from a delay between the experimenters
voice and his/her releasing of the string. Also, the 30 students
in the preliminary section were not chosen at random, which could
have an effect on the overall study. In a follow up study (given
enough time) one could select these 30 at random from the student
body, then test them, and then take an SRS from the rest of the
student body without the previous 30 included for the actual experiment.
This could prevent bias coming from the preliminary subject taking
the test twice.