Topic One: Statistical Analysis (Biology)
What are error bars used for?
To represent range, standard deviation, or other estimates of variability.
an overlap of error bars gives us what
a clue as to the significance of the results
what is a t-test
a statistical test which helps us determine the significance of the difference between the means of two populations
what does the mean not tell us
about the spread of the data
if t is less than cv,
accept h0
why do we reject h0 if t is greater than cv?
as the value of t increases, we become more confident that the results are real and not due to chance. If t is less than the critical value, we are less confident that the difference between the means is significant
what must the sample size do
be large enough to generate reliable data, and large enough to perform statistical tests
how do you calculate sample size
given amount - 2
what is the null hypothesis
h0 "there is no significant difference"
what does the t-test tell us (about probability)
it tells us the probability of two data sets being the same
what does a large overlap of error bars indicate
lots of shared data, results are not likely to be significantly different
what is the standard deviation
measure of the spread of most of the data
the higher value of p, the
more data overlap
what does no overlap of error bars indicate
no (or very little) shared data, results are likely to be significantly different
with the t-test we always start by saying the
null hypothesis
if t is greater than cv,
reject h0
what are two important things to consider when collecting large scale data
sample size and uncertainty/error of measurements
if p=1?
the data sets are exactly the same
if p=0?
the data sets are not at all the same
what is uncertainty
the margin of error in a measurement
if the t-test tells you to accept h0 then
the means are not significantly different
if the t-test tells you to reject h0 then
the means are significantly different
What is the definition of mean?
the measure of the central tendency of the data
the lower the value of p,
the more significant are our results
what is the definition of statistical analysis
the treatment of data collected in scientific investigations
what percent falls within one standard deviation of the mean
68%
how sure do we want to be on a t-test
95%
what percept falls within two standard deviations of the mean
95%
What are error bars?
Graphical representation of the variability of the data
