Dear Allstat,
I am looking at some blood pressure data and I have noticed that there seems
to be a preference for a particular blood pressure. Is there a significance
test I can carry out on frequency data to test whether a value is
significantly greater / less than its expected value ? The data would be
bell shaped and quite close to normality.
Say if the data looked something like this. (I have made this data up, but
it illustrates how it looks)
Blood Pressure Freq
120 23
122 34
124 45
126 67
128 143
130 67
132 45
134 12
Here we see that the value of 128 seems to be preferred with 143 subjects
recorded with this value. I know we can obtain an expected value, but can we
say that a value is significantly greater / less than its expected value. in
the example I have given here it might have been the case that the expected
value
for the 128 reading was 90 (made up). As 143 > 90 by some margin I want to
determine how large that margin has to be to be significant.
Does anyone know how to do this ?
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|