Comparing two values for significance

[radycardia]

Looking for a little assistance please. I need to determine whether or not there is a statistical significant difference (p<0.05, with CI of 95%) between two values. I have tried a chi square or t-test but I don't think that works.

Here we go:

Total number of patients: 816

Of the 816 patients, they were broken down into two groups based on blood pressure.

Group A = 27 patients had a blood pressure of 89 or less
Group B = 789 patients had a blood pressure of 90 or greater

Now we looked at the incidence of broken hip bones in these two groups

In group A, 7/27 or 25.9% had a broken hip
In group B, 97/789 or 12.3% had a broken hip

My question is this: is there a significant difference between the incidence of broken bones for group A and group B. ie: Is group A really more likely to have a broken bone. Or, due to my small group A sample size, if the difference not significant.

Once we figure this out, I have another question as I have broken these groups down even further.

Thanks to anyone who can help.

JD

[fed1]

Hello Rady cardia,
I think chi square is just great for this.
Here is a table for your data. Just significant!!
This message board is addictive.


Table of bp by hip

bp hip

Frequency‚
Percent ‚broken ‚not ‚ Total
ƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆ
hi ‚ 97 ‚ 692 ‚ 789
‚ 11.89 ‚ 84.80 ‚ 96.69
ƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆ
lo ‚ 7 ‚ 20 ‚ 27
‚ 0.86 ‚ 2.45 ‚ 3.31
ƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆ
Total 104 712 816
12.75 87.25 100.00


Statistics for Table of bp by hip

Statistic DF Value Prob
ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ
Chi-Square 1 4.3624 0.0367

[radycardia]

Thank you very much for the help.

Is it possible for you to explain how to do this? I see your chart but it doesn't make sense to my ignorant self.

I have many more calculations that I need to do for my project and I would like to learn how to do this myself.

thanks

[TheEcologist]

Thank you very much for the help.

Is it possible for you to explain how to do this? I see your chart but it doesn't make sense to my ignorant self.

I have many more calculations that I need to do for my project and I would like to learn how to do this myself.

thanks

Here's a place to start:

http://udel.edu/~mcdonald/statchiind.html

And here's a nice selection of more tests:
http://udel.edu/~mcdonald/statbigchart.html

[radycardia]

Hello Rady cardia,
I think chi square is just great for this.
Here is a table for your data. Just significant!!
This message board is addictive.


Table of bp by hip

bp hip

Frequency‚
Percent ‚broken ‚not ‚ Total
ƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆ
hi ‚ 97 ‚ 692 ‚ 789
‚ 11.89 ‚ 84.80 ‚ 96.69
ƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆ
lo ‚ 7 ‚ 20 ‚ 27
‚ 0.86 ‚ 2.45 ‚ 3.31
ƒƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆƒƒƒƒƒƒƒƒˆ
Total 104 712 816
12.75 87.25 100.00


Statistics for Table of bp by hip

Statistic DF Value Prob
ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ
Chi-Square 1 4.3624 0.0367

Thanks for your help fred, I can follow part of your post, but I don't know where you get some of your numbers.

I can see the hi = blood pressure above or = to 90. That the first three numbers are the number with a fracture, number without a fracture and the total. But where do you get the next series of three numbers right below that? (I bolded them)

I can follow the same thing for the low, but where are the second set of three numbers coming from?