RobertNiles.com
About Robert
Reporting Help
Finding Data on the Internet
Questions for Candidates
Stats Lessons
Mean
Median
Percent
Per capita
Standard Deviation
Margin of Error
Data Analysis
Sample Sizes
Stats Tests
Help Board
Bookstore


A Mathematician Reads the Newspaper

Read about reporters who blew it when trying to write about stats. A mathematician leads you through the tricks, citing contemporary examples.
More information
-->

Home > Statistics Every Writer Should Know > The Stats Board > Discusssion

How to calculate this problem
Message posted by Dr.Sankar (via 152.163.201.56) on November 9, 2001 at 3:38 AM (ET)

In a hospital during last year 394 patients were treated and 30 of them developed complications and during this year 192 patients were treated and 12 of them developed complications.How to compare this two and determine which is statistically significant and better.

1.By chi square analysis
2.How to calculate P-Value in this problem
3.Is there any other methods.

Please help me.


READERS RESPOND:
(In chronological order. Most recent at the bottom.)

Re: How to calculate this problem
Message posted by Tomi (via 154.32.142.194) on November 9, 2001 at 10:14 PM (ET)

In a hospital during last year 394 patients were treated and 30 of them developed complications and during this year 192 patients were treated and 12 of them developed complications.How to compare this two and determine which is statistically significant and better.


Although Jack has given you another method, it is possible to use chi squared analysis.

What you do is create a contingency table:

[ ]#[Year1]#[Year2]#[Total]
[ Complications]#[ 30]#[ 12]#[ 42]
[No Complications]#[ 364]#[ 180]#[544]
[ Total]#[ 394]#[ 192]#[ 586]

The expected values for each cell in the table can be calculated from the row and column totals:
{Year1 Complications} = 42*394/586
{Year2 Complications} = 42*192/586
{Year1 No Complications} = 544*394/586
{Year1 Complications} = 544*192/586

to give the following values:

[ ]#[Year1]#[Year2]#[Total]
[ Complications]#[ 28.2]#[ 13.8]#[ 42]
[No Complications]#[365.7]#[178.2]#[544]
[ Total]#[ 394]#[ 192]#[ 586]

A quick glance at the expected values shows that they are very similar to the observed values, indicating that there is probably no difference between the years. However, we need to work out the X squared statistic:

X squared = sum {(observed - expected)^2} / expected = 0.110 + 0.225 + 0.008 + 0.017 = 0.361

2*2 contingency table has degrees of freedom df = 1.

p score can be calculated in Excel by {=CHIDIST(0.361,1)} which gives 0.548

This p value is much larger than the 0.05 = 5% usually taken as a significance level.

Oops! Should have stated at the beginning:

Null hypothesis: No of complications independent of Year.

Alternative hypothesis: No of complications dependent on Year.

Conclusion: No of complications is independent of Year.



Your $5 contribution helps cover part the $500 annual cost of keeping this site online.

Niles Online Sites:RobertNiles.comTheme Park InsiderViolinist.com

RobertNiles.com™, the site, content and services 咀opyright 1996-2002, Robert Niles.
All rights reserved. Questions? Comments? Read my Privacy Policy, or E-mail me!