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

Chi-Square and Multiple Response?
Message posted by Dan Woodlief (via 206.27.60.196) on June 26, 2001 at 4:41 PM (ET)

I am hypothesizing that there are differences between the responses of teachers from various grade levels on a multiple response question (which three program attributes they prefer). The grade level response is not set up as multiple response, but the other question is. I do have the data from the multiple response question organized under discrete variables as either "1" or "0." Can I perform chi-square analysis to determine independence of each of the multiple response choices against (individually) the variable for grade? Should I use something else instead?


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

Re: Chi-Square and Multiple Response?
Message posted by Peter (via 131.193.141.96) on June 26, 2001 at 5:04 PM (ET)

could you clarify with a sample data example? is your data set up like this?: grade level, multiple response #1, multiple response #2, multiple response #3. sample data entry: 5, 0,1,1


Re: Chi-Square and Multiple Response?
Message posted by dan woodlief (via 206.27.60.196) on June 27, 2001 at 8:34 AM (ET)

The data is set up as (1) Grade variable - either K2, 3-5, or 6-8 in a single column. (2) Program Variables - "1" or "O," depending on whether it was indicated as one of the top three most important out of a long list. Each program attribute is set up as a separate variable in my SPSS file, with either "1" or "O;" in other words, multiple columns. I do have a variable that combines them as well - e.g., AttribB, AttribC, AttribM, etc., but there may be too many combinations to use any statistics with the combined column. From just looking at the crosstab results (expected versus observed values and just the percentage differences between the grades), it is apparent there are correlations, but I want to back it up with chi-square or some other statistic. Plus some relationships are easier to see than others without the statistics.


Re: Chi-Square and Multiple Response?
Message posted by dan woodlief (via 206.27.60.196) on June 29, 2001 at 9:07 AM (ET)

Any answers?



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!