Home > Statistics Every Writer Should Know > The Stats Board > Discusssion
Chi-Square
Message posted by T. Bennett on September 3, 2000 at 12:00 AM (ET)
I am doing an Applied Research Project using the Chi-Square and I am at a lost. My project is on doing an analysis on the amount of Vietnamese projects that were funded within the last 4 years versus the other ethnic minorities. Should I do a chi-square based on the amount of monies that were funded and the amount of projects?
READERS RESPOND:
(In chronological order. Most recent at the bottom.)
Re: Chi-Square
Message posted by Phil on September 3, 2000 at 12:00 AM (ET)
If you are going to use the "Chi-Square test for Homogeniety" (found in most statistics textbooks), then the table requires "count" data--the number of projects. So a 2x2 table of projects by year and ethnicity could be appropriate.If you were using total dollars funded as a test statistic, then you would use a different kind of analysis (e.g. ANOVA).
Re: Chi-Square
Message posted by JG on September 6, 2000 at 12:00 AM (ET)
Chi-squared is a test of two or more patterns or distributions are similar to each other allowing for statistical variation especially in small samples. A 2 by 2 test for independence essentially asks if to rows or colums are proportionally related - such as whether the proportion of viatnamese projects funded is the same as that of some other group - allowing for statistical variation. The chi-squared test is based on the normal approximation to the binomial and basically asks whether two or more frequency distributions are derived from the same probability distribution.
Your $5 contribution helps cover part the $500 annual cost of keeping this site online.