The value of chi-square is always positive. Why we use the term "one-sided" and not "one-tailed"? To avoid confusion. Zar points out (p.503, 5th edition) that there is one extremely rare situation where the one-sided P value can be misleading: If your experimental design is such that you chose both the row totals and the column totals. With the chi-square test, the one-sided P value is half the two-sided P value. Prism gives you the choice of reporting a one-sided or two-sided P value. It is a very confusing topic, which explains why different statisticians (and so different software companies) use different methods. Also see the section on Fisher's test in Categorical Data Analysis by Alan Agresti. If you want to learn more, SISA provides a detail discussion with references. Most statisticians seem to recommend this approach, but some programs use a different approach. Prism computes the two-sided P value using the method of summing small P values. While everyone agrees on how to compute one-sided (one-tail) P value, there are actually three methods to compute "exact" two-sided (two-tail) P value from Fisher's test. The Fisher's test is called an "exact" test, so you would think there would be consensus on how to compute the P value. How the P value is calculatedĬalculating a chi-square test is standard, and explained in all statistics books. These apparent contradictions happens rarely, and most often when one of the values you enter equals zero. (You can make a similar rule for P values 0.05 with a 95% CI that does not include 1.0. If the P value is less than 0.05, then the 95% confidence interval cannot contain the value that defines the null hypothesis. P values and confidence intervals are intertwined. Why isn't the P value always consistent with the confidence interval? You will interpret the results differently depending on whether the P value is small or large. Altman, published in 1991 by Chapman and Hall.ĭon't forget that “statistically significant” is not the same as “scientifically important”. If there is no linear trend between row number and the fraction of subjects in the left column, what is the chance that you would happen to observe such a strong trend as a consequence of random sampling?įor more information about the chi-square test for trend, see the excellent text, Practical Statistics for Medical Research by D. It is also called the Cochran-Armitage method. The chi-square test for trend is performed when there are two columns and more than two rows arranged in a natural order. If there really is no association between the variable defining the rows and the variable defining the columns in the overall population, what is the chance that random sampling would result in an association as strong (or stronger) as observed in this experiment? The P value from a Fisher's or chi-square test answers this question:
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |