Question

Help with Stats? I will fan and medal...

For a χ2 test of independence, a researcher computes a χ2 value of 22.5. The two-way table from which this value was computed had 5 rows and 4 columns.
What’s the P-value for this statistic?

A. P > .25
B. .025

Answer 1

Degrees of freedom = (number of rows - 1) * (number of columns - 1) = 4*3 = 12

Use this and the given \(\chi^2\) value to find the p-value.

Answer 2

inst a chi square always a right tail test?

Answer 3

I think usually, yes, but nothing is stopping you from using a two-tailed test. I remember having used a two-tail test when estimating variance.

http://stats.stackexchange.com/a/84329

Answer 4

yeah, i recall that while browsing in some old stat books. wasnt real sure when they would apply tho

Answer 5

Well, at any rate, I don't think this experiment is necessarily two-tailed.

http://www.psychstat.missouristate.edu/introbook/sbk28m.htm

Answer 6

@SithsAndGiggles and @amistre64 -- I am getting B, do you guys agree?

Answer 7

id have to go home and review to be sure :)

Answer 8

Okay. If you have time I would appreciate the help, if not, don't worry.

Answer 9

According to this table: http://sites.stat.psu.edu/~mga/401/tables/Chi-square-table.pdf

you should get a \(p\) value between \(\chi^2_{0.025}\) and \(\chi^2_{0.05}\), which are cutoff values that are equivalent to 0.025 and 0.05, respectively.

Answer 10

the book i reviewed last night said that chi-square test for independence is always right tailed. It is the confidence interval for the standard deviation that got into a 2 tailed setup.

Answer 11

and yeah, its making sense now :) the table doesnt give a solid Pvalue. the actual Pvalue rests within an interval, technology would need to be used to aproximate it better.