topic Re: Polychoric Correlation in Statistical Procedures

Polychoric Correlation

deleted_user — Mon, 15 Jun 2009 13:28:09 GMT

Hello,

I ran a polychoric correlation and recieved the following output.

Polychoric Correlation Value = -0.1791 ASE = 0.0178

How can I interpret the ASE?

Thanks,

Kelly

Re: Polychoric Correlation

deleted_user — Mon, 15 Jun 2009 14:00:52 GMT

Is it the asymptotic standard error of the correlation?

Re: Polychoric Correlation

deleted_user — Mon, 15 Jun 2009 14:30:20 GMT

It is but, pardon the potentially stupid question, what does it signify in terms of the -.17 coefficient I got? Am I to assume that this coefficient is not significant?

Re: Polychoric Correlation

Dale — Wed, 24 Jun 2009 16:44:24 GMT

I would say quite the opposite. The absolute value of the ratio of the polychoric correlation to its asymptotic standard error is 10. Assuming that the ratio follows a normal distribution, the probability that the polychoric correlation is zero in the population from which your sample is drawn is essentially zero.

That said, the correlation is small. The amount of shared variance is on the order of 3%. Whether this is meaningful is a totally different question than whether the correlation is significant.

Re: Polychoric Correlation

deleted_user — Mon, 06 Jul 2009 17:10:16 GMT

Sorry for the late reply - but thanks very much!