Re: Ttest on nonnormal data

As others have said here, the t-test is pretty robust in many
situations. Particularly in social sciences, I can't see why skewness
is a big issue since everything is approximate anyway.

But your understanding of the central limit theorem is incorrect. Your
data may not be normal but you are not looking at the distribution of
the means. In fact, your one survey gives you one estimate of the mean
and one point does not a distribution make!

Your initial statement of "somehow I am not getting satisfactory
answers" worries me. We advise you on what's right to do and if that
doesn't give you the result you have in mind, then you should examine
the results. There are many methodologies out there and if you search
long enough, you will find one that gives you the result you want but
that doesn't make it right.

.

Relevant Pages

  • Re: significance of skewness
    ... The true underlying distribution of this data is unknown to me. ... >> I want to determine whether the skewness of the data is significant. ... > There is a literature on "robust tests of symmetry". ... robust measures of skewness and kurtosis. ...
    (sci.stat.math)
  • Re: issues with statistical test suite from http://csrc.nist.gov/rng/
    ... >> ses is the standard error of skewness. ... >distribution differs from the expected one. ... >Unfortunately I can't know if a generator under test is good or bad if it ... Kurtosis could be ...
    (sci.crypt)
  • Re: issues with statistical test suite from http://csrc.nist.gov/rng/
    ... > ses is the standard error of skewness. ... >>> this conforms to the expected distribution, ... with a proper test for the goodness of fit you are able to ... That final KS p-values seems really useless calculated that way, ...
    (sci.crypt)
  • Re: issues with statistical test suite from http://csrc.nist.gov/rng/
    ... ses is the standard error of skewness. ... In this case the skew was to the left. ... >conversation about testing the generators. ... not with respect to the expected distribution. ...
    (sci.crypt)
  • Re: test of normality
    ... I am not an expert statistician, but I NEVER made a mistake alike. ... skewness is a measure of the asymmetry of the probability distribution of a real-valued random variable. ... a distribution has positive skew if the higher tail is longer and negative skew if the lower tail is longer (confusing the two is a common ...
    (sci.stat.math)
Loading