statsmodels.stats.weightstats.ttest_ind#

statsmodels.stats.weightstats.ttest_ind(x1, x2, alternative='two-sided', usevar='pooled', weights=(None, None), value=0)[source]#

ttest independent sample

Convenience function that uses the classes and throws away the intermediate results, compared to scipy stats: drops axis option, adds alternative, usevar, and weights option.

Parameters:
x1array_like, 1-D or 2-D

first of the two independent samples, see notes for 2-D case

x2array_like, 1-D or 2-D

second of the two independent samples, see notes for 2-D case

alternative{“two-sided”, “larger”, “smaller”}, optional

The alternative hypothesis, H1, has to be one of the following

  • ‘two-sided’ (default): H1: difference in means not equal to value

  • ‘larger’ : H1: difference in means larger than value

  • ‘smaller’ : H1: difference in means smaller than value

usevar{“pooled”, “unequal”}, optional

If pooled, then the standard deviation of the samples is assumed to be the same. If unequal, then Welch ttest with Satterthwait degrees of freedom is used

weightstuple of array_like, optional

Case weights for the two samples. For details on weights see DescrStatsW

valuefloat, optional

difference between the means under the Null hypothesis.

Returns:
tstatfloat

test statistic

pvaluefloat

pvalue of the t-test

dfint or float

degrees of freedom used in the t-test