Despite this, the Mann and Whitney test (1947) has its
limits. With the Monte Carlo methods, methods that
calculate a numerical value by using random or probabilistic
processes, it was shown that the t‐test is most of the time
more powerful than the U‐test. Indeed, this fact remains
whatever the amplitude of the differences between the
averages of the populations under investigation and even if
20
the distributions of these populations do not meet the
criteria of normality (Zimmerman, 1985). On the other hand,
very little statistical power is lost if the Mann‐Whitney U test
is used instead of the t‐test and this, under statistically
controlled conditions (Gibbons and Chakraborti, 1991).
Despite this, the Mann and Whitney test (1947) has its limits. With the Monte Carlo methods, methods that calculate a numerical value by using random or probabilistic processes, it was shown that the t‐test is most of the time more powerful than the U‐test. Indeed, this fact remains whatever the amplitude of the differences between the averages of the populations under investigation and even if 20 the distributions of these populations do not meet the criteria of normality (Zimmerman, 1985). On the other hand, very little statistical power is lost if the Mann‐Whitney U test is used instead of the t‐test and this, under statistically controlled conditions (Gibbons and Chakraborti, 1991).
การแปล กรุณารอสักครู่..
