Dear all,
While I was reading some papers, I have come up with a statistical question
concerning the T-test and Chi-square test. For clarifying, let me please assume
that a researcher was interested in examining whether or not there is a significant difference between a two spot speed samples (assume n1=n2=25) have been taken on two motorways. In other words, does road1 speed sample mean really differ from
road2 speed sample mean?
Now the standard tests to examine the null hypothesis (Ho: no difference
between them) are either the T-test or Mann-Whitney test (assuming independent
samples) depending on the both samples normality and variances homogeneity
issues.
The question is; can I use Chi-Square test here to test the Ho? This could be
done by, for example, assuming the first speed sample measurements as the
(observed values) while the second speed sample mean will be the (expected
value). So, again, is this statistically acceptable?.Further, what is the
difference between this one and those above?
Could somebody suitably experienced help me please?
Kind Regards and Many Thanks for Your Time.
Firas
[log in to unmask]
***********************************************************************
|