Question

True or false?:

1) If the relationship between 2 variables is perfect the standard error of estimate equals 0.

2) F=-1.0 is never possible in ANOVA.

3) If F<1.0 in an ANOVA we don’t even have to bother to look at F critical values table since the result cannot be significant.

4) In an ANOVA, sample size n1=10, n2=12, n3=30, and n4=100, we can use Tukey’s HSD to conduct multiple comparisons.

Answer 1

(1) If the relationship between 2 variables is per fect then y-i, where yi-ith observation, fi predicted value of y when independent variable r En- AP Now, standard error of estimate = Hence the statement is true (2) F is a ratio of two positive quantities i.e sum of square due to ef fect/sum of square due to error so F 20 hence the statement is true (3) Since expected F is always greater than equal to 1 so the statement as true (4) False since the sample sizes must be equal when usinq the studentized range approach