Suppose a controlled experiment will test 5 different insulation materials (for keeping the contents of a box cold). For each material, the investigator will take 8 independent measurements of the temperature of the contents (for a total of 40 measurements). Then, he will compute the relevant F-statistic. If these different insulation materials have about the same effect (no real difference in average temperature) and the ANOVA assumptions apply, then how often could we expect such an investigator to get an F-statistic bigger than 4? Hint: use the pf function in R, or the custom function from the Whalen-Stat-13 R script.
o Practically 0% of the time.
o About 1% of the time.
o About 2% of the time.
o About 3% of the time.
o About 4% of the time.
o About 5% of the time.
o About 6% of the time.
o About 7% of the time.
o About 10% of the time.
o About 15% of the time.
o About 20% of the time.
o About 25% of the time.