Question 1
If there is significantly more variability between samples than within samples, this tends to be indicative of:
◦
use of an inefficient study design.
◦
actual mean differences attributed to the factor.
◦
incidental differences attributed to sampling error.
◦
a need to resample.
Question 2
In 1995, a survey of 109 countries was conducted. The survey recorded (among several other variables) the dominant religion of the country as well as its population. It was thought that perhaps religion was somehow correlated with population, so a one-way ANOVA was conducted on the data. An ANOVA summary from the analysis appears below:
Source | df | SS | MS | F | p |
Factor | 9 | 14,101.2 | 1566.8 | 16.95 | 0.000 |
Error | 99 | 9149.7 | 92.4 |
|
|
Total | 108 | 23,250.9 |
|
|
|
Does it appear that there is a difference in mean population size for the various religions? Select the proper conclusion. (Useα= 0.05.)
◦
p-value <α so fail to rejectH0; mean population appears to differ for at least two of the religions.
◦
p-value <α so rejectH0; mean population appears to differ for at least two of the religions.
◦
p-value <α so rejectH0; religion does not appear to be correlated with population.
◦
p-value >αso fail to rejectH0; mean population appears to differ for at least two of the religions.