In: Statistics and Probability
Here is the start of the SAS code to address this
question:
DATA READING;
INPUT GROUP $ WORDS @@;
DATALINES;
X 700 X 850 X 820 X 640 X 920
Y 480 Y 460 Y 500 Y 570 Y 580
Z 500 Z 550 Z 480 Z 600 Z 610
;
Is there a significant difference between the mean reading score
among the groups?
Assuming you answer "yes" in the previous question, between what groups was there a significant difference in means?
The means of the 3 groups are:
= 786, = 518, = 548
And the std deviations are:
sx = 113.93, sy = 54.037, sz = 58.05
The 2-sample t-statistic for samples X and Y =
= [ - ] / sqrt[(sx2/n1) + (sy2/n2)]
4.752 > t0.975, 8 = 2.262 (critical 2-sided t-score for 5% significance and 8 degrees of freedom)
Hence, we can reject the null hypothesis that the means are equal at the 5% significance level. So, there is a significant difference between means of X and Y.
Similarly, for X and Z, t-statistic = 4.162 > t0.975, 8 = 2.262. Hence, there is a significant difference between means of X and Z as well.
Now, for Y and Z, t-statistic = -0.846 > t0.025, 8 = -2.262 (we look at the left tail probability as the t-statistic lies towards the left of the mean 0). Hence, we cannot reject the null hypothesis that there is a significant difference between means of Y and Z at the 5% significance level.
Hence