Question

An industrial psychologist is investigating the effects of work environment on employee attitudes. A group of 20 recently hired sales trainees were randomly assigned to one of four different “home rooms” – five trainees per room. Each room is identical except for wall color. The four colors used were green, blue, gray and red. The psychologist wants to know whether room color has an effect on attitude, and, if so, wants to compare the mean attitudes of the trainees assigned to the four room colors. At the end of the training program, the attitude of each trainee was measured on a 60-point scale (the lower the score, the poorer the attitude).

Data from the sample are provided in the following ta

	green	blue	gray	red
group
a	47	55	33	41
b	59	49	49	43
c	50	53	46	46
d	56	54	54	35
e	44	48	40	52

1. Conduct a one-way ANOVA with Room Color as your treatment effect (level of significance = .10). Clearly mention your null and alternate hypotheses, calculations and your conclusion. [Show your code in “R Code” section. Show the answer in “Answer” section. Include the null and alternate hypotheses and the phrase “accept null hypothesis” or “reject null hypothesis” along with a justification in a few sentences in “Comments” section.]
2. Conduct a randomized complete block ANOVA with Group as your blocking effect (level of significance = .10). Clearly mention your null and alternate hypotheses, calculations and your conclusion. [Show your code in “R Code” section. Show the answer in “Answer” section. Include the null and alternate hypotheses and the phrase “accept null hypothesis” or “reject null hypothesis” along with a justification in a few sentences in “Comments” section.]

Answer 1

Anova: Single Factor
SUMMARY
Groups	Count	Sum	Average	Variance
green	5	256	51.2	38.7
blue	5	259	51.8	9.7
gray	5	222	44.4	66.3
red	5	217	43.4	39.3
ANOVA
Source of Variation	SS	df	MS	F	P-value	F crit
Between Groups	292.2	3	97.4	2.52987	0.09392	3.238872
Within Groups	616	16	38.5
Total	908.2	19

Ha: means are all not equal

p-value = 0.0939 < alpha (0.10)

we reject the null hypothesis

So answer is At alpha=0.10, it can be said that at least one color has a different mean