Question

a.   Compare Means of Three of More Groups

Descriptives
X24 -- Likely to Recommend
	N	Mean	Std. Deviation	Std. Error	95% Confidence Interval for Mean		Minimum	Maximum
					Lower Bound	Upper Bound
No Children at Home	159	4.86	1.581	.125	4.61	5.11	1	7
1-2 Children at Home	101	4.68	1.449	.144	4.40	4.97	1	7
More Than 2 Children at Home	89	5.09	1.520	.161	4.77	5.41	1	7
Total	349	4.87	1.531	.082	4.71	5.03	1	7

ANOVA
X24 -- Likely to Recommend
	Sum of Squares	df	Mean Square	F	Sig.
Between Groups	7.839	2	3.919	1.678	.188
Within Groups	808.098	346	2.336
Total	815.937	348

Multiple Comparisons
Dependent Variable:   X24 -- Likely to Recommend
Scheffe
(I) X33 -- Number of Children at Home	(J) X33 -- Number of Children at Home	Mean Difference (I-J)	Std. Error	Sig.	95% Confidence Interval
					Lower Bound	Upper Bound
No Children at Home	1-2 Children at Home	.178	.194	.657	-.30	.66
	More Than 2 Children at Home	-.228	.202	.530	-.73	.27
1-2 Children at Home	No Children at Home	-.178	.194	.657	-.66	.30
	More Than 2 Children at Home	-.407	.222	.189	-.95	.14
More Than 2 Children at Home	No Children at Home	.228	.202	.530	-.27	.73
	1-2 Children at Home	.407	.222	.189	-.14	.95

2. Compare and contrast the means of three or more groups. Are households with more children more likely OR less likely to recommend their favorite Mexican restaurant than are households with fewer children? Explain fully your conclusion. Don't guess. Support your answer by providing the mean that was computed.    

Answer 1

Before starting with the comparision of means of the three or more groups groups we need to first check if the model is significant or not?

For that we use ANOVA table and in that we check the F- test. Keeping significance level as .05 and p- value is .188 (given) we say overall the model is not significant.

But this does not conclude that there cant be relation between two means...So we compare and contrast the means of different groups.

It is as follows:-

1) No children at home and 1-2 children at home. We see the mean difference to be .178. But to make sense from this mean we need to see that p- value of this group. p value which is .657 (given) is more than .05 therefore we say that we accept the null hypothesis and household with more children are not more likely OR less likely to recommend their favorite Mexican restaurant than are households with fewer children.

2) No children at home and more than two children at home.  We see the mean difference to be -.228. But to make sense from this mean we need to see that p- value of this group. p value which is .53 (given) is more than .05 therefore we say that we accept the null hypothesis and household with more children are not more likely OR less likely to recommend their favorite Mexican restaurant than are households with fewer children.

3) 1-2 children at home and more than two children at home. We see the mean difference to be -.407. But to make sense from this mean we need to see that p- value of this group. p value which is .189 (given) is more than .05 therefore we say that we accept the null hypothesis and household with more children are not more likely OR less likely to recommend their favorite Mexican restaurant than are households with fewer children.