This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: /20 QUESTION #3 /2 a) H : The average chicken weight is the same for all three levels of the drug in the chick feed. OR μ 1 = μ 2 = μ 3 where μ i represents the population average chicken weight for the ith dose. OR τ i = 0 where τ i represents the effect (difference between trt mean and overall mean) of the ith dose. H a : The average chicken weight differs for at least one level of the drug in the chick feed. OR Not a i are equal. OR not all τ i = 0. NOTE: ANOVA tests a general alternative hypothesis – it is not testing for an increas compared to the control. That will have to be done using contrasts. /2 b) Clearly from the scatterplot, supplementing the chick food with the drug does appear to increase the a weight at maturity. There appears to be a substantial difference between the control (no drug) and th whereas, there is a lesser increase between the low and the high dose. Note in the plot high dose ap dose – a flaw with the plotting routine. Page 1 of 7 From this plot we can see that the profiles for the different blocks are not parallel but cross each other is either no difference among blocks - indicating that blocking was not an effective tool - or there is a b treatment interaction - which will inflate the error term making it harder to detect differences among tre ANOVA below indicates that blocks are not significant while treatments are – so probably blocking wa effective. However our analysis must be a RCB since that was how the study was designed – reducin 1-way ANOVA is not appropriate, after the fact....
View Full Document
- Winter '09
- Normal Distribution, Multiple comparisons, Dunnett, average chicken weight