Question

We conducted 2 independent randomised experiments on mice separately to determine whether exposure to excess UV (ultra-violet) light would cause cancer among young and old mice. The following data were tabulated. ("Healthy" means no cancer, and "Normal" means no excess UV) Young mice Old mice UV Healthy Cancer Total Normal92 Excess855 Total UV Healthy Cancer Total Normal 488 Excess12 Tota500 512 1000 84 596 1096 102 10 95 105 1052 950 96 947 (a) What are the explanatory variable and response variable here? (b) Calculate the relative risk of "Cancer" for excess to normal UV exposure in young mice (c) Calculate the relative risk of "Cancer" for excess to normal UV exposure in old mice (d) Calculate the relative risk of Cancer" for excess to normal UV exposure in all mice combined (no need to adjust for confounding variable) (e) Do you think "Simpson's Paradox" is at play here? (f) If your answer to part (d) is yes, give a plausible reason why this is so. If your answer is no, explain why not. (g) Based on the relative risk concept, do you believe the 2 random samples above suggest there is a practically significant relationship between "Excess" UV exposure and "Cancer", separately for young and old mice? Briefly explain

Answer 1

uth evels ) Nomal (2) Excess (1)Healt ) Canten ツ. Relative Psket "tance, excess t rmmalUv excess to nrmal Uv 7102 102102 Relative eak-192 .Relette Risk in oldl mice

Date abte for all muice Comblmed 442 mo46一ㄧ·← o.nll_ = 0.3612 S22/1102 RRz 0.4437