non-parametric, k related samples

Please use the Cochran-Q Test

Suppose we were interested in the inuence of interviewer riendlinem upon householders' responses in an opinion survey. We might train an interview to conduct three kinds of interviews: [1} showing inter- est. friendlinessT and enthusiasm; (2} showing formality, reserve, and courtesy; and [3} showing disinterest, abruptnem, and harsh formality. The interviewer would be assigned to visit 3 groups of 13 houses and was told to use interview 1 with the rst group1 intervimv 2 to the second group, and interview 3 to the last group. That is. we would obtain 13 sets of householders with 3 matched householders. {equated on relevant tariables} in each set. For each set, the 3 members wouid be designed randome to the 3 types of interviews. Thus, we would have 3 matched samples with 13 members each. Test whether the great dierencee among the three styles of interviews inuenced the number of "yes\" rte-spouses given to a particular item by the three matched groups. Perform pairwise comparisons if signicant dierenees are ob- served. H .3 Set Interview 1 Interview 2 Interview 3 I {I I] [I 2 1 1 3 1 H 4i. {I I} II 5 l I] 6 l l E! 'i" l l I} 3 {I l [I I] 1 [II Ill 1] I] ll 11 1 1 I 12 l 1 1 l3 1 l 14 l l 15 l l 16 l l l l l l nab-ma g... be