This preview shows pages 1–3. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: Practical Data Analysis I A not uncommon situation: I A client brings you data from a food development study: 3 treatments: I Old “on market” formulation of soup I new formulation, Same salt content as old I new formulation, Reduced salt content I They recruited 25 people. Asked each to taste all 3 soups and provide a numerical score describing “how well did you like this soup”. I Range from 0 (absolutely hate) to 15 (absolutely love) I They ask you to analyze the data. I Focus on trt. means: μ O , μ S , and μ R I They tell you they are interested in: Whether the new formulation is better liked than the old Whether reducing the salt content has any effect Copyright c 2011 Dept. of Statistics (Iowa State University) Statistics 511 1 / 11 Practical Data Analysis How would you analyze the data? (N.B. all analyses account for blocking / pairing of obs. within subject) 1. ANOVA F test of μ O = μ S = μ R , report p-value and trt. means 2. 3 paired t-tests: μ O = μ S , μ O = μ R , and μ S = μ R , report p-values and differences of means 3. above with multiple comparisons adjustment for p-values 4. ANOVA F test, followed by pairwise comparison of means, no mult. comp. adj. 5. ANOVA F test, followed by linear contrasts to estimate μ O- ( μ S + μ R ) / 2 and μ S- μ R . report ANOVA p-value and estimates 6. linear contrasts (above) without F test first....
View Full Document
This note was uploaded on 02/11/2012 for the course STAT 511 taught by Professor Staff during the Spring '08 term at Iowa State.
- Spring '08