OUTLINE FOR CHAPTER
12
Post Hoc and Planned Comparisons
 Introduction
 The ambiguity of a significant F and the purpose of
post hoc comparisons
 Two features of planned comparisons
 The relative popularity of planned and post hoc comparisons
 Post Hoc Comparisons
 Definition and purpose
 Terminology
 Synonyms for "post hoc"
 "Contrast" and "comparison"
 The "omnibus" Ftest
 "Pairwise" and "nonpairwise" comparisons
 Test procedures frequently used in post hoc analyses
 The 6 tests used most often by applied researchers
 Fisher's LSD
 Duncan's multiple range test
 NewmanKeuls test
 Bonferroni
 Tukey's test
 Scheffe's test
 "Liberal" vs. "conservative" procedures
 Comparing other groups vs. a control condition
 Dunnett's test
 Tamhane's test
 The null hypotheses of a post hoc investigation
 Presentation of results
 Summarize results in a passage of text
 Three different ways to summarize results in a table or
figure
 Attaching letters to means
 Notes using "<" and ">" symbols (and
maybe commas as well)
 Attaching letters to bars in a bargraph
 The Bonferroni
procedure as a post hoc technique
 Planned Comparisons
 How these comparisons can be examined directly without reference
to the omnibus F
 Test procedures used to make planned comparisons
 Pairwise and nonpairwise planned comparisons
 Comments
 Terminology
 A priori
 One degreeoffreedom Ftest
 Orthogonal
 Assumptions
 The researcher's choice of test procedure
 Statistical significance vs. practical significance
 Other test procedures
