Planned Comparisons & Post Hoc Tests

Slides:



Advertisements
Similar presentations
Dr. AJIT SAHAI Director – Professor Biometrics JIPMER, Pondicherry
Advertisements

Chapter Thirteen The One-Way Analysis of Variance.
Analysis of Variance (ANOVA) ANOVA methods are widely used for comparing 2 or more population means from populations that are approximately normal in distribution.
Analysis of variance (ANOVA)-the General Linear Model (GLM)
Chapter Fourteen The Two-Way Analysis of Variance.
Regression Part II One-factor ANOVA Another dummy variable coding scheme Contrasts Multiple comparisons Interactions.
Analysis of Variance (ANOVA) Statistics for the Social Sciences Psychology 340 Spring 2010.
One-Way ANOVA Multiple Comparisons.
POST HOC COMPARISONS A significant F in ANOVA tells you only that there is a difference among the groups, not which groups are different. Post hoc tests.
Lecture 10 PY 427 Statistics 1 Fall 2006 Kin Ching Kong, Ph.D
Research methods and statistics
Comparing Means.
Intro to Statistics for the Behavioral Sciences PSYC 1900
Statistics for the Social Sciences Psychology 340 Spring 2005 Analysis of Variance (ANOVA)
One-way Between Groups Analysis of Variance
Comparing Means.
Introduction to Analysis of Variance (ANOVA)
If = 10 and = 0.05 per experiment = 0.5 Type I Error Rates I.Per Comparison II.Per Experiment (frequency) = error rate of any comparison = # of comparisons.
Quiz 12  Nonparametric statistics. 1. Which condition is not required to perform a non- parametric test? a) random sampling of population b) data are.
Repeated ANOVA. Outline When to use a repeated ANOVA How variability is partitioned Interpretation of the F-ratio How to compute & interpret one-way ANOVA.
ANOVA Chapter 12.
Factorial ANOVA 2 or More IVs. Questions (1)  What are main effects in ANOVA?  What are interactions in ANOVA? How do you know you have an interaction?
T-test Mechanics. Z-score If we know the population mean and standard deviation, for any value of X we can compute a z-score Z-score tells us how far.
ANOVA Greg C Elvers.
Comparing Means. Anova F-test can be used to determine whether the expected responses at the t levels of an experimental factor differ from each other.
Stats Lunch: Day 7 One-Way ANOVA. Basic Steps of Calculating an ANOVA M = 3 M = 6 M = 10 Remember, there are 2 ways to estimate pop. variance in ANOVA:
Chapter 13: Introduction to Analysis of Variance
Confidence Intervals for Means. point estimate – using a single value (or point) to approximate a population parameter. –the sample mean is the best point.
Chapter 11 HYPOTHESIS TESTING USING THE ONE-WAY ANALYSIS OF VARIANCE.
Regression Part II One-factor ANOVA Another dummy variable coding scheme Contrasts Multiple comparisons Interactions.
Statistics 11 Confidence Interval Suppose you have a sample from a population You know the sample mean is an unbiased estimate of population mean Question:
Orthogonal Linear Contrasts This is a technique for partitioning ANOVA sum of squares into individual degrees of freedom.
Between-Groups ANOVA Chapter 12. >When to use an F distribution Working with more than two samples >ANOVA Used with two or more nominal independent variables.
Post Hoc Tests. What is a Post Hoc Test? Review: – Adjusting Alpha Level – Multiple A Priori Comparisons What makes a test Post Hoc? – Many tests could.
Jeopardy Opening Robert Lee | UOIT Game Board $ 200 $ 200 $ 200 $ 200 $ 200 $ 400 $ 400 $ 400 $ 400 $ 400 $ 10 0 $ 10 0 $ 10 0 $ 10 0 $ 10 0 $ 300 $
Statistics for the Social Sciences Psychology 340 Fall 2013 Tuesday, October 15, 2013 Analysis of Variance (ANOVA)
Single Factor or One-Way ANOVA Comparing the Means of 3 or More Groups Chapter 10.
Analysis of Variance (One Factor). ANOVA Analysis of Variance Tests whether differences exist among population means categorized by only one factor or.
Principles of Biostatistics ANOVA. DietWeight Gain (grams) Standard910 8 Junk Food Organic Table shows weight gains for mice on 3 diets.
Education 793 Class Notes Decisions, Error and Power Presentation 8.
Marshall University School of Medicine Department of Biochemistry and Microbiology BMS 617 Lecture 13: One-way ANOVA Marshall University Genomics Core.
PS215: Methods in Psychology II W eek 8. 2 Next Friday (Week 9) Evaluating research, class test First ten minutes of lecture ( ) Please come a.
Chapter 13 For Explaining Psychological Statistics, 4th ed. by B. Cohen 1 Chapter 13: Multiple Comparisons Experimentwise Alpha (α EW ) –The probability.
Psy 230 Jeopardy Related Samples t-test ANOVA shorthand ANOVA concepts Post hoc testsSurprise $100 $200$200 $300 $500 $400 $300 $400 $300 $400 $500 $400.
Quiz 11  Analysis of Variance (ANOVA)  post hoc tests.
Statistics for the Social Sciences Psychology 340 Spring 2009 Analysis of Variance (ANOVA)
Simple ANOVA Comparing the Means of Three or More Groups Chapter 9.
O A post-hoc test is needed after we complete an ANOVA in order to determine which groups differ from each other. o Do not conduct a post-hoc test unless.
ANOVA: Why analyzing variance to compare means?.
Stats/Methods II JEOPARDY. Jeopardy Estimation ANOVA shorthand ANOVA concepts Post hoc testsSurprise $100 $200$200 $300 $500 $400 $300 $400 $300 $400.
Independent Samples ANOVA. Outline of Today’s Discussion 1.Independent Samples ANOVA: A Conceptual Introduction 2.The Equal Variance Assumption 3.Cumulative.
Chapter 11: Test for Comparing Group Means: Part I.
Posthoc Comparisons finding the differences. Statistical Significance What does a statistically significant F statistic, in a Oneway ANOVA, tell us? What.
Six Easy Steps for an ANOVA 1) State the hypothesis 2) Find the F-critical value 3) Calculate the F-value 4) Decision 5) Create the summary table 6) Put.
ANALYSIS OF VARIANCE (ANOVA)
Comparing several means: ANOVA (GLM 1)
Psych 706: Stats II Class #2.
Post Hoc Tests on One-Way ANOVA
Post Hoc Tests on One-Way ANOVA
Planned Comparisons & Post Hoc Tests
Differences Among Group Means: One-Way Analysis of Variance
Psych 231: Research Methods in Psychology
Comparing Means.
Psych 231: Research Methods in Psychology
Factorial ANOVA 2 or More IVs.
Psych 231: Research Methods in Psychology
Analysis of Variance: repeated measures
Psych 231: Research Methods in Psychology
Conceptual Understanding
The Structural Model in the
Presentation transcript:

Planned Comparisons & Post Hoc Tests Comparing Cell Means Planned Comparisons & Post Hoc Tests

Questions What is the main difference between planned comparisons and post hoc tests? Generate numbers (like 0 1, -1 or 1 –1/2, -1/2) to create a contrast appropriate for a given problem. How many independent comparisons can be made in a given design? What is the difference between a per comparison and a familywise error rate? How does Bonferroni deal with familywise error rate problems? What is the studentized range statistic? How is it used?

Questions (2) What is the difference between the Tukey HSD and the Newman-Keuls? What are the considerations when choosing a post hoc test (what do you need to trade-off)? Describe (make up) a concrete example where you would use planned comparisons instead of an overall F test. Explain why the planned comparison is the proper analysis. Describe (make up) a concrete example where you would use a post hoc test. Explain why the post hoc test is needed (not the specific choice of post hoc test, but rather why post hoc test at all).

Planned vs. Post Hoc Planned Comparisons or Contrasts Use instead of overall F test. Planned before the study. Post Hoc or Incidental tests. Use after significant overall F test to investigate specific means. No specific plan before study. Control Comp Tutor Comp Tutor+ Lab Comp Tutor + lab + quiz

Planned Comparisons (1) Population Comparison: Weights are real numbers not all zero. Sum of weights must equal zero. Sample Comparison:

Planned Comparison (2) A1 A2 A3 A4 Source SS df MS F (Data) (3 possible comparisons) A1 A2 A3 A4 22 26 28 21 15 27 31 17 24 18 23 20 25 Comparison A1 A2 A3 A4 1 1/2 -1/2 2 -1 3 Source SS df MS F Cells (A1-A4) 219 3 73 12.17 Error 72 12 6 Total 291 15 (Summary Table)

Sampling Variance of Planned Comparisons The sample comparison is an unbiased estimate of the population comparison. The variance of the sampling distribution of the comparison: Sampling variance will be large when within cells variance is large, the weights are large, and the number of people in each cell is small. Estimated by: We substitute for

Significance Test A1 A2 A3 A4 Source SS df MS F df=N-J; 16-4=12=dfe. 22 26 28 21 15 27 31 17 24 18 23 20 25 Source SS df MS F Cells (A1-A4) 219 3 73 12.17 Error 72 12 6 Total 291 15 df=N-J; 16-4=12=dfe. t(12) =-2.86, p < .05

Significance Test A1 A2 A3 A4 Source SS df MS F df=N-J= 22 26 28 21 15 27 31 17 24 18 23 20 25 Source SS df MS F Cells (A1-A4) 219 3 73 12.17 Error 72 12 6 Total 291 15 df=N-J=

Review What is the main difference between planned comparisons and post hoc tests? Suppose I do a blind orange juice taste test and discover that my means are: Tropicana Florida Fresh Pulpmaster 7.3 5.5 6.4 If my hypothesis is that Tropicana is better than all others, what are my contrast weights?

Independence of Planned Comparisons You can make several planned comparisons on the same data. Some of these comparisons are independent; some are dependent. We want them independent. Two comparisons from a normal population with equal sample sizes in each cell are independent if the sum of the products of weights is zero. With unequal sample sizes, it’s:

Independence (2) One and two are orthogonal; one and three are not. Comparison A1 A2 A3 A4 1 -1/3 2 -1/2 3 1/2 One and two are orthogonal; one and three are not. There are J-1 orthogonal comparisons. Use only what you need.

Choosing Comparisons Usually done on basis of theory. But there are methods to generate all possible orthogonal comparisons. Group 1 2 3 4 5 Comparison 1 -1

Error Rates With 1 test, we set alpha = Type I error rate. With multiple tests, original (nominal) alpha is called the per comparison error rate ( ). With comparisons, we have a family of tests on the same data. Want to know the probability of at least 1 Type I error in the family of tests. Such a probability is called familywise error rate ( ). For independent tests, E.g., 10 tests:

Bonferroni Tests Familywise error depends on the number of tests (K) and the nominal alpha, . Bonferroni’s solution is to set: Suppose we want FW error to be .05 and we will have 4 comparisons. Then Where is an aspiration level. We use the adjusted alpha (.0125) for each of the 4 tests.

Bonferroni Test (2) Use the adjusted alpha (e.g., .0125) for each comparison. Look at the p value on the printout (use .0125 instead of .05). Use a statistical function (e.g., Excel, SAS) if you want to find the critical value. E.g., Excel function TINV says with p=.0125 and df=12, t is 2.93.

Review How many independent comparisons can be made in a given design? What is the difference between a per comparison and a familywise error rate? How does Bonferroni deal with familywise error rate problems?

Post Hoc Tests Given a significant F, where are the mean differences? Often do not have planned comparisons. Usually compare pairs of means. There are many methods of post hoc (after the fact) tests.

Scheffé Can use for any contrast. Follows same calculations, but uses different critical values. Instead of comparing the test statistic to a critical value of t, use: Where the F comes from the overall F test (J-1 and N-J df).

Scheffé (2) Source SS df MS F (Data from earlier problem.) Cells (A1-A4) 219 3 73 12.17 Error 72 12 6 Total 291 15 (Data from earlier problem.) The comparison is not significant because |-2.86|<3.24.

Paired comparisons Newman Keuls and Tukey HSD are two (of many) choices. Both depend on q, the studentized range statistic. Suppose we have J independent sample means and we find the largest and the smallest. MSerror comes from the ANOVA we did to get the J means. The n refers to sample size per cell. If two cells are unequal, use 2n1n2/(n1+n2). The sampling distribution of q depends on k, the number of means covered by the range (max-min), and on v, the degrees of freedom for MSerror.

Tukey HSD HSD = honestly significant difference. For HSD, use k = J, the number of groups in the study. Choose alpha, and find the df for error. Look up the value qα. Then find the value: Compare HSD to the absolute value of the difference between all pairs of means. Any difference larger than HSD is significant.

HSD 2 Grp -> 1 2 3 4 5 M -> 63 82 80 77 70 Source SS df MS F p Grps 2942.4 725.6 4.13 <.05 Error 9801.0 55 178.2 K = 5 groups; n=12 per group, v has 55 df. Tabled value of q with alpha =.05 is 3.98. Group 1 5 4 3 2 63 7 14 17* 19* 70 10 12 77 80 82

Newman-Keuls Layer refers to how many means apart. Layer 4 Layer 3 Group 1 2 3 4 5 1 63 7 14* 17* 19* 2 70 10 12 3 77 4 80 5 82 Layer 4 Layer 3 Layer 2 Layer 1 Same as HSD except the value of q changes with layers. For layer k-1 (here 4), use HSD. For each layer down, subtract 1 from the value of k for the tabled value of q.

Comparing Post Hoc Tests The Newman-Keuls found 3 significant differences in our example. The HSD found 2 differences. If we had used the Bonferroni approach,we would have found an interval of 15.91 required for significance (and therefore the same two significant as HSD). Thus, power descends from the Newman-Keuls to the HSD to the Bonferroni. The type I error rates go just the opposite, the lowest to Bonferroni, then HSD and finally Newman-Keuls. Do you want to be liberal or conservative in your choice of tests? Type I error vs Power.

Review What is the studentized range statistic? How is it used? What is the difference between the Tukey HSD and the Newman-Keuls? What are the considerations when choosing a post hoc test (what do you need to trade-off)? Describe (make up) a concrete example where you would use planned comparisons instead of an overall F test. Explain why the planned comparison is the proper analysis. Describe (make up) a concrete example where you would use a post hoc test. Explain why the post hoc test is needed (not the specific choice of post hoc test, but rather why post hoc test at all).