Comparison of means test

Slides:



Advertisements
Similar presentations
Quntative Data Analysis SPSS Exploring Assumptions
Advertisements

Multiple Analysis of Variance – MANOVA
Factorial ANOVA More than one categorical explanatory variable.
2  How to compare the difference on >2 groups on one or more variables  If it is only one variable, we could compare three groups with multiple ttests:
Model Adequacy Checking in the ANOVA Text reference, Section 3-4, pg
N-way ANOVA. Two-factor ANOVA with equal replications Experimental design: 2  2 (or 2 2 ) factorial with n = 5 replicate Total number of observations:
“Unplanned comparisons”
Analysis of variance (3). Normality Check Frequency histogram (Skewness & Kurtosis) Probability plot, K-S test Normality Check Frequency histogram (Skewness.
Analysis of variance (2) Lecture 10. Normality Check Frequency histogram (Skewness & Kurtosis) Probability plot, K-S test Normality Check Frequency histogram.
What Is Multivariate Analysis of Variance (MANOVA)?
Biostatistics in Research Practice: Non-parametric tests Dr Victoria Allgar.
Today Concepts underlying inferential statistics
Chapter 14 Inferential Data Analysis
Statistics Idiots Guide! Dr. Hamda Qotba, B.Med.Sc, M.D, ABCM.
Which Test Do I Use? Statistics for Two Group Experiments The Chi Square Test The t Test Analyzing Multiple Groups and Factorial Experiments Analysis of.
Non-parametric Tests. With histograms like these, there really isn’t a need to perform the Shapiro-Wilk tests!
MGT-491 QUANTITATIVE ANALYSIS AND RESEARCH FOR MANAGEMENT OSMAN BIN SAIF Session 26.
Educational Research Chapter 13 Inferential Statistics Gay, Mills, and Airasian 10 th Edition.
ITEC6310 Research Methods in Information Technology Instructor: Prof. Z. Yang Course Website: c6310.htm Office:
Chapter 15 – Analysis of Variance Math 22 Introductory Statistics.
Analysis and Interpretation: Analysis of Variance (ANOVA)
Chapter 12 Introduction to Analysis of Variance PowerPoint Lecture Slides Essentials of Statistics for the Behavioral Sciences Eighth Edition by Frederick.
Psychology 202a Advanced Psychological Statistics November 19, 2015.
Soc 3306a Lecture 7: Inference and Hypothesis Testing T-tests and ANOVA.
Analysis of Variance II Interactions Post-Hoc. ANOVA What question are we asking? On a dependent variable, are several group means different from one.
Statistics for the Social Sciences Psychology 340 Spring 2009 Analysis of Variance (ANOVA)
Chapter 9 Two-way between-groups ANOVA Psyc301- Spring 2013 SPSS Session TA: Ezgi Aytürk.
Analysis of Variance (ANOVA) Scott Harris October 2009.
Chapter 12 Introduction to Analysis of Variance
Multi Group Comparisons with Analysis of Variance (ANOVA)
Chapter 11 Analysis of Variance
Course Review Questions will not be all on one topic, i.e. questions may have parts covering more than one area.
DTC Quantitative Methods Bivariate Analysis: t-tests and Analysis of Variance (ANOVA) Thursday 20th February 2014  
Done by : Mohammad Da’as Special thanks to Dana Rida and her slides 
Statistics for Managers Using Microsoft Excel 3rd Edition
Factorial Experiments
Comparing Three or More Means
Comparing ≥ 3 Groups Analysis of Biological Data/Biometrics
Hypothesis testing. Chi-square test
University of Warwick, Department of Sociology, 2014/15 SO 201: SSAASS (Surveys and Statistics) (Richard Lampard) Analysing Means II: Nonparametric techniques.
Chapter 10: Analysis of Variance: Comparing More Than Two Means
Y - Tests Type Based on Response and Measure Variable Data
Data Analysis and Interpretation
Analysis of Data Graphics Quantitative data
آمار مقدماتی و پیشرفته مدرس: دکتر بریم نژاد دانشیار واحد کرج
Chapter 13: Comparing Several Means (One-Way ANOVA)
STAT120C: Final Review.
Introduction to Statistics
Comparing Groups.
Comparing Several Means: ANOVA
Single-Factor Studies
Analysis variance (ANOVA).
Single-Factor Studies
Statistics in SPSS Lecture 8
Hypothesis testing. Chi-square test
MOHAMMAD NAZMUL HUQ, Assistant Professor, Department of Business Administration. Chapter-16: Analysis of Variance and Covariance Relationship among techniques.
I. Statistical Tests: Why do we use them? What do they involve?
CS 594: Empirical Methods in HCC Experimental Research in HCI (Part 2)
Factorial ANOVA 2 or More IVs.
Parametric versus Nonparametric (Chi-square)
Understanding Statistical Inferences
1-Way Analysis of Variance - Completely Randomized Design
Exercise 1 Use Transform  Compute variable to calculate weight lost by each person Calculate the overall mean weight lost Calculate the means and standard.
NONPARAMETRIC STATISTICS FOR BEHAVIORAL SCIENCE
Analysis of variance (ANOVA)
BUS-221 Quantitative Methods
Univariate analysis Önder Ergönül, MD, MPH June 2019.
Practical Solutions Analysis of Variance
STATISTICS INFORMED DECISIONS USING DATA
Problem 3.26, when assumptions are violated
Presentation transcript:

Comparison of means test Multi-factorial ANOVA as a linear model Hypotheses being tested Interaction effects Post-hoc tests Non-parametric Slide 1

Factorial ANOVA as regression

Factorial ANOVA as regression

Factorial ANOVA as regression How do we code the interaction term? Multiply the variables A x B

Factorial ANOVA as regression How do we code the interaction term? Multiply the variables A x B Gender Alcohol Dummy (Gender) Dummy (Alcohol) Interaction Mean Male None 66.875 4 Pints 1 35.625 Female 60.625 57.500

Factorial ANOVA as regression Gender Alcohol Dummy (Gender) Dummy (Alcohol) Interaction Mean Male None 66.875 4 Pints 1 35.625 Female 60.625 57.500

Factorial ANOVA as regression Gender Alcohol Dummy (Gender) Dummy (Alcohol) Interaction Mean Male None 66.875 4 Pints 1 35.625 Female 60.625 57.500 The cell mean

Factorial ANOVA as regression Gender Alcohol Dummy (Gender) Dummy (Alcohol) Interaction Mean Male None 66.875 4 Pints 1 35.625 Female 60.625 57.500 The cell mean

Factorial ANOVA as regression Gender Alcohol Dummy (Gender) Dummy (Alcohol) Interaction Mean Male None 66.875 4 Pints 1 35.625 Female 60.625 57.500 The cell mean

Factorial ANOVA as regression Gender Alcohol Dummy (Gender) Dummy (Alcohol) Interaction Mean Male None 66.875 4 Pints 1 35.625 Female 60.625 57.500 The cell mean

Two-factor analysis of variance Hypotheses being tested Simultaneous analysis of two factors and measurement of mean response Case. 1: equal replication Terminology: One factor termed A and one factor termed B We use this notation a number of levels in A b is the number of levels in B

Two-factor analysis of variance Hypotheses being tested Researchers have sought to examine the effects of various types of music on agitation levels in patients in early and middle stages of Alzheimer’s disease. Patients were selected based on their form of Alzheimer’s disease. Three forms of music were tested: easy listening, Mozart, and piano interludes. The response variable agitation level was scored.

Two-factor analysis of variance Hypotheses being tested What is (are) the null hypothesis(ese) being tested?

Two-factor analysis of variance Hypotheses being tested Plot these data (means) on a single figure such that cell-level means can be evaluated.

Three-factor analysis of variance Hypotheses being tested Evaluate respiratory rate of crabs (ml O2 hr-1) Factors: Sex Species Temperature

Multiway Factorial ANOVA Hypotheses being tested

Three-factor analysis of variance Hypotheses being tested Three factor ANOVA: HO: Yield is the same in all three Batch sizes HO : Yield is the same in all three Oil amounts HO : Yield is the same in all three Popcorn types HO : The mean yield is the same for all levels of batch, independent of oil amount (Batch X Oil) HO : The mean yield is the same for all levels of Oil amount, independent of popcorn type (Oil X Type) HO : The mean yield is the same for all levels of batch, independent of popcorn type (Batch X Type) HO : Differences in mean Yield among the batch, oil amount, and popcorn type are independent of the other factors (Batch X Type X Oil)

Three-factor analysis of variance Hypotheses being tested

Interaction effects Experiment: we are interested in oxygen consumption of two species of limpets in different concentration of seawater. Factor A is the species of limpet (levels, a) Factor B is the concentration of SW as a function of maximum salinity – 100, 75, and 50 % (levels, b)

Interaction effects Response: respiratory rate of limpets (ml O2 hr-1)

Interaction effects When the two factors are identified as A and B, the interaction is identified as the A X B interaction. Variability not accounted for by A and B alone. Interaction: The effect of one factor in the presence of a particular level of another factor. There is an interaction between two factors if the effect of one factor depends on the levels of the second factor.

Interaction effects Response: respiratory rate of limpets (ml O2 hr-1)

Interaction effects

Interaction effects The response to salinity differs between the two species At 75% salinity A. scabra consumes the least oxygen and A. digitalis consumes the most. Therefore a simple statement about the species response to salinity is not clear; all we can really say is: The pattern of response to changes in salinity differed in the two species.

Interaction effects The difference among levels of one factor is not constant at all levels of the second factor “it is generally not useful to speak of an individual factor effect – even if its F is significant – if there is a significant interaction effect” Zar

Post-hoc tests Tukey test – balanced, orthogonal designs Step one: is to arrange and number all five sample means in order of increasing magnitude Calculate the pairwise difference in sample means. We use a t-test “analog” to calculate a q-statistic

Post-hoc tests Scheffe’s test Examine multiple contrasts: ideas is to compare combinations of samples to each other instead of the comparison among individual k levels.

Post-hoc tests Scheffe’s test Compare the mean outflow volume of four different rivers: 5 vs 1,2,3,4

Post-hoc tests Scheffe’s test Compare the mean outflow volume of four different rivers: 1 vs. 2,3,4,5 Alternatives multiple contrasts:

Post-hoc tests Scheffe’s test Test Statistic:

Non-Parametric tests Violations of the assumptions We assume equality of variance – ANOVA is a robust test. Robust to unbalanced design. How to deal with outliers: use in analysis if they are valid data. Test of normality: Shapiro Wilks Test of equality of variance: Bartletts test.

Non-Parametric tests Nonparametric analysis of variance. If k > 2 Kruskal-Wallis test – analysis of variance by rank Power increases with sample size. If k = 2 the Kruskal-Wallis is equivalent to the Mann-Whitney test.

Non-Parametric tests If there are tied ranks H needs to be corrected using a correction factor C.

Non-Parametric tests If there are tied ranks H needs to be corrected using a correction factor C. ti is the number of tied ranks.

Non-Parametric tests

Non-Parametric tests

Non-Parametric tests

Non-Parametric tests

Non-Parametric tests

Non-Parametric tests

Non-Parametric tests

Non-Parametric tests