Selecting Statistical Procedures Graziano and Raulin Research Methods

Selecting Statistical Procedures n n Selecting statistical analysis procedures is part of the design

Describing Your Research 1. 2. 3. What is its level of constraint, dependent variables,

Flowchart Structure n Descriptive Statistics (up and right) – These should always be done

Functional Flowchart (Student Resource Website) n n Functional flowchart walks you through the process

Secondary Analyses n n Defined as analyses that do not directly test the hypothesis

Caveats and Disclaimers n n Flowchart is a teaching device only NOT every statistical

Robust Statistical Procedures n n A statistical procedure is robust to an assumption IF

Summary n Select statistical procedure as part of design process – Statistical procedures selected

Slides: 10

Download presentation

Selecting Statistical Procedures Graziano and Raulin Research Methods: Appendix D This multimedia product and its contents are protected under copyright law. The following are prohibited by law: (1) Any public performance or display, including transmission of any image over a network; (2) Preparation of any derivative work, including the extraction, in whole or in part, of any images; (3) Any rental, lease, or lending of the program. Copyright © Allyn & Bacon (2010)

Selecting Statistical Procedures n n Selecting statistical analysis procedures is part of the design process Based on the characteristics of the study Use the information about the study to follow the flowchart to arrive at decisions about statistics to use Need not memorize the flowchart Copyright © Allyn & Bacon (2010)

Describing Your Research 1. 2. 3. What is its level of constraint, dependent variables, and level of measurement of each dependent variable? What are the independent variables, what type (between-subjects or withinsubjects), and how many levels? What is (are) the research hypothesis (hypotheses)? Copyright © Allyn & Bacon (2010)

Flowchart Structure n Descriptive Statistics (up and right) – These should always be done n Other statistics (down the left side) – Manipulation check (if relevant) – Correlations n n Relevant if it is correlational research Often included in higher constraint research to identify potential confounding variables – Testing Mean Differences n The most common inferential statistical procedure Copyright © Allyn & Bacon (2010)

Functional Flowchart (Student Resource Website) n n Functional flowchart walks you through the process of selecting a statistical test Links to the manual and computerized computation tutorials Copyright © Allyn & Bacon (2010)

Secondary Analyses n n Defined as analyses that do not directly test the hypothesis of the study Three types of secondary analyses – Post hoc analyses: Such as specific means comparisons in ANOVA – Secondary analyses to explain results – Data snooping: Exploring the data for clues that might form hypotheses for later research Copyright © Allyn & Bacon (2010)

Caveats and Disclaimers n n Flowchart is a teaching device only NOT every statistical procedure is included – Only the most common (e. g. , testing mean differences) n Procedures suggested by flowchart, although the correct procedures to use, are NOT the only correct procedures for that situation Copyright © Allyn & Bacon (2010)

Robust Statistical Procedures n n A statistical procedure is robust to an assumption IF the assumption can be violated without distorting results Evaluated with Monte Carlo procedures – Computer simulations of the effects of various situations on the effectiveness of a statistic n Evidence suggests that assumption violations have little effect IF your sample sizes about equal in each group Copyright © Allyn & Bacon (2010)

Summary n Select statistical procedure as part of design process – Statistical procedures selected before data collection – Selection based on study characteristics – Functional version of the flowchart is on the website, with links to how to compute each statistical procedure Copyright © Allyn & Bacon (2010)