East Carolina University
Department of Psychology
Karl Wuensch's Statistical Help Page
Use of These Documents
Binomial Effect Size Display
Effect of n1/n2
on Estimated d and rpb
Notational Confusion for d and g Statistics
Odds Ratios -- why I prefer them over
Odds Ratios, Phi, and Base Rates -- phi can be greatly affected by base
rates, odds ratios are not.
Eigenvalues -- origin of the term
Exploratory Factor Analysis -- review of 1999 article by Fabrigar et al.
e-history -- base of the natural log
EX16-11 -- from Howell's fundamentals text, page 333
Analysis, Number of Factors, code for Parallel Analysis and Velicer's
-- many of these letters are used as statistical symbols.
Statistics -- useful links to Internet Resources, University of York
HOWL13-6 -- exercise 13-6 from the Fundamentals text
-- jokes involving statistics
Significant, Simple Main Effects Not -- suggestion on how to deal with
IV-DV -- Independent and Dependent
Kolgomorov, A. N. -- Obituary for the mathematician whose work had an
enormous influence on statistics and other domains as well.
KR20 -- Kuder-Richardson 20
Kurtosis -- discussion of kurtosis
Likert -- what is a Likert scale and how do you
Logit -- pronunciations of the word
LESSONS -- these are my online lecture
notes for my stats classes
Low Power, Low Reliability -- Is it a big problem when results are significant?
Mediation and Moderation
Effects -- Jeremy Dawson
Moderation/Mediation Help Centre -- Paul Jose
Quantpsy.org -- Kristopher
Prevention Laboratory -- David MacKinnon
Mediation and Moderation Analysis -- Facebook Group
SAS Macros -- Yung-Jui Yang
SPSS and SAS
Macros, Andrew Hayes -- check out Process for SPSS and SAS
Meta-Analysis -- links to useful resources
Mindless Statistics -- Gigerenzer
Monty Hall Dilemma-- would you switch doors?
Missing Data -- suggested readings on the topic of how to deal with missing data
Assumptions -- Osborne, J., & Waters, E.
(2002). Four assumptions of multiple regression that researchers should
always test. Practical Assessment, Research & Evaluation, 8(2).
Multivariate Normality -- a few
links to get one started on this topic
NHST-SHIT -- Jack Cohen's wit and a list of
references on significance testing.
NHST-Quiz-- A little True-False quiz to test your
understanding of Statistical Hypothesis Inference Testing
Nonparametrics -- common misconception that nonparametrics have no assumptions
-- comparing your sample mean and variance to
that of a normative group
Normality Assumption -- EDSTAT-L
discussion on checking data for compliance with an assumption of normality
Omega-Squared -- advice on interpreting the omega-squared statistic
One-Tailed NHST -- does it ever make sense to test directional
OPSCAN -- using OPSCAN data sheets in your research
p Values and the American Statistical Association
and the American Statistical Society
p-exact -- why researchers should report exact p values
p values suspiciously distributed -- something here smells like
Pairwise Comparisons -- discussion of pairwise/multiple comparisons procedures, including common misconceptions about them
ANOVA, Half-Tailed Tests -- ANOVA, F,
t, directional tests, and the foolishness of "post hoc" power analysis.
ANOVA2-Followup -- heterogeneity of variance in the factorial ANOVA in Howell (grad) page 404
Bayesian Stats, Brief Intro
Beyond Significance Testing -- my notes from the
text by Rex B. Kline (2004).
Receptivity -- nice example of how to present the psychometrics
associated with the development of a new scale
Canonical -- why is it called "canonical" correlation?
Certificate in Statistics --
From ECU, Psychology
Comparing Correlated, Overlapping Correlation Coefficients -- Steiger's
Comparing Correlated but Nonoverlapping Correlation Coefficients -- the Pearson-Filon statistic and the ZPF statistic
Comparing k > 2 Independent Correlation Coefficients
Comparing Reliability Coefficients --
procedures for comparing one reliability coefficient to another
Confidence Interval for R2 -- putting a confidence
interval on R2
Contingency Tables, Low Expected Frequencies --
Is this a problem or not?
Consulting Client Types
Correlation and Causation -- When does correlation imply causation?
Correlation and Regression, Assumptions
-- fixed versus random X
Data-Help -- suggestions on setting up a data file
Dichotomizing Continuous Variables -- a bad idea: Discussion from Edstat-L.
Effect Size Estimation -- An
overview of this topic
Path-Matrix -- using matrix algebra to get effect coefficients in path analysis
r, Critical Values
Pitman's T -- testing variances with correlated samples
Poems -- Poems about statistics
-- An overview
Probability of Replication -- is this new statistic
of any value?
Quotes about Statistics
-- The average human has one breast and one testicle.
R: Learning Statistics with R
Randomness -- what does it really mean when we
say an event is "random?"
Regression Towards the Mean
Reliability -- a few notes on reliability from Nunnally's text
Resampling Statistics -- a brief introduction to bootstrapping and permutation/randomization tests
Research Wahlberg --
Facebook page for stats/research geeks
Reversal-Paradox -- discussion of the reversal paradox,
also known as Simpson's paradox.
Sample Size Does NOT Affect the Probability
of a Type I Error
Sample Size Required, Finite
Population, Survey with Dichotomous Responses
SAS -- help using the SAS stat pack
Scales-Transform -- nonlinear data transformations
Signif-Testing -- why Frank Schmidt thinks significance testing the devil's work
Simpson -- Simpson's paradox, with bibliography
Skew -- location of mean, median, mode in skewed distributions
Split in Half --Splitting a data file into two
SPSS -- Help using the SPSS stats pack
SPSS2Excel2SAS -- write SPSS data to Excel and then bring into SAS
SS-Type -- different types of sums of squares
Standardized Confidence Intervals -- why not standardize confidence intervals
Statistician as Detective
Stepwise -- stepwise procedures are the devil's work
Student's t Statistic --
why "t" ?
t-CLT -- nonrobustness of the
t-Crit -- t approximated by z and some criticisms of significance testing
Type I Errors -- How frequent are Type I errors in the published literature?
Type-I-II-Errors -- Evaluating the Relative Seriousness of Type I versus Type II Errors in Classical Hypothesis Testing
Type III Errors -- Correctly rejecting the null but incorrectly inferring the direction of effect
Is .039 Larger or Smaller than .05?
Why Standardize? -- when is it helpful to
standardize effect size estimates rather than presenting them in
Proportions -- Pairwise
comparions for three or more independent proporotions
Return to Dr. Wuensch's Statistical Resources Page.
Contact Information for the Webmaster,
Dr. Karl L. Wuensch
This page most recently revised on 21-January-2017.