Presentations

Psychometrics from the Ground Up

Measurement Scales and variables

Nominal, Ordinal, Interval, and Ratio scales

Frequency Distributions

How to visualize which values are most frequent in a distribution

Probability Density Functions

Functions that tell us which values are more likely to occur

Expected Value and the Meaning of the Mean

An introduction to expected value and how it is related to the concept of the mean of a variable.

Expected Value and Variance

How variance is calculated

The Normal Distribution and the Central Limit Theorem

Sum of the Reasons Variables Are Normally Normal

Skewness: Lopsided variability

An introduction to skewness and its relationship to variability.

Kurtosis (Beyond Peakedness)

An introduction to kurtosis and how it is not just a measure of peakedness.

Standard Scores (and Why We Need Them)

An introduction to z-scores, stanines, stens, scaled scores, T scores, and index scores

Covariance

Covariance is rarely useful by itself but is indispensable for many statistics.

Other Assessment Videos

Emotional Intelligence and CHC Theory

How might emotional intelligence relate to psychometric models of intelligence?

Misunderstanding Regression to the Mean

What regression to the mean is and how it is often misunderstood. Examples are provided as to how it is applied to IQ and the death penalty.

Taking latent variable models seriously

Applying latent score estimates to individuals

Specific cognitive processing weaknesses are rarely full explanations for academic deficits

In which I confess to having had a longstanding misunderstanding about diagnosing learning disorders.

Two kinds of cognitive ability hierarchies

Some latent variables represent “hierarchical abstractions.”

A Taxonomy of Influences on Ability Tests

Some influences are test-specific whereas others are shared across tests. Some influences are transient whereas others are stable. Some influences are relevant to the construct of interest whereas others are irrelevant.

Within-composite differences: Why measures of the same ability differ?

There are a number of lesser-known reasons that two tests that are intended to measure the same ability might differ substantially.

Within-composite differences: Do large subtest score differences invalidate composite scores?

It is often asserted that composite scores should not be interpreted when the scores that make up that composite are discrepant. I show here why this is not typically true, depending on what the composite score is used for.