In today’s episode, Greg and Patrick dig into Confirmatory Composite Analysis, a very clever way to get formative factors and their causal indicators into the traditional structural equation modeling framework, along with any other latent factors and their effect indicators that might already be in the model. Along the way they also discuss full-contact Wordle, …
S5E08 Confirmatory Composite Analysis: Enter the Hexagon Read More »
In today’s episode, Patrick and Greg talk about the challenge of having causal indicators of formative factors within an analytical framework that is historically dominated by effect indicators and latent factors — and the critical importance of getting your arrows right. Along the way they also mention: self help books, habits, Hagrid and the giants, …
S5E07 Perspectives on Formative Factors & Causal Indicators Read More »
In today’s episode, Patrick and Greg talk about fun extensions to the basic confirmatory factor model, including higher order models, bifactor or residualized models, and multitrait-multimethod models. Along the way they also discuss microscope lab, burning ants, substitute teaching, Cool or Creepy?, Monet, Clueless, haystacks, hotdogs, What are you thinking?, pennies and pounds, party like …
In this week’s episode, the first for Season 5, Greg and Patrick explore the very cool world of receiver operating characteristic, or ROC, curves: what they are, how they work, and why we can give partial thanks to Winston Churchill for their existence. Along the way they also discuss advent calendars, lasagna for 8, honey …
S5E01 Receiver Operating Characteristic (ROC) Curves Read More »
In this week’s episode Patrick and Greg talk about the critical and often unmet assumptions underlying the use of measured variables at multiple time points, whether for simple analyses like tests of means or more complex analyses like modeling individual growth. Along the way they also mention: following your blocker, Pascal’s wager, Wikipedia empathy, ham …
In this week’s episode Greg and Patrick talk about exploratory structural equation modeling (or ESEM), a technique drawing upon the strengths of exploratory and confirmatory factor analysis methods and able to be embedded within structural equation models. Along they way they also discuss crab vs. crabs, Most Dangerous Catch, Next Day Blinds, smash-and-grab, sharp pencils, …
In this week’s episode Greg and Patrick explore all the ways they lie about things when they teach, not the least of which is that there are actually no individual growth trajectories estimated in an individual trajectory growth model. They discuss why this is, how individual trajectory estimates can be obtained, and how these might …
S4E14 Growth Trajectory Estimates…What’s My Line? Read More »
In this week’s episode Greg and Patrick spend a wonderful, if not at times awkward, hour talking with Dan Bauer about the genesis, application, and future directions of the what may be the world’s worst acronym: MNLFA, or moderated nonlinear factor analysis. Along the way they also discuss unsolicited help from teenagers, gold stars, acronyms, …
S4E05 Moderated Nonlinear Factor Analysis: A Conversation with Dan Bauer Read More »
In this week’s episode Patrick and Greg talk about partial least squares, a technique that resembles structural equation modeling but with a lot of flexibility, including but not limited to its ability to accommodate both reflective and formative constructs. Along the way they also mention dark Red Bubble, Sheeps Kin, snorting SEM, the Coors Light …
S4E04 Partial Least Squares: Straight Outta Uppsala Read More »
In this week’s episode Greg and Patrick discuss latent means models to conduct group mean comparisons while controlling for measurement error, which gives you more power and more accurate standardized effect size estimates. Along the way, they also mention Aunt Roz, table 8, naughty pigs, crossing the streams, big twinkies, asbestos, It’s a Small World, …
