afhayes.bsky.social
🇨🇦 Distinguished Research Professor at the University of Calgary. Creator of the PROCESS macro for SPSS, SAS, and R. Director of the Canadian Centre for Research Analysis and Methods. afhayes.com haskayne.ucalgary.ca/ccram/academy 🇨🇦 (he/him)
61 posts
1,797 followers
1,290 following
Prolific Poster
Active Commenter
comment in response to
post
www.narcity.com/why-kitkat-t...
comment in response to
post
Installment #2 on local lettuce. I thank the current geopolitical moment for making me realize I can make a great salad using all Canadian produce.
comment in response to
post
I found this Canadian lettuce at a Safeway in Calgary, perhaps grown in a greenhouse hydroponically. Fresher and tastier than anything I’ve had from the US, in my opinion.
comment in response to
post
If you can’t make it to this, consider learning about these topics and more at Rocky Mountain Methodology Academy in Calgary in July. A trip to Banff is included. haskayne.ucalgary.ca/ccram/academy
comment in response to
post
This course would be great for anyone in industry doing research, and as an in-person course, industry and academic folks in attendance can learn from each other.
comment in response to
post
Find a good supportive publisher (I’m with Guilford Press), write books, say what you want how you want in as many words as you want or need, and feel liberated.
comment in response to
post
• Elimination of the need to specify a model number for certain models
• More modern custom dialog extension (.spe) for the SPSS GUI (the .spd builder file has been discontinued).
4/4
comment in response to
post
• All subsets regression
• Ordinary and heteroskedasticity-consistent errors-in-variables regression to account/correct for random measurement error in mediation and ordinary regression models
• Cluster and stratified bootstrap sampling
• Cluster-robust standard errors and inference
3/4
comment in response to
post
• Partial and semipartial correlations and their squares (as eta-squared, partial eta-squared) as well as Cohen’s f-squared
• Spline regression
• Sums of squares/ANOVA summary table for regression variance components
• Inference for sets of variables
2/4
comment in response to
post
Class started yesterday, but registrations are still being accepted.
comment in response to
post
Not at the moment. I’m not a big believer in this, though there is no harm in doing it and maybe I will add it to PROCESS one day.
comment in response to
post
By the way, for the last couple of years, PROCESS has been able to estimate mediation models with counterfactually-defined effects in the presence of treatment by mediator interaction. Coming in a month or so is the debiasing that results from random measurement error and lots of other new features.
comment in response to
post
Just wait until you get a chance to use the new errors-in-variables routine coming in PROCESS v5. You'll get SEM-like latent variable regression and mediation models to deal with random measurement error in one simple line of code. Unfortunately, not yet available for moderation models.
comment in response to
post
Learn about some of its complexities at haskayne.ucalgary.ca/CCRAM/mediat... you’ll have access to PROCESS v5 before it is released in 2025.
comment in response to
post
I have never regretted the outcome of the work.
comment in response to
post
The second edition is in the works.
comment in response to
post
Writing entire books, however, is a good idea and I have never regretted that.
comment in response to
post
Always seems like a bad idea to me, then I sometimes agree anyway, then pretty much always regret.
comment in response to
post
Faculty and grad students at most Canadian universities are unionized. Not a solution to all problems, and even causes some, but definitely a net plus compared to academia in the US (where i worked most of my career until recently).
comment in response to
post
Here is the proper link: haskayne.ucalgary.ca/ccram/academy