Don Lynam

Interactions are difficult to detect in field studies as they are typically tiny--very small to start with and made smaller by the joint unreliabilities of the components. Here, we find some but the contribution to explained variance is negligible. Call off the search. It is not worth the effort.

Additive and Interactive Relations of Personality and Cognition With Externalizing Behaviors - Nathaniel L. Phillips, Nathan T. Carter, Kevin M. King, Courtland S. Hyatt, Max M. Owens, Donald R. Lynam... Personality and cognition offer robust frameworks to understand the individual differences associated with externalizing behaviors. However, these literatures h...

𝐀𝐝𝐝𝐢𝐭𝐢𝐯𝐞 𝐚𝐧𝐝 𝐈𝐧𝐭𝐞𝐫𝐚𝐜𝐭𝐢𝐯𝐞 𝐑𝐞𝐥𝐚𝐭𝐢𝐨𝐧𝐬𝐨𝐟 𝐏𝐞𝐫𝐬𝐨𝐧𝐚𝐥𝐢𝐭𝐲 𝐚𝐧𝐝 𝐂𝐨𝐠𝐧𝐢𝐭𝐢𝐨𝐧 𝐖𝐢𝐭𝐡𝐄𝐱𝐭𝐞𝐫𝐧𝐚𝐥𝐢𝐳𝐢𝐧𝐠 𝐁𝐞𝐡𝐚𝐯𝐢𝐨𝐫𝐬 | "Although interaction effects were detected, they were small and practically negligible in their explanation of variance in externalizing behaviors" journals.sagepub.com/doi/10.1177/...

Another great rock and roll show. @hayescarll.bsky.social at the Turntable. Such a great singer songwriter. Terrific band. Awesome venue. I am expanding my travel radius for Carll and looking for another show at Turntable. Check them both out.

Sarah Mullally was named the new archbishop of Canterbury on Friday, becoming the first woman to serve as the spiritual leader of the Church of England, and of the global Anglican community, in the church’s nearly 500-year history. https://wapo.st/3KAojoc

I will try to collect and report answers (if i get any). But I am slightly disorganized and sometimes confused by the structure of bluesky.
2/2

I think about this stuff sometimes. For other open science / “the sky is falling enthusiasts, if you could make a skeptic read earnestly just one article on open science or the credibility crisis, which one would you pick?
1/2

I am glad we were able to chat.

They will write papers (including how tos), build slide decks, and provide materials to interested parties. They are interested in having people aboard but frustrated that they are still having to convince people to get on That's my view.
3/3

The shared excitement sometimes bothers me. People will come in and give talks if you invite them. On whatever you want. Credibility crisis. Preregistration. RRs.They will answer questions.
2/3

I do some of that because I am a snarky guy and I get frustrated. And I have a social media account that let's me scream things. I like to scream things. But the rest (the stuff not on social media) of open science space is pretty friendly in my experience. They/we are excited about it.
1/3

7. Transparency is the only hope for our shifty field (we sometimes conference at the same places, so you know what I mean). 8. I recommend reading Feynman's Cargo Cult Science now and then to help stay the course. TYFCTMTT
5/5

6. Other subdisciplines (especially personality and especially in Europe) are already engaging in these practices with much less hand wringing. So it can be done.
4/5

4. Twitter is not real life. For any person who drags you on Twitter there will hundreds (dozens at least) of people in the field who respect you more. And you set an example. 5. We were all sold a bill of goods when we were taught science is easy.
3/5

2. My early preregistration sucked. You have to sit down with your preregistration when you sit down with your data to see all of the microdecisions you missed. You fix them next time. 3. They got better.
2/5

Just a couple of non-snarky thoughts. 1. Most of us older folks doing this stuff are self-taught. When I started preregistering stuff 10 years ago, I had no idea what I was doing. I just pulled a template and had at it.
1/5

Disappointment abounds on both sides.

You're so mean. Now I definitely won't do the right thing

only for that paper. there will be other papers. I see that as a mulligan. I think your next tee shot counts.

some people are never going to be honest about what they have done to their data. The only question is how many?

I stand corrected.

those menus will also spit out the code if you ask nicely. But we could extend your idea to stats body cameras. They come on as soon as you engage with your data and record continuously until you tug your data in at night.

Unless you are doing analyses by hand or in excel, you (or the program you re using) are/is using code. Every program i have ever worked with is capable of spitting the code it used out.

Everyone is using code. Even with canned programs. I still use SPSS with some frequency. I can easily produce that syntax. Even if I were using only the buttons.

There are no principled arguments against any of these practices. Other fields are already engaging with them at reasonable rates. There is nothing special about our field that prevents us. Except maybe the fragility of our findings and grant money.

"I will give up my researcher degrees of freedom when you pry them from my cold dead hands."

Thanks.

thanks.

It is not that you have not been clear, I just don't think your ideas work in the absence of preregistration. Multiverse analyses, robustness checks, etc. only work when transparent (i.e., preregistered). How else are we to guarantee that you reported all of the analyses you conducted?

As is often the case, the Drive By Truckers have a relevant line: "but they got paid so there ain't nothin' else to think about."

I am not fully convinced that medicine is that much better. But something needs to happen.