Protzko
@protzko.bsky.social
Experimental Psychologist, social perceptions, metascience. I study the assumptions you make. Here just trying to share good research
https://sites.google.com/view/assumptionlabccsu/home
https://sites.google.com/view/assumptionlabccsu/home
Me thinking I like reading #philosophy until I get hit with a sentence like this.
Seriously, who writes this, slaps it, and says that's good that's not going anywhere?
P. S. this is a footnote calling to another footnote.
Seriously, who writes this, slaps it, and says that's good that's not going anywhere?
P. S. this is a footnote calling to another footnote.
January 22, 2026 at 1:16 PM
Me thinking I like reading #philosophy until I get hit with a sentence like this.
Seriously, who writes this, slaps it, and says that's good that's not going anywhere?
P. S. this is a footnote calling to another footnote.
Seriously, who writes this, slaps it, and says that's good that's not going anywhere?
P. S. this is a footnote calling to another footnote.
Here, we start the process towards global random sampling. We can't fully get there, but we can make steps. Our framework:
Cluster random sampling, proportional weighting, BTS, nonresponse bias correction.
osf.io/preprints/ps...
with @miguelsilan.bsky.social @grinschglsandra.bsky.social
#psych
Cluster random sampling, proportional weighting, BTS, nonresponse bias correction.
osf.io/preprints/ps...
with @miguelsilan.bsky.social @grinschglsandra.bsky.social
#psych
January 21, 2026 at 2:14 PM
Here, we start the process towards global random sampling. We can't fully get there, but we can make steps. Our framework:
Cluster random sampling, proportional weighting, BTS, nonresponse bias correction.
osf.io/preprints/ps...
with @miguelsilan.bsky.social @grinschglsandra.bsky.social
#psych
Cluster random sampling, proportional weighting, BTS, nonresponse bias correction.
osf.io/preprints/ps...
with @miguelsilan.bsky.social @grinschglsandra.bsky.social
#psych
When running an ANOVA (say, 3-groups), F is the omnibus test.
Translated to OLS, 2 dummy variables, the change in R2 is the omnibus test.
BUT, what if analyzing as a Structural Equation Model (in the Figure below)?
Is there something that matches the Omnibus F? Or closely approximates?
#stats
Translated to OLS, 2 dummy variables, the change in R2 is the omnibus test.
BUT, what if analyzing as a Structural Equation Model (in the Figure below)?
Is there something that matches the Omnibus F? Or closely approximates?
#stats
January 17, 2026 at 1:33 PM
When running an ANOVA (say, 3-groups), F is the omnibus test.
Translated to OLS, 2 dummy variables, the change in R2 is the omnibus test.
BUT, what if analyzing as a Structural Equation Model (in the Figure below)?
Is there something that matches the Omnibus F? Or closely approximates?
#stats
Translated to OLS, 2 dummy variables, the change in R2 is the omnibus test.
BUT, what if analyzing as a Structural Equation Model (in the Figure below)?
Is there something that matches the Omnibus F? Or closely approximates?
#stats
Can you make people feel wealthier than their peers?
In two direct replications, we answer: maybe?
But if you can’t it does not transfer to things subjective SES correlate with.
Our new work with Lily Green &
@samanthajamesbrown.bsky.social
#psych #meta
osf.io/preprints/ps...
In two direct replications, we answer: maybe?
But if you can’t it does not transfer to things subjective SES correlate with.
Our new work with Lily Green &
@samanthajamesbrown.bsky.social
#psych #meta
osf.io/preprints/ps...
January 12, 2026 at 2:12 PM
Can you make people feel wealthier than their peers?
In two direct replications, we answer: maybe?
But if you can’t it does not transfer to things subjective SES correlate with.
Our new work with Lily Green &
@samanthajamesbrown.bsky.social
#psych #meta
osf.io/preprints/ps...
In two direct replications, we answer: maybe?
But if you can’t it does not transfer to things subjective SES correlate with.
Our new work with Lily Green &
@samanthajamesbrown.bsky.social
#psych #meta
osf.io/preprints/ps...
Fuck! Swearing may make you a little bit stronger.
In 2 preregistered studies, swearing every 2s increases static strength vs. saying a neutral word.
From Richard Stephens & colleagues
psycnet.apa.org/fulltext/202...
#psych #exercise
In 2 preregistered studies, swearing every 2s increases static strength vs. saying a neutral word.
From Richard Stephens & colleagues
psycnet.apa.org/fulltext/202...
#psych #exercise
December 22, 2025 at 2:06 PM
Fuck! Swearing may make you a little bit stronger.
In 2 preregistered studies, swearing every 2s increases static strength vs. saying a neutral word.
From Richard Stephens & colleagues
psycnet.apa.org/fulltext/202...
#psych #exercise
In 2 preregistered studies, swearing every 2s increases static strength vs. saying a neutral word.
From Richard Stephens & colleagues
psycnet.apa.org/fulltext/202...
#psych #exercise
Corrections to misinformation work & may increase perceptions of truth & feelings towards the target of the misinfo!
For 1-2 weeks at best.
From @jhchae.bsky.social, Groeling, & @hyunjinsong.bsky.social
academic.oup.com/joc/advance-...
#psych
We need to see how long our interventions last!
For 1-2 weeks at best.
From @jhchae.bsky.social, Groeling, & @hyunjinsong.bsky.social
academic.oup.com/joc/advance-...
#psych
We need to see how long our interventions last!
November 25, 2025 at 1:33 PM
Corrections to misinformation work & may increase perceptions of truth & feelings towards the target of the misinfo!
For 1-2 weeks at best.
From @jhchae.bsky.social, Groeling, & @hyunjinsong.bsky.social
academic.oup.com/joc/advance-...
#psych
We need to see how long our interventions last!
For 1-2 weeks at best.
From @jhchae.bsky.social, Groeling, & @hyunjinsong.bsky.social
academic.oup.com/joc/advance-...
#psych
We need to see how long our interventions last!
Fasting for 8-48 hours shows no major effect on cognitive abilities.
Longer fasting shows even less cognitive effects, but there may be some small 8-hour effects. Therefore, lunch to lunch is probably best for a 24hour fast.
From Bamberg & @davidmoreau.bsky.social
psycnet.apa.org/psycarticles...
Longer fasting shows even less cognitive effects, but there may be some small 8-hour effects. Therefore, lunch to lunch is probably best for a 24hour fast.
From Bamberg & @davidmoreau.bsky.social
psycnet.apa.org/psycarticles...
November 24, 2025 at 3:02 PM
Fasting for 8-48 hours shows no major effect on cognitive abilities.
Longer fasting shows even less cognitive effects, but there may be some small 8-hour effects. Therefore, lunch to lunch is probably best for a 24hour fast.
From Bamberg & @davidmoreau.bsky.social
psycnet.apa.org/psycarticles...
Longer fasting shows even less cognitive effects, but there may be some small 8-hour effects. Therefore, lunch to lunch is probably best for a 24hour fast.
From Bamberg & @davidmoreau.bsky.social
psycnet.apa.org/psycarticles...
Draw 2 boxes: exposure & outcome on opposite sides of screen.
Draw 3 causes of exposure w arrows going from cause to exposure.
Draw 3 unique causes of outcome same way.
Draw 3 causes for each of the 6 causes.
Connect any of left side of screen to right side of screen.
Repeat until out of causes
Draw 3 causes of exposure w arrows going from cause to exposure.
Draw 3 unique causes of outcome same way.
Draw 3 causes for each of the 6 causes.
Connect any of left side of screen to right side of screen.
Repeat until out of causes
November 11, 2025 at 1:42 PM
Draw 2 boxes: exposure & outcome on opposite sides of screen.
Draw 3 causes of exposure w arrows going from cause to exposure.
Draw 3 unique causes of outcome same way.
Draw 3 causes for each of the 6 causes.
Connect any of left side of screen to right side of screen.
Repeat until out of causes
Draw 3 causes of exposure w arrows going from cause to exposure.
Draw 3 unique causes of outcome same way.
Draw 3 causes for each of the 6 causes.
Connect any of left side of screen to right side of screen.
Repeat until out of causes
Sending 3yolds (N = 588) to Montessori preschools causes increases in theory of mind, letter-word identification ability.
Lottery study with followup through Kindergarten! Excited to see the longer-term results of this study!
From Angeline Lillard
www.pnas.org/doi/pdf/10.1...
#psych
Lottery study with followup through Kindergarten! Excited to see the longer-term results of this study!
From Angeline Lillard
www.pnas.org/doi/pdf/10.1...
#psych
October 30, 2025 at 2:59 PM
Sending 3yolds (N = 588) to Montessori preschools causes increases in theory of mind, letter-word identification ability.
Lottery study with followup through Kindergarten! Excited to see the longer-term results of this study!
From Angeline Lillard
www.pnas.org/doi/pdf/10.1...
#psych
Lottery study with followup through Kindergarten! Excited to see the longer-term results of this study!
From Angeline Lillard
www.pnas.org/doi/pdf/10.1...
#psych
Kids These days…are LESS shy than they were 20 years ago.
According to parent-reported shyness, it is going down among 4-5 year olds.
From Louis A. Schmidt
link.springer.com/article/10.1...
#psych
According to parent-reported shyness, it is going down among 4-5 year olds.
From Louis A. Schmidt
link.springer.com/article/10.1...
#psych
October 15, 2025 at 1:05 PM
Kids These days…are LESS shy than they were 20 years ago.
According to parent-reported shyness, it is going down among 4-5 year olds.
From Louis A. Schmidt
link.springer.com/article/10.1...
#psych
According to parent-reported shyness, it is going down among 4-5 year olds.
From Louis A. Schmidt
link.springer.com/article/10.1...
#psych
In between-Ss experiments: without a reference point, people say playing a prank is bad but give the same ‘bad’ rating to committing a war crime.
Not in within-Ss experiments or when giving reference points for judgments.
From @vladchituc.bsky.social
www.crockettlab.org/s/1-s20-S001...
#psych
Not in within-Ss experiments or when giving reference points for judgments.
From @vladchituc.bsky.social
www.crockettlab.org/s/1-s20-S001...
#psych
October 14, 2025 at 2:10 PM
In between-Ss experiments: without a reference point, people say playing a prank is bad but give the same ‘bad’ rating to committing a war crime.
Not in within-Ss experiments or when giving reference points for judgments.
From @vladchituc.bsky.social
www.crockettlab.org/s/1-s20-S001...
#psych
Not in within-Ss experiments or when giving reference points for judgments.
From @vladchituc.bsky.social
www.crockettlab.org/s/1-s20-S001...
#psych
Even if you tell people cognitive biases are good & lead to good outcomes, we still think they have the bias less than others.
Only when we see the bias as very desirable we might think we are = to others on it.
From @cruzf.bsky.social & André Mata
link.springer.com/content/pdf/...
#psych #phdsky
Only when we see the bias as very desirable we might think we are = to others on it.
From @cruzf.bsky.social & André Mata
link.springer.com/content/pdf/...
#psych #phdsky
October 1, 2025 at 1:49 PM
Even if you tell people cognitive biases are good & lead to good outcomes, we still think they have the bias less than others.
Only when we see the bias as very desirable we might think we are = to others on it.
From @cruzf.bsky.social & André Mata
link.springer.com/content/pdf/...
#psych #phdsky
Only when we see the bias as very desirable we might think we are = to others on it.
From @cruzf.bsky.social & André Mata
link.springer.com/content/pdf/...
#psych #phdsky
We should fear common method variance; especially in self-report scales.
A nonsense likert scale ‘predicts’ later relationship satisfaction.
We are getting results using methods when we are measuring nothing.
From @datingdecisions.bsky.social
journals.sagepub.com/doi/pdf/10.1...
#psych #stats
A nonsense likert scale ‘predicts’ later relationship satisfaction.
We are getting results using methods when we are measuring nothing.
From @datingdecisions.bsky.social
journals.sagepub.com/doi/pdf/10.1...
#psych #stats
September 29, 2025 at 2:53 PM
We should fear common method variance; especially in self-report scales.
A nonsense likert scale ‘predicts’ later relationship satisfaction.
We are getting results using methods when we are measuring nothing.
From @datingdecisions.bsky.social
journals.sagepub.com/doi/pdf/10.1...
#psych #stats
A nonsense likert scale ‘predicts’ later relationship satisfaction.
We are getting results using methods when we are measuring nothing.
From @datingdecisions.bsky.social
journals.sagepub.com/doi/pdf/10.1...
#psych #stats
Hostile and benevolent sexism are both on the decline in 1,097 studies, N = 339,740 since 1996.
Note that "3" on these scales is already a neutral response. Even bigger progress in countries with more hostile sexism.
From Matthew D. Hammond
psycnet.apa.org/psycarticles...
#phdsky #psych
Note that "3" on these scales is already a neutral response. Even bigger progress in countries with more hostile sexism.
From Matthew D. Hammond
psycnet.apa.org/psycarticles...
#phdsky #psych
September 25, 2025 at 1:37 PM
Hostile and benevolent sexism are both on the decline in 1,097 studies, N = 339,740 since 1996.
Note that "3" on these scales is already a neutral response. Even bigger progress in countries with more hostile sexism.
From Matthew D. Hammond
psycnet.apa.org/psycarticles...
#phdsky #psych
Note that "3" on these scales is already a neutral response. Even bigger progress in countries with more hostile sexism.
From Matthew D. Hammond
psycnet.apa.org/psycarticles...
#phdsky #psych
We have to start finding out how long our interventions last.
So we did.
Manipulating people’s belief in free will works, & lasts for somewhere between less than a day to 1 week at best.
Our new paper, led by @olivergenschow.bsky.social & with @sebraem.bsky.social
osf.io/7ftsk_v1/dow...
#psych
So we did.
Manipulating people’s belief in free will works, & lasts for somewhere between less than a day to 1 week at best.
Our new paper, led by @olivergenschow.bsky.social & with @sebraem.bsky.social
osf.io/7ftsk_v1/dow...
#psych
September 24, 2025 at 1:34 PM
We have to start finding out how long our interventions last.
So we did.
Manipulating people’s belief in free will works, & lasts for somewhere between less than a day to 1 week at best.
Our new paper, led by @olivergenschow.bsky.social & with @sebraem.bsky.social
osf.io/7ftsk_v1/dow...
#psych
So we did.
Manipulating people’s belief in free will works, & lasts for somewhere between less than a day to 1 week at best.
Our new paper, led by @olivergenschow.bsky.social & with @sebraem.bsky.social
osf.io/7ftsk_v1/dow...
#psych
Chimpanzees in the wild of Uganda & Cote d’Ivoire have about 1.4 – 2.6 drinks-worth of alcohol per day in over-ripened fruit.
That’s about the same amount per day as modern humans.
From Aleksey Maro
www.science.org/doi/10.1126/...
#primateresearch #phdsky
That’s about the same amount per day as modern humans.
From Aleksey Maro
www.science.org/doi/10.1126/...
#primateresearch #phdsky
September 23, 2025 at 2:26 PM
Chimpanzees in the wild of Uganda & Cote d’Ivoire have about 1.4 – 2.6 drinks-worth of alcohol per day in over-ripened fruit.
That’s about the same amount per day as modern humans.
From Aleksey Maro
www.science.org/doi/10.1126/...
#primateresearch #phdsky
That’s about the same amount per day as modern humans.
From Aleksey Maro
www.science.org/doi/10.1126/...
#primateresearch #phdsky
Preschool-aged kids whose parents are working class are less likely to be called on when they raise their hand compared to kids with middle/upper class occupations.
Early socialization indeed.
|
From @lewisdoyle.bsky.social & @andreicimpian.bsky.social
www.pnas.org/doi/pdf/10.1...
#psych #phdsky
Early socialization indeed.
|
From @lewisdoyle.bsky.social & @andreicimpian.bsky.social
www.pnas.org/doi/pdf/10.1...
#psych #phdsky
September 17, 2025 at 1:51 PM
Preschool-aged kids whose parents are working class are less likely to be called on when they raise their hand compared to kids with middle/upper class occupations.
Early socialization indeed.
|
From @lewisdoyle.bsky.social & @andreicimpian.bsky.social
www.pnas.org/doi/pdf/10.1...
#psych #phdsky
Early socialization indeed.
|
From @lewisdoyle.bsky.social & @andreicimpian.bsky.social
www.pnas.org/doi/pdf/10.1...
#psych #phdsky
There is a growing inequality in reading for pleasure.
Those who read are reading for longer now than 20 years ago.
BUT when asking people what they did in the past 24 hours fewer said they read (28% to 16%).
From Jessica Bone
www.cell.com/iscience/pdf...
#psych #phdsky
Those who read are reading for longer now than 20 years ago.
BUT when asking people what they did in the past 24 hours fewer said they read (28% to 16%).
From Jessica Bone
www.cell.com/iscience/pdf...
#psych #phdsky
September 11, 2025 at 11:09 AM
There is a growing inequality in reading for pleasure.
Those who read are reading for longer now than 20 years ago.
BUT when asking people what they did in the past 24 hours fewer said they read (28% to 16%).
From Jessica Bone
www.cell.com/iscience/pdf...
#psych #phdsky
Those who read are reading for longer now than 20 years ago.
BUT when asking people what they did in the past 24 hours fewer said they read (28% to 16%).
From Jessica Bone
www.cell.com/iscience/pdf...
#psych #phdsky
Algorithms are not necessary for the creation of echo chambers.
All that is required is people muting or leaving when they have a bad interaction. Then an echo chamber arises with minimal preferences.
From @pettertornberg.com
#psych
arxiv.org/pdf/2508.10466
All that is required is people muting or leaving when they have a bad interaction. Then an echo chamber arises with minimal preferences.
From @pettertornberg.com
#psych
arxiv.org/pdf/2508.10466
September 9, 2025 at 1:32 PM
Algorithms are not necessary for the creation of echo chambers.
All that is required is people muting or leaving when they have a bad interaction. Then an echo chamber arises with minimal preferences.
From @pettertornberg.com
#psych
arxiv.org/pdf/2508.10466
All that is required is people muting or leaving when they have a bad interaction. Then an echo chamber arises with minimal preferences.
From @pettertornberg.com
#psych
arxiv.org/pdf/2508.10466
Biggest concern in mediation is mediator & outcome confounding, even when X is random.
Comparing methods: how the indirect effect acts when errors of M & Y are allowed to correlate is a great option.
From Alvarez-Bartolo & MacKinnon
journals.sagepub.com/doi/pdf/10.1...
#stats #psych #phdsky
Comparing methods: how the indirect effect acts when errors of M & Y are allowed to correlate is a great option.
From Alvarez-Bartolo & MacKinnon
journals.sagepub.com/doi/pdf/10.1...
#stats #psych #phdsky
September 8, 2025 at 12:08 PM
Biggest concern in mediation is mediator & outcome confounding, even when X is random.
Comparing methods: how the indirect effect acts when errors of M & Y are allowed to correlate is a great option.
From Alvarez-Bartolo & MacKinnon
journals.sagepub.com/doi/pdf/10.1...
#stats #psych #phdsky
Comparing methods: how the indirect effect acts when errors of M & Y are allowed to correlate is a great option.
From Alvarez-Bartolo & MacKinnon
journals.sagepub.com/doi/pdf/10.1...
#stats #psych #phdsky
Here is a how-to guide for writing the next social change or technology panic article.
Included is a fillable template, a how to guide to running a correlational study nearly guaranteed to work, suggested language (e.g. save the children), & more.
Free!
#phdsky
osf.io/cvnhu_v1/dow...
Included is a fillable template, a how to guide to running a correlational study nearly guaranteed to work, suggested language (e.g. save the children), & more.
Free!
#phdsky
osf.io/cvnhu_v1/dow...
September 3, 2025 at 12:35 PM
Here is a how-to guide for writing the next social change or technology panic article.
Included is a fillable template, a how to guide to running a correlational study nearly guaranteed to work, suggested language (e.g. save the children), & more.
Free!
#phdsky
osf.io/cvnhu_v1/dow...
Included is a fillable template, a how to guide to running a correlational study nearly guaranteed to work, suggested language (e.g. save the children), & more.
Free!
#phdsky
osf.io/cvnhu_v1/dow...
Generalizability issue: we use participants in our studies & the type of people who are participants are a different type of person than non-participants.
⬆️age, intelligence & education, conventional
⬇️sociable, arousal-seeking, interested in religion, self-disclosing
osf.io/gh2ck_v2/dow...
#phdsky
⬆️age, intelligence & education, conventional
⬇️sociable, arousal-seeking, interested in religion, self-disclosing
osf.io/gh2ck_v2/dow...
#phdsky
August 21, 2025 at 2:02 PM
Generalizability issue: we use participants in our studies & the type of people who are participants are a different type of person than non-participants.
⬆️age, intelligence & education, conventional
⬇️sociable, arousal-seeking, interested in religion, self-disclosing
osf.io/gh2ck_v2/dow...
#phdsky
⬆️age, intelligence & education, conventional
⬇️sociable, arousal-seeking, interested in religion, self-disclosing
osf.io/gh2ck_v2/dow...
#phdsky