Everything Hertz

By Dan QuintanaScience Social Sciences Medicine Health & Fitness

Everything Hertz episodes

August 27, 201625: Misunderstanding p-values
P-values are universal, but do we really know what they mean? In this episode, Dan and James discuss a recent paper describing the failure to correctly interpret p-values in a sample of academic psychologists.
Some of the topics discussed:
Common p-value misconceptions
James tests Dan on his p-value knowledge
p-values vs. effect size
The problem of sample size with p-value interpretation
The Facebook mood manipulation study
Data peeking
Equivalent p-values do not represent equivalent results
Meta-analytical thinking
Using significance as a categorical factor
Statistical vs. clinical significance
Clinical trial registration and 'secondary outcome creep'
Dan and James answer listener questions
Science communicator vs. scientist
Grant titles and the 'Pub test'
NASA and social media
Links
The article
http://journal.frontiersin.org/article/10.3389/fpsyg.2016.01247/full
Geoff Cumming's book (we got the name completely wrong - sorry Geoff!)
http://www.amazon.com/Understanding-The-New-Statistics-Meta-Analysis-ebook/dp/B007M9D76G/ref=pd_sim_kstore_1?ie=UTF8&refRID=1QWKES82EP85DBAEKNT1
The story on research passing the 'pub' test
https://theconversation.com/if-youre-going-to-ridicule-research-do-your-homework-64238
Real scientists
http://realscientists.org
Facebook page
https://www.facebook.com/everythinghertzpodcast/
Twitter account
https://www.twitter.com
Support Everything Hertz
56 min
August 17, 201624: Incentive structures in science
Science funding has a series of built in incentive structures, but what sort of science does this produce?
Some of the topics discussed:
Feedback from our 'Public health and Pokemon' episode (#22)
Incentive structures in science
What we should be doing in science compared to what we are doing
Quantity vs. Quality
The analysis of Trump's tweets for negativity vs. positivity
Pre-registration
How much detail do you need to go into when it comes to pre-registering an analysis
APS journal badges - they're working!
Data sharing makes you more careful with your data
Solutions to the incentive problem have to come from the policy level
The grant funding lottery system proposal
The PhD oversupply
Gaming the system
James wants to mandate science communication
Dan wants to include replication studies in PhD programs
Scientist names that suit their research area
Links
The article on incentive structures
https://medium.com/the-spike/how-a-happy-moment-for-neuroscience-is-a-sad-moment-for-science-c4ba00336e9c#.x3sea13i1
The Allen brain atlas
http://observatory.brain-map.org
Analysis of Trump's tweets
http://varianceexplained.org/r/trump-tweets/
Facebook page
https://www.facebook.com/everythinghertzpodcast/
Twitter account
https://www.twitter.com/hertzpodcast
Support Everything Hertz
1 hr 1 min
August 11, 201623: Serious academics
Can you be a "serious academic" while still posting photos on Instagram? In this episode, James and Dan discuss a recent article bemoaning the infiltration of the "selfie epidemic" into academia.
Some of the topics discussed:
James and Dan share their thoughts on the article
The REAL 'c' word....
Social media and conferences
Snapchat + academics = snapademics
Using Instagram stories to share you research
Why "PHD comics" is so successful
Criticism in academia
Listener question 1: What's your favourite part of research?
Listener question 2: What's your favourite technique or experiment to perform?
Listener question 3: What's a funny story from being an academic?
Links
The article
https://www.theguardian.com/higher-education-network/2016/aug/05/im-a-serious-academic-not-a-professional-instagrammer
A response to the article
https://www.theguardian.com/science/brain-flapping/2016/aug/05/im-a-non-serious-academic-i-make-no-apologies-for-this
Snapademics
https://twitter.com/snapademia
Facebook page
https://www.facebook.com/everythinghertzpodcast/
Twitter account
https://www.twitter.com/hertzpodcast
Support Everything Hertz
53 min
August 03, 201622: Pokemon and public health
Pokemon Go is sweeping the world and getting people walking again! But is the Pokemon Go 'model' a golden opportunity to tackle obesity or just another fad?
Some of the topics discussed:
James plays "Pokemon or Cholesterol medication?"
Dan tries to explain Pokemon Go to James
James' first contact with Pokemon Go "trainers"
Should health interventions be modeled on Pokemon Go?
Other exercise augmented reality health apps
What's the app's endgame?
Can health authorities copy this model?
We make a correction from episode 17: PLoS is in fact a non-profit journal, not a for-profit journal
Dan and James answer two listener questions: i) The dumbest things they've ever done in the lab (both related to email faux pas) ii) How often should lab meetings be run
The importance of PROPERLY piloting your experiment
If you don't know the person in the meeting that takes up too much time, it's probably you.
Links
The quiz
http://www.slate.com/articles/technology/gaming/2016/07/pokemon_or_cholesterol_medication_a_quiz.html
Facebook page
https://www.facebook.com/everythinghertzpodcast/
Twitter account
https://www.twitter.com/hertzpodcast
Support Everything Hertz
1 hr
July 22, 201621: This is your brain on steroids
It's well established that steroid use is associated with many adverse healthy outcomes, but what does it actually do to your brain?
Dan and James discuss an interesting new paper that compared brain structure in long-term steroid users and non-using weightlifters.
Some of the topics discussed:
A summary of the study
How are steroids typically used?
What are the differences in use between sports?
The recruitment of 'real' users
James gives Dan a surprise Norwegian test (he doesn't do too well)
The things Dan and James liked about the study (hint: many things)
Steroid use in women
Dose-dependent effects of steroids
Folk beliefs surrounding steroid use
James' goal of making his cat as jacked as possible
If you have a great study, there's no need to oversell
James' experience of participating in a growth hormone trial
Links
The paper
http://www.sciencedirect.com/science/article/pii/S000632231632529X
Facebook page
https://www.facebook.com/everythinghertzpodcast/
Twitter account
https://www.twitter.com/hertzpodcast
Support Everything Hertz
59 min
July 13, 201620: Sample sizes in psychology studies
Can psychologists learn more by studying fewer people?
Some of the topics discussed:
Brexit and science
Can the UK take the 'Norway' option?
Horizon 2020
The impact on personnel and research training
Italian coffee
Listener feedback
We're sorry for the chewing sound from episode 17!
Intraindividual replication vs. larger sample sizes
What sort of experiments are better suited to detailed within-subject studies?
Is 'quantified self' data more valid than experimental data?
What if you happen to recruit a 'weird' person?
Links
The paper
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4911349/
Facebook page
https://www.facebook.com/everythinghertzpodcast/
Twitter account
https://www.twitter.com/hertzpodcast
Support Everything Hertz
1 hr 2 min
July 06, 201619: Let us spray: oxytocin and spirituality
Dan and James discuss a recent paper on intranasal oxytocin and spirituality
Some of the topics discussed:
A summary of a recent paper on oxytocin and spirituality
Why within-subject designs are a better choice for oxytocin research
The physiology of nasal administration
How do you control for differences in nasal environment
Hypothesis-driven vs. exploratory research
Oxytocin pathway gene
ANCOVA and Lord's paradox (yep, it's called that)
Dan applauding the authors for posting ALL their data online
James disagrees with Dan on approaches to pre-registering studies
James promises never to chew during a recording (sorry!)
Producing research vs. actually reading other people's research
Our first citation (sort of)
How the replication crisis is the fact that we're not doing any
Heart rate variability in oysters (and other animals)
Links
The paper
http://scan.oxfordjournals.org/content/early/2016/06/22/scan.nsw078
Dan's blog post on intranasal oxytocin administration considerations
https://medium.com/@dsquintana/intranasal-drug-administration-in-psychiatry-80d076f1abdd#.dsdrrohdu
Dan's paper on the promise and pitfalls of intranasal administration
http://www.ncbi.nlm.nih.gov/pubmed/26552590
The data from the paper
https://osf.io/mjhzw/
Our first (blog post) citation
https://rhythmicpsychology.wordpress.com/2016/06/22/shock-shock-horror-horror/
Can we replicate the replication crisis?
https://theconversation.com/is-psychology-really-in-crisis-60869
Facebook page
https://www.facebook.com/everythinghertzpodcast/
Twitter account
https://www.twitter.com/hertzpodcast
Support Everything Hertz
48 min
June 29, 201618: Data sharing
Withholding data: bad science or scientific misconduct?
Some of the topics discussed:
Dan raises privacy issues surrounding sharing data
What are the limits of AI to identify people from 'un-identifiable' data?
The new age of sharing data
What grinds Dan's gears
Requesting data from people who said that they'd actually share their data
James' experiences with requesting data
Dan offers a solution for accessing data via an independent 3rd party
When is a data request 'vexatious'?
Hiding data vs. just being lazy with data management
The importance of sharing your analysis code along with the data
Version controlling manuscripts
How much will open data actually improve science?
'Fake' journals masquerading as real ones
Australia's windfarm commissioner (a real thing, unfortunalty)
Australian coffee vs. the rest of the world
Our Twitter followers numbers have surpassed our Facebook fans!
Links
The paper on requesting data
http://www.ncbi.nlm.nih.gov/pubmed/22686633
The blog post on whether withholding your data is scientific misconduct
http://blogs.lse.ac.uk/impactofsocialsciences/2015/07/03/data-secrecy-bad-science-or-scientific-misconduct/
Facebook page
https://www.facebook.com/everythinghertzpodcast/
Twitter account
https://www.twitter.com/hertzpodcast
Support Everything Hertz
52 min
June 22, 201617: Journals: Do we need them?
Do we really need scientific journals?
Some of the topics discussed:
James trolling predatory journals with jibberish papers on the 'DONG' effect
How do these spammy journal invitation emails actually work?
Formal journals vs. preprint servers
The shift to preprints in psychology
Why do some journals forbid preprints?
An article defending the big publishers
How much does it really cost to have an online journal?
What if the public funders were to set up thier own journals?
What's the role of post-pubication review?
Do journals provide added value?
James' brush with the 'established' media
Links
Publisher policies on preprints
https://en.wikipedia.org/wiki/List_of_academic_journals_by_preprint_policy
How much do universities pay for journal subscriptions?
https://gowers.wordpress.com/2014/04/24/elsevier-journals-some-facts/
Facebook page
https://www.facebook.com/everythinghertzpodcast/
Twitter account
https://www.twitter.com/hertzpodcast
Support Everything Hertz
52 min
June 15, 201616: What makes a good psych study?
What are the defining characteristics of a good psychology study? We received this excellent question from a listener and decided to do a whole episode on this idea.
Some of the topics discussed:
When’s the last time you saw a psych study that only reported a t-test?
Dan and James’ new paper on worry and heart rate variability
Skepticism towards studies with many variables and ‘novel’ statistical approaches (that tend to always provide marvellous results)
Repeated measures ANOVAs vs. linear mixed models
Publishing convenient ideas, even if they’re wrong
Fishing expeditions
The ‘nocebo’ effect in action
What are markers of study quality?
Pre-registering studies - can it be gamed?
The gradual improvement of psychology studies
Links
Dan and James' new paper on worry and heart rate variability
http://www.ncbi.nlm.nih.gov/pubmed/27255891
The paper that said "Encouraging experimental psychologists to use LMMs was like giving shotguns to toddlers.”
http://www.ncbi.nlm.nih.gov/pubmed/25657634
Facebook page
https://www.facebook.com/everythinghertzpodcast/
Twitter account
https://www.twitter.com/hertzpodcast
Support Everything Hertz
50 min