Following up on last year’s summary of 2017, I wanted to write up the highlights of my academic 2018: publications, collaborations, achievements and failures, open science, blogging, and some thoughts about the upcoming year.
In summary, due to my new job, and many new obligations that came with it, 2018 was a much busier year, and I managed to get a lot less done in terms of traveling and writing. I gave 13 talks and 7 workshops in 2017, and only 2 talks and 1 workshop in 2018; I had to decline a lot of opportunities to travel, which hurt. I wrote 8 first author papers in 2017, and only 1 in 2018. On the other hand, I had a ton of fun teaching several new courses in Leiden; supervising over 15 Bachelor and Master students. Maybe best of all, I got to travel to London, Cambridge, Glasgow, Lille, Northern Spain, Grand Rapids, and San Francisco. Thanks to all of you who contributed to making it an interesting year full of new experiences.
I decided to leave my personal life out of this blog post for the most part, but briefly wanted to note that it was the first year in my academic life where I struggled quite severely with stress, to the degree that it affected my personal life. This is in part due to crazy things that I had no control over, but in part due to bad decisions I made in terms of workload. I hope things change for the better in 2019, and I’ll put effort into getting a better grasp on the whole work/life balance thing!
1. Assistant Professor Leiden University
After 4 years as a postdoc, I applied for two Assistant Professor positions in both Quantitative Psychology and Clinical Psychology at Leiden University. Clinical Psychology gave me the green light, and I started in April 2018.
It’s a 60% teaching position, meaning I have to teach several 8-week courses per year, plus supervisions (~25 students per year). I expected that to be a lot, but it’s much more than I anticipated, and it’s a remarkable transition from 3 years PhD and 4 years postdoc with zero administration and teaching obligations. Since starting in April, I’ve taught 3 courses, supervised a group of 8 bachelor students, and am supervising 7 master students.
The bachelor group was an interesting challenge: I wanted to finish the complete cycle of research (ideas, planning, data collection, analysis, writing) in 4 months. It was also a challenge because I am not very experienced in gathering data myself. But together with Leonie Cloos who helped me organize the course, we did very well, the students very much enjoyed the course, and we collected Ecological Momentary Assessment data in about 100 international students for 2 weeks, with about 70 measurement points. I hope to put the data online soon, and we’ll likely write up the 8 bachelor theses as a paper.
And while teaching was a burden, it was simultaneously the most rewarding part of working in Leiden — due to the overwhelmingly positive feedback of the students. I taught methods courses for clinical master students that are usually quite poorly evaluated, and an intro course to clinical psychology for bachelor students. I had no prior experience teaching these specific courses, but managed to score 5.0 / 5.0 (n=9) in my first course, 4.9 / 5.0 in my second course (n=11), and 4.9 / 5.0 (n=14) in my third course. My favorite comment was from a student in the clinical psychology course where we read a textbook by Alan Kazdin: “Eiko rocks and Kazdin talks a lot”. There were several other positive comments that students had left in the form, and considering that the evaluation takes place after a gruelling 3-hour exam, it really warmed my heart that students found the time to put in some kind words about their teacher. I also received several longer emails from students such as this one (posting with permission). This sort of positive teaching feedback made me orders of magnitudes happier than any “paper accepted” email from academic journals.
Additionally, I’ve been very lucky with my new colleagues in Leiden, who have all been extremely helpful and forthcoming, and made me feel welcome. Without them, I would have never managed to teach that well either. The language in some of the Department meetings is Dutch, unfortunately, which I find disappointing for a University which prides itself for its international character.
Finally, my 18-minute train commute from Haarlem to work is so amazing, I cannot stop talking about it …
Another photograph of the "I hate my commute to work" series ;) pic.twitter.com/EATKHEXsZR— Eiko Fried (@EikoFried) May 6, 2018
Commute home from work: Traffic, grey ugly buildings, pollution, rain.— Eiko Fried (@EikoFried) November 18, 2018
Moving to the Netherlands was the worst mistake of my life. pic.twitter.com/YiDVjCD4fl
Unsurprisingly, I got less work done this year. But I collaborated a lot more, and was PI on 8 papers (numbers 2 through 9 below), which was a fun experience and taught me skills I had not picked up before. For what it’s worth, I had about 1000 citations a year ago, and am closing in on 2000. A topic I’ve been thinking a lot about this year is authorship, and I declined to be on a few papers that will likely be much more cited than my own work, because I feel I didn’t contribute enough (I was asked to ‘sign on’ to a few papers). I still struggle with my decision, to be honest, and Jennifer Tackett and I hope to find the time 2019 to write a paper on the topic more generally.
- Fried et al. (2018). Replicability and generalizability of PTSD networks: A cross-cultural multisite study of PTSD symptoms in four trauma patient samples. Clinical Psychological Science. PDF.
- Epskamp & Fried (2018). A Tutorial on Regularized Partial Correlation Networks. Psychological Methods. PDF.
- Aalbers et al. (2018). Social Media and Depression Symptoms: A Network Perspective. Journal of Experimental Psychology: General. PDF.
- Santos et al. (2018). Network Structure of Perinatal Depressive Symptoms in Latinas: Relationship to Stress-Related and Reproductive Biomarkers. Research in Nursing & Health. PDF.
- Greene et al. (2018). Dynamic networks of PTSD symptoms during conflict. Psychological Medicine. PDF.
- Fonseca-Pedrero et al. (2018). The network structure of schizotypal personality traits. Schizophrenia Bulletin. PDF.
- Kendler et al. (2018). The Centrality of DSM and non-DSM Depressive Symptoms in Han Chinese Women with Major Depression. Journal of Affective Disorders. PDF.
- Epskamp, Borsboom & Fried (2018). Estimating psychological networks and their accuracy: a tutorial paper. Behavioral Research Methods. PDF.
- Santos et al. (2018). Longitudinal network structure of depression symptoms and self-efficacy in low- income mothers. PLoS ONE. PDF.
- Molendijk, Fried & van der Does (2018). The SMILES trial: do undisclosed recruitment practices explain the remarkably large effect? BMC Medicine. PDF.
- Briganti, Fried & Linkowski (2018). Network analysis of Contingencies of Self-Worth Scale in 680 university students. Psychiatry Research. PDF.
- Fritz et al. (2018). A Network Model of Resilience Factors for Adolescents with and without Exposure to Childhood Adversity. Scientific Reports. PDF.
- Rodebaugh et al. (2018). Does centrality in a cross-sectional network suggest intervention targets for social anxiety disorder? Journal of Consulting and Clinical Psychology. PDF.
- Rouquette et al. (2018). Emotional and Behavioral Symptom Network Structure in Elementary School Girls and Association With Anxiety Disorders and Depression in Adolescence and Early Adulthood — A Network Analysis. JAMA Psychiatry. PDF.
- Borsboom et al. (2018). Robustness and replicability of psychopathology networks. World Psychiatry. PDF.
- Briganti et al. (2018). Network analysis of empathy items from the Interpersonal Reactivity Index in 1973 young adults. Psychiatry Research. PDF.
- Bos et al. (2018). Cross-sectional networks of depressive symptoms before and after antidepressant medication treatment. Social Psychiatry and Psychiatric Epidemiology. PDF.
- von Stockert et al. (2018). Evaluating the stability of DSM-5 PTSD symptom network structure in a national sample of U.S. military veterans. Journal of Affective Disorders. PDF.
- van Loo et al. (2018). Robust symptom networks in recurrent major depression across different levels of genetic and environmental risk. Journal of Affective Disorders. PDF.
I learned a lot from working with others, but I also found out that being the methods expert on a paper is less rewarding for me than it used to be. I ended up having to rewrite more than I had anticipated, because I found papers to be somewhat overstated, especially in terms of drawing causal inferences in the absence of causal evidence. I’m a fairly strict reviewer when it comes to this, but am at least equally strict with papers I’m involved in, of course. This is by no means criticism of my co-authors, who have a ton of substantive knowledge in other areas such as clinical theory and treatment that I do not have. It also only happened rarely. But I found it to be a time-consuming and repetitive enterprise, and one of my main goals for 2019 is to say ‘no’ more to helping others fit models, focus more on my own work, and only participate in projects where I get to learn something new.
That being said, there were many projects in which I learned a lot, many more than I could discuss here. Let me highlight 3 projects specifically. First, I got to work with Donald Robinaugh and colleagues on Don’s formalized theory of panic disorder; the preprint will go online in a few months at the latest. Don and I submitted a workshop to SIPS 2019 on the topic of Don’s paper yesterday, and I am enthusiastic about computational models that explicitly describe psychological theories. Here’s the abstract:
Most psychological theories are “verbal models”: narrative descriptions of hypotheses aimed at explaining psychological phenomena. Because of the vagaries of language, such models can be imprecise and, thus, vulnerable to hidden assumptions and other unknowns. Moreover, reliance on these models has likely contributed to the replication crisis in psychological science, as their ambiguity creates fertile ground for questionable research practices (e.g., hypothesizing after the results are known). In this workshop, we provide an introduction to theory development through computational modeling in three steps. First, we discuss the advantages of computational modeling and show how it fosters more transparent, replicable, and cumulative science. Second, we present a step-by-step tutorial on how to generate, develop, and evaluate formalized psychological theories. Third, we use examples to illustrate the process and benefits of computational modeling, including agent-based modeling in social psychology and computational models of learning in cognitive psychology. We focus especially on a cognitive behavioral theory of Panic Disorder and show how the theory can be implemented as a computational model using a set of difference equations in R. We conclude by illustrating how that model facilitates theory development and evaluation. Our hope is that participants leave this workshop with (a) an appreciation of the importance of formalized psychological theories, (b) a basic understanding of what difference equations are and how they can be used to implement psychological theories as computational models, and (c) ideas for how to develop and evaluate formalized theories in their own work.
Second, Jessica Flake and I wrote up our Measurement Schmeasurement paper, during a writing retreat in Glasgow, hosted by Lisa DeBruine & Ben Jones. The paper is based on a symposium we had organized for APS 2018, our APS Observer piece Measurement Matters, and the SIPS 2018 worskhop on questionable measurement practices. We will have a preprint online within the next few weeks. Oh, and Jessica and I also put up a measurement resource list, featuring papers & books on different measurement topics we think are recommended reading material.
As expected, Glasgow was incredibly boring, and all we did was working. I swore an oath to never return.
Given all the work I did with Jessica 2018, it’s hard to pick a Flake-highlight. One of my favourites was definitely Jessica trying to tempt me into questionable research practices at our SIPS workshop though:
At the highly entertaining measurement workshop this morning, @JkayFlake plays devil’s advocate to @EikoFried : “why not gather all the measures and see what’s significant?” 😈 #SIPS2018 pic.twitter.com/ugVwL4xyCo— Jackie Thompson (@NeuroJackie) June 25, 2018
Third, Linda Nab & I set up an interdisciplinary measurement colloquium, with participants from Clinical Epidemiology, Clinical Psychology, and Quantitative Psychology. Both meetings so far were incredibly rewarding, and I hope we will find the time to continue this in 2019.
1/2 Just finished our first interdisciplinary meeting on measurement issues in social and life sciences 📏 at @UniLeiden that I co-organized with @lindanab1, including psychologists, statisticians & epidemiologists. Linda gave a presentation on 📏 error.— Eiko Fried (@EikoFried) November 2, 2018
Want to join next time? pic.twitter.com/227cbHjiiz
4. Lectures, workshops, conferences
This was my first year since 2014 where I only went to 2 conferences, due to both funding and time constraints, and I had to select them carefully. I went to APS, a conference that has made a lot of progress in terms of methodology and open science since I went there the first time 2014, and to SIPS, which was the most amazing conference I have ever been to. You can find some reflections on both conferences here (APS) and here (SIPS). I had to say no to a few invitations to talk at conferences (e.g. ABCT, ADAA) and Universities, due to teaching. I also didn’t get to teach any network analysis workshops, which I had done a lot in 2016 and 2017 and were always a lot of fun. I hope I can make more time for lectures and workshops in 2019.
5. Open science
Apart from learning a ton about open science at SIPS, I joined the BMC Medicine Editorial Board (an open access open review journal), continued my work as associate editor at Collabra Psychology (a new open access journal) and on the advisory board of the new journal Meta Psychology (free of open access publishing), and agreed to join an open science task force for a journal specialized on measurement after the incoming editor in chief takes over. Additionally, we published 4 papers 2018 for which we were able to share data. Lastly, I got to meet Anna van T’Veer at SIPS who works in the Quantitative Psychology Group in Leiden. I assisted Anna in founding the Open Science Community Leiden (Twitter, Website with signup), and we had our inaugural meeting in December.
6. Blogging on Eiko-Fried.com
The biggest new development was starting the “Brief Psychology News” rubric, which started as an email I sent out to my new colleagues in the Clinical Psychology Department at Leiden University. I figured I could also put the news online, so it became a blog series. I’m not fully committed to continue the series yet, and feedback is appreciated whether you think that’s a resource that is useful to you.
In total, I wrote 14 blogs for my personal website in 2018, which reflect my broad interests somewhere between clinical psychology, methodology, philosophy of science, and rigorous/open science:
- Brief psychology news 11/2018
- James Coyne sued me for cyberbullying
- Brief psychology news 10/2018
- Failure of the week
- Brief psychology news 08/2018
- 10,377 ways for major depression, but 341,737 ways for melancholia
- SIPS18 collected resources, and reflections of a SIPS virgin
- Brief psychology news 06/2018
- Measurement Schmeasurement
- Reflections on APS 18: open science, transparency, and inclusion
- 7 Sternberg papers: 351 references, 161 self-citations
- Psychology: in the upside down of academic publishing
- Assistant professor, APS rising star, and some papers
- All mental disorders are brain disorders … not.
7. Blogging on Psych-Networks.com
I found a lot less time to blog on psych-networks this year, and want to transform it more into a community platform. I did find the time to keep the tutorial site updated, and the R-packages site, and did write a few additional tutorials.
In total, there were 9 blogs in 2018:
- Fixed-margin sampling & networks: New commentary on network replicability (Sacha Epskamp)
- (Mis)interpreting Networks: An Abbreviated Tutorial on Visualizations (Payton Jones)
- How to interpret centrality values in network structures (not) (Eiko)
- Bootstrapping edges after regularization: clarifications & tutorial (Eiko)
- APS 2018: Collection of all network presentations (Eiko)
- New paper on the role of stabilizing and communicating symptoms (Tessa Blanken & Marie Deserno)
- Estimating psychological networks via Information Filtering Networks (Eiko)
- FAQ on network stability, part II: Why is my network unstable? (Eiko)
- Collection of PTSD network papers & recent conference talks (Tobias Spiller)
In the future, I would like to publish a lot more guest blogs, and want to especially encourage people who do not identify as male to submit guest blogs (to clarify: don’t want men to submit less, but others to submit more). Last year, in my 2017 summary, I wrote this: “Seeing this list [of guest bloggers], however, made me realize than nearly all guest bloggers were guys, so when I post my 2018 summary, my goal is to have 50% female contributors to psych-networks (and do call me out if I don’t reach that goal).” I count 3 men and 2 women, so I failed my 2018 goal. We currently have 5 guest blogs planned, from 4 female researchers, so I hope things will turn around in 2019. Again, please reach out to me: anything broadly related to dynamical systems (philosophy, psychology, methodology) will work, and there were several guest blogs in the past I do not (entirely) agree with, so also reach out to me in these cases. The guest blogs are not supposed to reflect my opinions, but to be contributions that lead to insightful discussions.
I had planned to apply for 2 large grants 2018, but failed to the degree that I didn’t even start naming a folder on my computer “grants 2018”. Given my new job, I guess I shouldn’t be too hard on myself, but I really, really need to apply for these 2 grants in 2019. Further, I had to submit a correction for one of my 2017 papers for which I made a pretty stupid error, which I wrote up in more detail here. This also made me realize that we should talk more about failures in academia, which I’ll try to do more in the future.
I was more active on Twitter this year. Inspired by Veronika Cheplygin, here are some of my 2018 tweets that led to interesting discussions that I enjoyed:
1) My MBTI rant:
1/6 When you see someone use the Myer-Briggs Type Indicator in a professional context, call them out. It does not meet basic criteria of psychometric scales, was "developed" in the 40s by people without any training, & is based on outdated psychoanalytic ideas.— Eiko Fried (@EikoFried) April 1, 2018
Long version 👇
2) A short video tutorial on sci-hub:
I put together a 1 minute tutorial video on how to set up sci-hub efficiently to gain access within a few seconds to PDFs of scientific papers your University library has access to #scihub. For #chrome, but also works for other browsers.— Eiko Fried (@EikoFried) January 8, 2018
(Sorry for the typos 🤭 …) pic.twitter.com/Z5wZBoLKk9
3) A comment on a rejection letter by an editor:
Paper of colleague was rejected recently.— Eiko Fried (@EikoFried) June 8, 2018
Editor: "The study reports a null finding, which, as you know, is uninterpretable. My decision is based on whether the work provides truly new and novel findings."
Let me repeat what I wrote on Facebook a few days back: pic.twitter.com/7kHMXAPd2q
4) A gif on model equivalence:
When different models have the same fit in the same data, we talk about 'model equivalence'. The gif below (via Denny Borsboom) highlights this phenomenon very well.— Eiko Fried (@EikoFried) March 29, 2018
For equivalence between network and latent variable models, see Denny's guest blog here: https://t.co/FWbiE1m0IZ pic.twitter.com/5qWzhDcHez
5) My blog post on SIPS 2018:
For those who couldn't join #SIPS2018 and either have the #fomo or want to know what was going on, I put together a few resources, summaries of projects people were working on, and praise/challenges here. https://t.co/n1bQ6PCn2o— Eiko Fried (@EikoFried) June 28, 2018
6) A tweet on reviewer comments that had recommended to reject the paper on the large replicability project:
The big psychology replicability study published a few days ago was rejected in both Nature & Science, with reviewer comments like the one below.— Eiko Fried (@EikoFried) August 29, 2018
Why favor these results over the older ones? I'll tell you:
1) An average of 5 times the power
2) They were pre-registered https://t.co/yw3jd5J28E
7) A short video tutorial on a useful R-studio hack:
8) Finally, I wanted to thank all of you for your amazing support (on Facebook, privately, and in response to the tweet below) regarding my experiences with Coyne. I’m not making this a bigger topic in this 2018 review because I’ve wasted too much time on this guy. I’m not going to lie though, it makes me both angry and sad that he is still invited by Universities to give talks and workshops, after all that happened, and did have a fairly profound impact on my mental health in 2018.
Thanks again to everybody who supported be over the last year. There are many more names that I could tag in one tweet: you know who you are. https://t.co/1hnySDKizW— Eiko Fried (@EikoFried) October 23, 2018
10. The coming year 2019
There are a few things I am most excited about for 2019. First, I’ll be spending June through August in Richmond (VA), working with Ken Kendler. Many of you will know that he’s my science superhero, has inspired a lot of my own work, and I’m incredibly excited to get the opportunity to pick Ken’s brain for 3 months. If you live closeby (say within 350 miles), please reach out, I’d be happy to meet up!
Second, I’m very much looking forward to APS 2019, not only because I get to see many of my science friends there, but also because I put together a symposium entitled “Recent advances in the use of modeling to explain and predict psychological phenomena: Examples from Social, Health, and Clinical Psychology”:
For APS 2019 @PsychScience, I put together a symposium on a topic I'd like to learn more about. The symposium was accepted, and I'm excited to hear @DepressionLab @talyarkoni @aaronjfisher Donald Robinaugh & Jonas Dalege discuss models to explain/predict psychological constructs. pic.twitter.com/wd5AWVTzNv— Eiko Fried (@EikoFried) December 23, 2018
Third, I’m looking forward to my second SIPS conference, SIPS 2019 in Rotterdam. It was an incredibly rewarding experience last time around, and I’m thrilled to keep learning from and contributing to the growing open science community in psychology. If our workshop gets accepted, that’s even better, but I’ll be there anyway, even if I have to leave beloved Richmond for a few days.
Finally, I get to participate in a 2-day workshop on Open Science & Reproducibility, and am looking forward to meet many tweeps there.
Pleased to announce our 2-Day IMC Workshop on Open Science and Reproducibility in March 12-13 - #markyourcalendars! Check out the confirmed speakers here: https://t.co/yIFNxbfrMQ org by @ce_parsons @JCSkewesDK @aroepstorff pic.twitter.com/VjlwjHetp5— Interacting Minds (@interact_minds) December 14, 2018
And now I have to be on my way, doing secret vigilante things …
Ok, who put googly eyes on my coat hook? pic.twitter.com/mYSnljY6IY— 🐘 @firstname.lastname@example.org (@LisaDeBruine) November 6, 2018