Lightnews — Scholar-powered news

Dylan Burton

@dylanburton.bsky.social

500 followers 160 following 20 posts

Assistant Professor at Georgia State University. MSU and Lancaster alum. My research focus is on language testing and assessment Learn more about me here: https://dylanburton.hcommons.org

dylanburton.hcommons.org

Posts Media Videos Starter Packs

Pinned

Dylan Burton @dylanburton.bsky.social · Feb 17

Excited that my article with Paula Winke has been published today in Studies in Second Language Acquisition! In this study we examined the relationship between non-language specialists' estimates of L2 language ability and perceived affect. www.cambridge.org/core/journal...

Affect as a component of second language speech perception | Studies in Second Language Acquisition | Cambridge Core

Affect as a component of second language speech perception

www.cambridge.org

1 4 12

Reposted by Dylan Burton

Pauliina Peltonen @pauliinapeltonen.com · 20h

Happy to announce that our article (w/ Sandra Götz & @pekkalintunen.bsky.social) exploring L1 & L2 speech #fluency in monologue + dialogue is now out in SSLA (OA)! Part of the recently completed #FDF2 project (funded by the Research Council of Finland). #FlowLangGroup @utu.fi doi.org/10.1017/S027...

The Impact of L1 Speaking Style, Task Mode, and L2 Proficiency on L2 Fluency: A Within-subject Study of Monologic and Dialogic Speech | Studies in Second Language Acquisition | Cambridge Core

The Impact of L1 Speaking Style, Task Mode, and L2 Proficiency on L2 Fluency: A Within-subject Study of Monologic and Dialogic Speech

doi.org

1 2 7

Reposted by Dylan Burton

Garrick Aden-Buie @grrrck.xyz · 27d

ggplot2 4.0.0 is out and the new `paper`, `ink`, `accent` theme variables look super cool! Just pick 2-3 colors 🎨 to make your plots look great! I'm excited to hook this up to brand.yml 😉

A ggplot2 plot showing a classic `mpg` scatter plot. The code uses `theme_gray()` with `paper = "cornstalk"` for a light yellow background, `ink = "navy"` for navy blue text and points, and `accent = "tomato"` which is applied to the smoothing line created with `geom_smooth()`

1 19 73

Dylan Burton @dylanburton.bsky.social · Aug 27

Have One On Me compite con Vulnicura por disco de desamor ❤️

1 1

Reposted by Dylan Burton

Julia M. Rohrer @dingdingpeng.the100.ci · Aug 25

Ever stared at a table of regression coefficients & wondered what you're doing with your life?

Very excited to share this gentle introduction to another way of making sense of statistical models (w @vincentab.bsky.social)
Preprint: doi.org/10.31234/osf...
Website: j-rohrer.github.io/marginal-psy...

Models as Prediction Machines: How to Convert Confusing Coefficients into Clear Quantities

Abstract
Psychological researchers usually make sense of regression models by interpreting coefficient estimates directly. This works well enough for simple linear models, but is more challenging for more complex models with, for example, categorical variables, interactions, non-linearities, and hierarchical structures. Here, we introduce an alternative approach to making sense of statistical models. The central idea is to abstract away from the mechanics of estimation, and to treat models as “counterfactual prediction machines,” which are subsequently queried to estimate quantities and conduct tests that matter substantively. This workflow is model-agnostic; it can be applied in a consistent fashion to draw causal or descriptive inference from a wide range of models. We illustrate how to implement this workflow with the marginaleffects package, which supports over 100 different classes of models in R and Python, and present two worked examples. These examples show how the workflow can be applied across designs (e.g., observational study, randomized experiment) to answer different research questions (e.g., associations, causal effects, effect heterogeneity) while facing various challenges (e.g., controlling for confounders in a flexible manner, modelling ordinal outcomes, and interpreting non-linear models).

Figure illustrating model predictions. On the X-axis the predictor, annual gross income in Euro. On the Y-axis the outcome, predicted life satisfaction. A solid line marks the curve of predictions on which individual data points are marked as model-implied outcomes at incomes of interest. Comparing two such predictions gives us a comparison. We can also fit a tangent to the line of predictions, which illustrates the slope at any given point of the curve.

A figure illustrating various ways to include age as a predictor in a model. On the x-axis age (predictor), on the y-axis the outcome (model-implied importance of friends, including confidence intervals).

Illustrated are
1. age as a categorical predictor, resultings in the predictions bouncing around a lot with wide confidence intervals
2. age as a linear predictor, which forces a straight line through the data points that has a very tight confidence band and
3. age splines, which lies somewhere in between as it smoothly follows the data but has more uncertainty than the straight line.

49 280 940

Reposted by Dylan Burton

Cambridge University Press - Linguistics @cambup-linguistics.cambridge.org · Aug 22

New issue of Studies in Second Language Acquisition now available

📚 cup.org/47eF6ql

#SSLA #appliedlinguistics #LangSky

cover of the journal Studies in Second Language Acquisition

2 3

Reposted by Dylan Burton

Jessi Grieser @jessgrieser.com · Aug 21

While I'm dispensing computer advice, to any new graduate students:

Save all your article PDFs using the same template. I like LASTNAME YEAR SHORT TITLE/DESCRIPTION, like

Labov 1963 Marthas Vineyard

Save them all in just one giant folder, not piecemeal by class. Thank me when you're tenured.

1 1 21

Reposted by Dylan Burton

Ryan Bowles @ryanpbowles.bsky.social · Aug 21

The new Cracker Barrel logo is the visual version of a 50 year old saying skibidi.

1 1

Reposted by Dylan Burton

Vincent Arel-Bundock @vincentab.bsky.social · Aug 19

{tinytable} 0.13.0 for #RStats can now draw interactive tables with sorting, filtering & pages!

And the 📦 is still an ultra-simple, powerful, and 0-dep way to create beautiful tables in basically any format: tex, typ, docx, pdf, png, md, etc

Gallery & Docs: vincentarelbundock.github.io/tinytable/

7 40 150

Reposted by Dylan Burton

Darren Dahly @statsepi.bsky.social · Aug 11

Papers often conclude "more research is needed" without explanation. This is a missed opportunity. You are the expert. This is your time to shine. Explain what the remaining uncertainties are, and give justified recommendations on what the research needed to resolve them should look like.

13 30 160

Reposted by Dylan Burton

Julia M. Rohrer @dingdingpeng.the100.ci · Aug 7

I've said it before, and I will say it again:

It's astonishing how emotionally aversive revisions are, even in the best possible case (reviewers are positive and have very actionable remarks).

Or at least that's how I work; if anybody has some good recommendations for coping, do let me know 😂

14 10 110

Reposted by Dylan Burton

Nicole Holliday @mixedlinguist.bsky.social · Aug 7

I shouldn't wade into this discourse, but LANGUAGE CHANGES and old people always hate what kids are doing. This is fine, especially if the author intends it to be euphemistic. Language policing always ends in systematic discrimination against marginalized people. We don't want to go down this road!

Indie Book Spotlight @bookspotlight.bsky.social · Aug 4

If you write ‘unalive’ in prose or character dialogue, in your NOVEL - which I have seen TWICE NOW - I will immediately wince and full body cringe. The TikTokification of words has to stop. Your character almost DIED or was KILLED in your novel. They were not almost ‘unalived’.

7 18 110

Reposted by Dylan Burton

Language Testing @langtestjournal.bsky.social · Aug 5

Now available in OnlineFirst, Shungo Suzuki and colleagues propose a new approach to weakness identification for L2 speaking skills in diagnostic language assessment (DLA), with an AI-driven DLA program that uses contextualized diagnostic feedback to improve lexical use. doi.org/10.1177/0265...

1 1

Reposted by Dylan Burton

Julia M. Rohrer @dingdingpeng.the100.ci · Jul 16

At this point, I might as well --
Here's an infographic showing different ways to include age as a predictor. The top shows two extremes, just as a plain old numerical predictor (imposes linear trajectory) vs. categorical predictor (imposes nothing whatsoever). And then three solutions in between!

Infographic illustrating different ways to model age.
First panel shows two "extreme" cases; including age as a linear numerical predictor (df = 1) or including age as a categorical predictor (df = number of years of age minus 1).
Second panel shows an intermediate solution in which age is categorized into broader bins (df = number of categories minus 1, here 5 - 1 = 4).
Third panel shows an intermediate solution in which age is included with a polynomial (df = degrees of freedom of the polynomial, here 4).
Fourth panel shows an intermediate solution in which age is modeled with the help of splines (df = degrees of freedom of the splines, here 4).

22 47 210

Reposted by Dylan Burton

Language Testing @langtestjournal.bsky.social · Jul 8

Call for Special Issue abstracts is now open! The Language Testing Special Issue theme of 2027 is “Language Assessment and Policy”. Abstract submission deadline: October 1, 2025. journals.sagepub.com/page/ltj/spe...

3 2

Reposted by Dylan Burton

Dan Isbell @danielrisbell.bsky.social · Jul 8

Our study of English proficiency tests in US R1 university admissions circa late 2023 has been published in #LanguageTesting. Some highlights:
- TOEFL, IELTS, DET, PTE are most widely used
- Most institutions accept around 5 different tests
- Some mismatches in cut scores across tests
...

A survey of English language proficiency tests in international student admissions at US research-intensive universities - Nicholas Coney, Daniel R. Isbell, 2025

Over one million international students are enrolled in US higher education institutions. To be admitted into one of these degree-granting institutions, interna...

journals.sagepub.com

1 1 5

Reposted by Dylan Burton

Julia M. Rohrer @dingdingpeng.the100.ci · Jul 8

Let's say you want to include age as a predictor in your model. How do you do that?

Here's an illustration of three options -- it's for a paper I'm working on (so if you feel like anything could be tweaked...).

Plot that depicts the average importance people in my data assign to their friendships (y-axis, on a scale from 1 to 5, depicted with 95% confidence intervals) by their age (x-axis, from 18 to 60).

Depicted are 3 different ways to model importance of friends as a function of age.
Using age as a linear predictor: this imposes a linear trajectory which comes with very tight confidence intervals (i.e., uncertainty is low).
Using age as a categorical predictor: this imposes no trajectory whatsoever but instead simply reproduces the means by age. The confidence intervals are very wide, in particular for those ages not well represented in the data (i.e., uncertainty is high).
Age splines: This results in a smooth trajectory that follows some of the bumps in the data, but not all of them. The confidence intervals are somewhere between the linear and the categorical case (i.e., uncertainty is medium)

33 31 160

Reposted by Dylan Burton

Language Testing @langtestjournal.bsky.social · Jul 1

Now available in Online First, Michael Suhan and Mikyung Kim Wolf compare the performance of GPT-4, human ratings, and an existing automated writing evaluation model in scoring the writing responses of young EFL learners on the TOEFL Junior Writing rest.
journals.sagepub.com/doi/10.1177/...

2 3

Reposted by Dylan Burton

Julia M. Rohrer @dingdingpeng.the100.ci · May 30

Kowall et al. published a multi-analyst epi study on the effect of marital status on cardiovascular disease (link.springer.com/article/10.1...). What can be learned from this effort? Check out our new commentary! link.springer.com/article/10.1...
with @mendelrandom.bsky.social & Marcus Munafò

What can be learned when multiple analysts arrive at different estimates

Kowall et al. (2025) have brought the multi-analyst approach to epidemiology, with instructive results. With this approach, multiple researchers analyze the same data set to answer the same research question. Or, at least, what appears to be the same research question. In fact, the very first paper that popularized this approach [1] has been criticized for using a vague research question (“Are soccer referees more likely to give red cards to players with dark skin tone than light skin tone?”), which apparently led the analysts to try to answer quite different questions—from a simple question about a bivariate statistical association to much more complex causal questions about racial bias [2].

In contrast, the research question passed on to analysts in Kowall et al. (2025) seems less ambiguous: Does marital status influence the incidence of cardiovascular disease? But less ambiguous does not equal unambiguous. Considerable...

5 31 100

Dylan Burton @dylanburton.bsky.social · May 26

Waking up in Salzburg ready for the kickoff of #EALTA2025! Karen Dunn and I are starting our workshop today on modeling clustered and repeated language assessment data in R.

Dylan Burton @dylanburton.bsky.social · May 25

Congratulations! 🎉🎉🎉

Reposted by Dylan Burton

Dan Isbell @danielrisbell.bsky.social · May 19

Luke Plonsky has put together a directory of Applied Linguistics graduate programs in the US (in progress! get in touch if you have any others to add) over at Applied Linguistics Press. Great resource for prospective grad students and those interested in taking stock of the field.

Grad Programs

This page serves as a directory of graduate-level programs in applied linguistics. Its main purpose is to help prospective students find PhD/EdD programs that might match their scholarly interests as ...

www.appliedlinguisticspress.org

2 7

Reposted by Dylan Burton

Julia M. Rohrer @dingdingpeng.the100.ci · May 11

Thanks to everybody who chimed in!

I arrived at the conclusion that (1) there's a lot of interesting stuff about interactions and (2) the figure I was looking for does not exist.

So, I made it myself! Here's a simple illustration of how to control for confounding in interactions:>

69 280 1.1K

Reposted by Dylan Burton

Betsy Sneller @betsysneller.bsky.social · Apr 28

✨ ATTN Education and Sociolinguistics friends! ✨

I'm hiring a postdoc, to work with the MI Diaries Project! This position is renewable for up to 3 years total.

Please share widely!

careers.msu.edu/jobs/researc...

Research Associate-Fixed Term - East Lansing, Michigan, United States

Working/Functional Title Linguistics; Postdoctoral Research Associate Position Summary The successful applicant will join the MI Diaries project, which is collecting sociolinguistic data from Michigan...

careers.msu.edu

1 20 22

Reposted by Dylan Burton

Dylan Burton @dylanburton.bsky.social · Apr 14

The deadline for ECOLT proposals has been extended to April 27th! We look forward to seeing you at Georgetown University in September.

Dylan Burton @dylanburton.bsky.social · Apr 1

Remember to submit your proposals for the East Coast Organization of Language Testers conference, which will be held at Georgetown University on September 26, 2025! The deadline is April 20th.

sites.google.com/view/ecolt/e...

ECOLT - Call for Proposals

The ECOLT 2025 Call for Proposals is now open Submit here today!

sites.google.com