Philip Skotzke

The final round of thesis proofreading has started. 🤓

On paper, to get an impression of the looks when printed.

🔜 Submission next week

Link to the study:

doi.org/10.3390/covi...

I did:

➡️ use multinomial logistic regression and bootstrapping
➡️ improved the graphical presentation of the results
➡️ and did all of this in a technically reproducible Quarto script

Again, supplementary materials including the data set, the R code and information on the analysis process are on OSF

Just published 🥳

Following the study in Ghanaian students, this study looks into factors affecting vaccination hesitancy in Ukrainian students 💉🦠

This was a great learning opportunity and I can see how my contribution to the paper (doing the statistics) has improved compared to the first study

⬇️

The use of muscle near-infrared spectroscopy in sport, health and medical sciences: recent developments | Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Scie... Near-infrared spectroscopy (NIRS) has been shown to be one of the tools that can measure oxygenation in muscle and other tissues in vivo. This review paper highlights the progress, specifically in thi...

Found in Hamaoka et al. 2011
doi.org/10.1098/rsta...

Cycling kit with an integrated full-body NIRS system at the Tour de France? 🚴‍♂️
We surely are far away from the vision presented in a review I was reading today.

But I have to admit, such a technology would be highly intriguing, especially with the integrated display to oxygenation status 🤓

My first time at the #ECSS2025

I'm taking home many positive impressions - I've never seen this many sport scientists in one place!

I enjoyed having Stephen Seiler as chair during my oral presentation. Finally I got to tell him what my thesis is all about 🤓

Our new durability paper is out in SJMSS!

We tested how endurance performance determinants change after heavy-intensity running (90 and 120 min) in well-trained marathoners.

📌Below details of what we’ve found.

🆓 Open-Access
📄 onlinelibrary.wiley.com/doi/10.1111/...

🧵1/14

Broadly speaking, I agree with @thuyvytnguyen.bsky.social‬’s basic points. Good preregs take time and effort, analyses plans are very important, and clear hypotheses guide the analysis plan.

Only recently I complained about the use of tests for differences with NIRS data compared to other methods.

I am pleased to see that testing for equivalence was used in this study, confirming that BPs and LTs are not the same ⬇️

www.frontiersin.org/journals/phy...

Exploratory research in sport and exercise science: Perceptions, challenges, and recommendations Quantitative exploratory research implies a flexible examination of a dataset with the purpose of finding patterns, associations, and interactions between variables to help formulate a hypothesis, ...

The latest from myself, @crist14n.bsky.social, Grant Abt, and @lakens.bsky.social on exploratory research in sport/exercise science.
doi.org/10.1080/0264...
We outline the differences between exploratory & confirmatory studies and best practice for conducting/reporting an exploratory study.

No, in this case I just uploaded .R files to OSF. Not happy with this, but still looking for better ways to do it.
Using GitHub in didn't feel "right" when I tried it for a different project. Maybe I don't know enough about how it works?

Sounds like a great course! I would've liked more guidance

You can find the study here:
www.mdpi.com/2673-8112/5/...

/end

At the same time it was introduced to @cos.io and open science.

I learned a lot about making my work accessible reproducible since I was working on this project last year.
Nonetheless, you can find the repository here:
doi.org/10.17605/OSF...

/7

How did I contribute?
I performed all the statistical analyses. Combining logistic regression and multivariate multiple regression methods with multiple imputation (using the {mice} package) presented a considerable challenge for me 🤯

/6

In a second step, we used multivariate multiple regression to further explore which characteristics influence the 5C:

Religion, study program, vaccine experience, conspiracy mentality, sources of information, vaccine attributes and the vaccine passport had an influence on at least one of the Cs

/5

Using logistic regression, we found that students who are more likely to get vaccinated are:
- Females
- Studying a health related subject
- Akans (ethnic goup)
- more Confident in the vaccine
- have higher Collective responsibility
- have lower Constraints
- have lower Calculation

/4

Surveying students, we tried to understand how students accepting the vaccine 💉 are different from students who are resistant ⛔

A theoretical framework has been proposed to impact vaccination hesitancy - the 5Cs:
- Confidence
- Complacency
- Constraints
- Calculation
- Collective Responsibility

/3

What is the study about? (my naive explanation)

Once having an effective vaccine against COVID-19 🦠, getting large portions of the population vaccinated was the key intervention in the pandemic.

In Ghana the vaccine was available early on in the pandemic, but many people did not get vaccinated. /2

Just published 🥳

I am grateful to have been given the chance to be part of this project, which is potentially much more relevant than anything I am currently doing in sports science 👀

Under the guidance of Werner Pitsch, I was able to put my statistical skills to the test 🤓
A short 🧵⬇️ /1

A one-stage linear mixed effects model meta-analysis based on our individual participant dataset (but not peer reviewed) showing individual athlete percentage change scores for both VO2max (maximal oxygen uptake) and time-trial (TT) performance, for 236 trained athletes in 6 studies. The relationship (correlation) between change in VO2max and change in performance was not significant and not meaningful. The overall change in VO2max only explained around 25% of the total variance in the change in endurance performance, across athletes.

In our recent meta-analysis with trained athletes, did the athletes who improved their VO2max more also improve their endurance performance more?🤔

No! 😮
In this dataset, +1% improvement in VO2max was associated with a negligible +0.04% change in TT performance. But why? 1/15🧵

Is there an equivalent of t test or anova when the looking of differences between functions rather than numeric values?

Random Sunday thought:
The Joyner Model is typically used to predict performance for a specific distance/time.
Has anyone tried to predict the power/speed duration curve using the model?

Note, I don't want to say this study did something wrong. They are simply doing what everyone is doing

Another study found no mean differences between local and systemic breakpoints, but observed large individual differences.

As this is reported again and again, maybe it is time to actually test for equivalence instead for differences? 🤔

In my case it's about making the results (how does a difference in predictor X explains outcome Y) accessible to people without much statistical training.
Phrased differently, decision makers in public health should be able to take informed actions to target specific predictors/the groups.

Thanks for the reply, I will check it out?

What is the best way to report/visualize a multinomial logistic regression model?
Any reading recommendations/examples?
#stats #rstats

Using regular testing not only allows to track progress, but also keeps me as a coach responsible.

Combining the insights of lab tests with field tests has proven very useful to track longitudinal changes with a higher frequency.

Interestingly, note how the estimated MLSS (🟡, mLMT method) is lower than CP (🔵), but both track very well with each other.