I came across this blog on how to give good scientific talks. I’m up to the second post and I agree with almost all of it so far. Eliminating words on slides and using presenter notes is something I’ve recently started to do. It takes some more practice but I feel like people are indeed more engaged in the talk.
A few months ago Philipp Berens and I ran a six week course introducing the Generalised Linear Model to graduate students. We decided do run the course in Python and to use the IPython Notebook to present the lectures. You can view the lecture notes here and get the source code here.
The six lectures were:
- Introduction by us, then student group presentations of a points of significance article that we hoped would provide a refresher on basic statistics.
- Finishing off student presentations (which took too long in the first week), followed by an introduction to Python, dataframes and exploratory plotting.
- Design matrices
- ANOVA and link functions
- Current issues in statistical practice, in which we touched on things like exploratory vs confirmatory data analysis, p-hacking, and the idea that statistics is not simply a cookbook one can mechanically follow.
- In general the feedback suggests that most students felt they benefitted from the course. Many reported that they enjoyed using the notebook and working through the course materials with a few exercises themselves.
- The main goal of the course was to give students a broad understanding of how GLMs work and how many statistical procedures can be thought of as special cases of the GLM. Our teaching evaluations suggest that many people felt they achieved this.
- The notebooks allow tying the theory and equations very concretely to the computations one performs to do the analysis. I think many students found this helpful, particularly in working through the design matrices
- Many students felt the course was too short for the material that we wanted to cover. I agree.
- Some students found the lectures (for which weeks 2–5 involved me working through notebooks and presenting extra material) boring.
- Despite the niceness of the Anaconda distribution, Python is unfortunately still not as easy to set up as (for example) MATLAB across different environments (primarily Windows, some OSX). We spent more time than we would have liked troubleshooting issues (mainly to do with different Windows versions not playing nicely with other things).
- We didn’t spend a lot of time discussing the homework assignments in class.
- Some students (graduate students in the neural behavioural school) are more familiar with MATLAB and strongly preferred that we teach the course using that.
If I were to run the course again
I think the main thing I would change if I ran the course again would be to have it run longer. As a “taste test” I think six weeks was fine, but to really get into the materials would require more lectures.
I also think it would be beneficial to split the content into lectures and practicals. I think the IPython notebooks in the way I used them would be excellent for teaching practical / small class tutorials, but that the lectures probably benefit from higher-level overviews and less scrolling over code*.
I would also plan the homework projects better. One student suggested that rather than having the introduction presentations at the beginning of the course, it would be nice to have the last lecture dedicated to students running through their analysis of a particular dataset in the notebook. I think that’s a great idea.
Finally, I would ignore requests to do the course in MATLAB. Part of why we wanted to use Python or R to do the course was to equip students with tools that they could continue using (for free) if they leave the university environment. Perhaps this is more valuable to undergraduates than PhD students (who already have some MATLAB experience), but I think it’s good for those students to get exposed to free alternatives as well. Plus, the IPython notebooks are just, well, pretty boss.
*I know you can hide cells using the slide mode, but I found slide mode in general quite clunky and so I opted not to use it.
I came across this article on retraction watch. It offers some practical and concrete advice for setting up expectations and checks when beginning new collaborations. I’m filing it here for posterity – I’d like to implement many of the suggestions.
Sensitivity to spatial distributions of luminance is a fundamental component of visual encoding. In vision science, contrast sensitivity has been studied most often using stimuli like sinusoidal luminance patterns (“gratings”). The contrast information in these stimuli is limited to a narrow band of scales and orientations. Visual encoding can be studied using these tiny building blocks, and the hope is that the full architectural design will become clear if we piece together enough of the tiny bricks and examine how they interact.
Real world visual input is messy and the interactions between the building blocks are strong. Furthermore, in the real world we make about 3 eye movements per second, whereas in traditional experiments observers keep their eyes still.
In this paper, we attempt to quantify the effects of this real-world mess on sensitivity to luminance contrast. We had observers watch naturalistic movie stimuli (a nature documentary) while their gaze position was tracked. Every few seconds, a local patch of the movie was modified by increasing its contrast. The patch was yoked to the observer’s gaze position, so that it always remained at the same position on the retina as the observer moved their eyes. The observer had to report the position of the target relative to their centre of gaze.
We quantify how sensitivity depended on various factors including when and where observers moved their eyes and the content (in terms of low-level image structure) of the movies. Then we asked how well one traditional model of contrast processing could account for our data, comparing it to an atheoretical generalised linear model. In short, the parameters of the mechanistic model were poorly constrained by our data and hard to interpret, whereas the atheoretical model did just as well in terms of prediction. More complex mechanistic models may need to be tested to provide concise descriptions of behaviour in our task.
I’m really happy that this paper is finally out. It’s been a hard slog – the data were collected way back in 2010, and since then it has been through countless internal iterations. Happily, the reviews we received upon eventual submission were very positive (thanks to our two anonymous reviewers for slogging through this long paper and suggesting some great improvements). I’ve got plenty of ideas for what I would do differently next time we try something like this, and it’s something I intend to follow up sometime. In the meantime, I’m interested to see what the larger community think about this work!
To summarise, you might like to check out the paper if you’re interested in:
- contrast perception / processing
- eye movements
- naturalistic stimuli and image structure
- Bayesian estimation of model parameters
- combining all of the above in one piece of work
You can download the paper here.
The data analysis and modelling were all done in R; you can find the data and code here. Note that I’m violating one of my own suggestions (see also here) in that the data are not shared in a text-only format. This is purely a constraint of file size – R binaries are much more compressed than the raw csv files, so they fit into a Github repository. Not ideal, but the best I can do right now.
A few months ago I attended a talk given by a professor of cognitive neuroscience / psychology. The professor presented several experiments to support a particular hypothesis, including both behavioural studies and fMRI. The final minutes of the presentation were used to tell us about some exciting new findings that could suggest an interesting new effect. In presenting these results, the professor stated “we’ve only run 10 subjects so far and this difference is not quite significant yet, but we’re collecting a few more people and we expect it to drop under .05″.
This is an example of a “researcher degree of freedom”, “questionable research practice”, or “p-hacking” (specifically, we could call this example “data-peeking”). In my experience it is very common in experimental psychology, and recent publications show it’s a problem much more broadly (see e.g. here, here and this article by Chris Chambers).
Why does data-peeking happen? I believe that in almost all cases there is no malicious intent to mislead, but rather that it arises from a faulty intuition.
Researchers intuit that having more data should lead to a better estimate of the true effect. The intuition is correct, but where people go wrong is assuming it applies to statistical testing too. Unfortunately, many researchers (including my former self) don’t understand this, and erroneously rely on their intuition.
The Garden of Forking Paths *
It essentially boils down to this: if your data depend on your analyses or your analyses depend on your data, you could be on thin inferential ice. Daniel Lakens has a nice post on the former, while Gelman and Loken have an article on the latter that’s well worth your time (now published in revised form here).
Data depend on analyses
An example of this is if you test a few subjects, check the result, and maybe collect some more data because it’s “trending towards significance” (as for our anonymous professor, above). If you just apply a p-value as normal, it means that your false positive rate is no longer equal to the nominal alpha level (e.g., 0.05), but is actually higher. You’re more likely to reject the null hypothesis and call something significant if you data peek – unless you apply statistical corrections for your stopping rules (called “sequential testing” in the clinical trials literature; see this post by Daniel Lakens has some info on how to correct for this).
Analyses depend on data
An example of this is if you collect 20 subjects, then realise two of them show some “outlier-like” behavior that you hadn’t anticipated: reaction times that are too fast to be task-related. You decide to exclude the trials with “too-fast” reaction times, defining “too-fast” based on the observed RT distribution**. This neatens up your dataset — but given different data (say, those two subjects behaved like the others), your analysis would have looked different. In this case, your analyses are dependent on the data.
I believe this happens all the time in experimental psychology / neuroscience. Other examples include grouping levels of a categorical variable differently, defining which “multiple comparisons” to correct for, defining cutoffs for “regions of interest”… When your analyses depend on the data, you’re doing exploratory data analysis. Why is that a problem? By making data-dependent decisions you’ve likely managed to fit noise in your data, and this won’t hold for new, noisy data: you’re increasing the chance that your findings for this dataset won’t generalise to a new dataset.
Exploratory analyses are often very informative — but they should be labelled as such. As above, your actual false positive rate will be higher than your nominal false positive rate (alpha level) when you use a p-value.
We should be doing more confirmatory research studies
For experimental scientists, the best way to ensure that your findings are robust is to run a confirmatory study. This means
- Collect the data with a pre-specified plan: how many participants, then stop. If you plan on having contingent stopping rules (doing sequential testing) then follow the appropriate corrections for any test statistics you use to make inferential decisions.
- Analyse the data with an analysis pipeline (from data cleaning to model fitting and inference) that has been prespecified, without seeing the data.
- Report the results of those analyses.
If you started out with a data-dependent (exploratory) analysis, report it as such in the paper, then add a confirmatory experiment ***. There’s a great example of this approach from experimental psychology (in this case, a negative example). Nosek, Spies and Motyl report finding that political moderates were better at matching the contrast (shade of grey) of a word than those with more extreme (left or right) political ideologies (p = 0.01, N = 1,979 — it was an online study). Punch line: “Political extremists perceive the world in black and white — literally and figuratively”. However, the authors were aware that they had made several data-dependent analysis decisions, so before rushing off to publish their finding, they decided to run a direct confirmatory study. New N = 1,300, new p-value = 0.59.
The best way to show the community that your analysis is confirmatory is to pre-register your study. One option is to go on something like the Open Science Framework, submit a document with your methods and detailed analysis plan, then register it. The project can stay private (so nobody else can see it), but now there’s a record of your analyis plan registered before data collection. A better option is to submit a fully registered report. In this case, you can send your introduction, method and analysis plan to a journal, where it is peer reviewed and feedback given — all before the data are collected. Taking amendments into account, you then run off and collect the data, analyse it as agreed, and report the results. In a registered report format, the journal agrees to publish it no matter the result. If the study is truly confirmatory, a null result can be informative too.
Of course, there’s still trust involved in both of these options – and that’s ok. It’s hard to stop people from outright lying. I don’t think that’s a big problem, because the vast majority of scientists really want to do the right thing. It’s more that people just don’t realise that contingent analyses can be a problem, or they convince themselves that their analysis is fine. Pre-registration can help you convince yourself that you’ve really run a confirmatory study.
I hope the considerations above are familiar to you already — but if you’re like many people in experimental psychology, neuroscience, and beyond, maybe this is the first you’ve heard of it. In practice, most people I know (including myself) are doing exploratory studies all the time. Full disclosure: I’ve never reported a truly confirmatory study, ever. In a follow-up post, I’m going to speculate about how the recommendations above might be practically implemented for someone like me.
** Note: this is a very bad idea, because it ignores any theoretical justification for what “too fast to be task-related” is. I use it only for example here. I have a more general problem with outlier removal, too: unless the data is wrong (e.g. equipment broke), then change the model, not the data. For example, if your data have a few outliers, run a robust regression (i.e. don’t assume Gaussian noise). Of course, this is another example of a data-dependent analysis. Run a confirmatory experiment…
*** An equivalent method is to keep a holdout set of data that’s only analysed at the end — if your conclusions hold, you’re probably ok.