What They Forgot to Teach You

Issue #10

👋 Welcome

Hey! There are a lot of new people who have signed up to get this in their inbox in the last couple of weeks so before we get into it, a quick hello.

I'm Jen. I'm a developmental psychologist and a former A/Prof at UNSW, and these days I run Reveal Research, where I coach Honours, GradDip and PhD students through the hidden curriculum of research training.

That's all the stuff nobody actually sits you down and teaches, from managing your supervisor to plotting your data to structuring your discussion. I have supervised more than 30 honours and PhD students in my time and marked more than 100 theses. Part of what made me step off the academic train into the unknown was a sense that research training doesn’t need to be as hard as it is. More and more students are enrolling in research programs, often online without a formal lab support structure, and are expected to just work it out. When your research project is a stepping stone into the career that you want (that may or may not be research based), I don’t think that is fair.

So this newsletter is where I hand over things that I think should be part of the formal curriculum, insider information that makes the process just a bit smoother. Every issue: one useful thing, one paper worth reading, and a standing invitation to come and hang out.

If you ever want to chat about what you think is missing from the curriculum, my calendar is always open. Book a chat using this link.

​https://calendly.com/revealresearchcoach/30min​

Right. This week I owe you a correction and I’m feeling kinda stink about it.

Back in issue #5, when we talked about deadlines and the lies we tell ourselves about being done by Friday, I pointed you to a famous study by Ariely and Wertenbroch (2002) reporting that students did best when someone else imposed deadlines on them. I linked to the study and recommended you try using the findings to help you overcome the problem of putting off your writing until the last minute.

This week that paper, which was originally published in Psychological Science, was retracted because the data appear to have been fabricated.

Turns out that paper was built on numbers that were, at best, not what they claimed to be. It makes me feel icky. But it does give me a reason to write about the all too common occurrence of what to do when your results do not behave.

📖 This week I read

Artificial Deadlines (Part 1): Evidence of Fraud in an Influential Study About Procrastination — Data Colada (2026). Read it here →​

Data Colada is a blog run by three researchers who, among other things, have become the psychology’s most careful data detectives. In this post, they take apart Study 2 of the deadlines paper and the red flags they identify are really interesting. This paper came to their attention because the effect failed to replicate. But on closer inspection that was perhaps not surprising because the effect size was implausibly large, like the pattern was too perfect to be true. As it turns out, the original spreadsheet shows rows of data are duplicated, numbers that should be rounded are not, and correlations that should be strong are somehow non-existent. Read together, it is very hard to explain how this might have some about via an honest mistake.

Psychological Science has now retracted the paper. Sadly this is not the first paper from Ariely that has been found to have “data issues”. A few years ago now, his work claiming that people are less dishonest in completing insurance forms if you have them sign an honesty statement at the top was also found to be fraudulent and was retracted from PNAS. (For a wry take on the most recent case, Andrew Gelman's post titled "I am shocked, shocked" is a good read)

​
​

🎓 The hidden curriculum

This story isn’t really about the integrity of a famous professor.

Nobody (well most people) don’t wake up and decide to fabricate their data. What happens, far more often, is smaller and much more relatable. You run your study. You have a hypothesis you are quietly in love with. And the results come back… nothing. Everything is p > .05 or worse, the effect is significant but in the wrong direction!

And in that moment there is a pull. Maybe I should drop those three participants who "clearly weren't paying attention." Or try the analysis a few different ways to see if one of them turns out to show p < .05. Or rewrite the hypothesis so it matches what I found. You are not Professor Famous changing values in your spreadsheet. But the instinct to want the data to say something it does not is kind of the same.

So what do you do instead, when the data doesn't behave? (And in my experience 80+% of honours students have misbehaving data).

You report what you found. A null result, clearly explained, is a real result. A messy result that you are honest about is worth more than a tidy one nobody can trust. If you drop participants, explain why, and do it before you look at the outcome. If your data makes you think about interesting questions that you hadn’t considered at the beginning of your project, that is ok, just flag those analyses as exploratory. If you changed your mind about your analysis approach after seeing the shape of your data, that is allowed, you just have to tell the reader that is what happened.

But what will my marker think? I hear you say. Examiners are not scoring you on whether your effect was significant because they know real data is lumpy. They are reading for whether you have thought critically about your data, understand your own results and tell the truth about them. "I predicted X, I found nothing, here is why I think that happened and what I think that means" shows more competence than a thesis where every hypothesis was magically confirmed.

So report the null, plot the mess, and write the true story. Your thesis will be stronger for it. But if you are staring at a result that didn't go your way and can't see how to write it up honestly without it sounding like a failure, that is exactly what I can help with.

I have sessions coming up on September 11 and 25 where we walk through outlining your General Discussion, warts and all. It will be a small group, plenty of room to throw ideas around and sense-check your explanations.

Booking link below...

💡 What I'm into

If this saga has you morbidly fascinated (you are not alone), Data Colada and Retraction Watch are both oddly gripping reading. Posts are often part detective story, part cautionary tale about the pressure to produce a significant result and a tidy story. It’s like work-related True Crime.

📅 Join me

Free Friday workshops — one hour, Zoom session, leave with tips and tricks

• How to set yourself up for research success — Friday September 18, 12pm AEST. Just getting started? Put a few project-management systems in place now and save yourself hours later.
• Back by popular demand… How to polish your thesis (and get an HD) is happening again on October 2. If you missed our session last week, it was a blast! Wondering how to push your thesis to HD level, this session is for you. Friday October 2, 12pm AEST.

Small group coaching — N = 5, two hours, Zoom session, bring your work and leave with something you can use.

• Keep Calm and Write Your General Discussion — Fridays September 11 or September 25, 12–2pm AEST. A$49. You have your results (including those that didn't go your way), you just don't know where to start. We work through the six discussion moves together and you leave with an outline.
• Plotting for survey research — Fridays September 18 or October 2, 1–3pm AEST. A$49. Survey data needs more looking, not less. Bring your data and I'll help you build the figures that show the honest shape of it and make your thesis stand out.
• The Messy Middle — Friday October 9, 12-2pm AEST. A$49. Is your research project getting out of hand? Nobody teaches you the bit between starting a project and writing it up, where the data piles up and "I'll sort it later" becomes a problem. Bring your mess and leave with a folder structure that makes sense and a plan that works backwards from your deadline so write-up isn't a panic.

Register for everything at www.revealresearch.org/events.

🛠️ For your toolkit

​A prompt to make GenAI make your think. Tempted to ask ChattieG what he thinks of your argument? Have Claude critique your experimental design? Use this prompt to set up some boundaries so that you avoid outsourcing your thinking entirely.

​The Polish Checklist. Six tests that you can run to check that you are holding your reader's hand, not assuming they know everything you do, and making your argument easy to follow from start to finish. All you need is a highlighter and Ctrl-F.

​Jenny’s very opinionated tips for academic writing. 12 pages listing 48 things that you wish someone had told you about academic writing. Words, sentences, paragraphs, structure, synthesis… this doc is full of all my tips and tricks.

👋 thanks!!

If a result of yours has refused to cooperate lately, hit reply and tell me about it. Honest messes are my favourite kind of email.

One more thing, some of you have followed my daughter's Google Sheet based puppy campaign. I am pleased to announce that Team Reveal Research's newest hire, Bingo, arrives this weekend.

This means that next week I will be attempting coaching calls, Writing Reviews and this very newsletter with a puppy underfoot. If you have worked from home with a new puppy and lived to tell the tale, hit reply with your best tips. I am going to need them!

Until next time, Jen

I'm Jen, a developmental psychologist and former UNSW A/Prof. I help new researchers go from feeling like they've missed something everyone else knows, to feeling like they genuinely belong in research, so they can get through their research degree and into the career they actually want. Learn more