XaiJu
ncase
ncase

patreon


An experiment... for SCIENCE!

For years, I've been making interactives to explain science.  Today, I launched an interactive to do science!

Specifically, I want to know if interactives actually are better than static content at helping people learn.  Question: does being asked to guess/predict an answer, before being shown the answer, help you retain information better?  Hypothesis: yes, but let's test that.  So on my web page, it randomly assigns you to a static version (control) or interactive version (experiment), then gives you a memory recall test.

The link below is directly to the experimental condition, and the survey at the end will not be analyzed.  As much as I'd love to have you as part of this experiment, I've already shown you too much info + a GIF.  Still, here's the link:

πŸ”¬ the experimental condition! (7 min play)

However, you can still help!  If you'd like to help me get a larger sample size, here's a tweet you can retweet:

🐦 for your retweeting pleasure

And finally, for more background on this experiment + intellectual transparency, here's a blog post "pre-registering" my hypotheses:

πŸ“Š my experiment "pre-registration" (5 min read)

This is just a small experiment to answer a small question, but it's a big step for me in two ways:  1) It'll be nice to finally prove (...or disprove...) the value of "explorables", for my sake & other educators'.  2) This is my first real science experiment, and I'm excited to learn about statistics & apply it to a real problem!  (and in the future, I can further consolidate what I've learnt by making an explorable explanation on statistics)

Either way, I'm excited to see what comes out of this!  I'll analyze & post the results – as an interactive paper – by the end of the month.  See you then!

<3,
~ Nicky

P.S: if you picked the $10/mo+ "badly-drawn peep" reward, I just sent out an email asking for reference photos! please let me know if you didn't get the email, thx

An experiment... for SCIENCE!

Comments

Are you perchance logging the amount of time each group spends on each fact? I'm a believer that interactivity can help, but I also wonder how much of the effect can be predicted by the amount of time spent learning. (Though interactivity and surprise may be essential to keeping focused engaged attention for longer amounts of time, so this wouldn't necessarily be a negative result.)

Vi Hart

Cool experiment. Just read about a similar thing in "Factfulness". One tip: the beautiful font is hard to read for a visually impaired. A tad more contrast (slightly thicker lines) would help a lot. ☺️

Erwin Kooi

I looked over the stats that say that many nations are what they say they are, and I am not buying it at all.

IHateSigningUpForThings

I like it

Kozmobot

Cool, though for your experiment, might have to consider using random fictitious numbers. (Or any other approach, idk.)In case anyone has seen the questions before. I’m guessing it might affect variance or error rate.

Iggy C

Ha, yup! All the data was found through Max Roser & co's amazing site, Our World In Data: <a href="https://ourworldindata.org/" rel="nofollow noopener" target="_blank">https://ourworldindata.org/</a> I haven't read Factfulness yet, but Hans' TED talk was great, and a major moment for dataviz taking the popular imagination!

Nicky Case

I'll take that bet! πŸ‘

Nicky Case

I love this. Thanks for being reflective and actually quantitatively assess what you do. Whatever the outcome, I hope it helps you grow and improve.

Tim S (Banana Juice Tech)

I wish you luck in your scientific endeavors!

CrabWithKnife

Been reading the book β€œFactfulness” by Hans Rosling. I think that biased my answers. Hans Rosling and Max Roser collaborated before.

Iggy C

Great stuff, Nicky :)

Will Harris-Braun

awesome experiment, I put my money on your hypothesis as well. In fact, my prediction is that the guesser group is 30% more accurate than the control :)

David Urbansky


More Creators