ivan_ah

Direct link to study abstract: https://psycnet.apa.org/record/2026-79270-001

PDF: https://psycnet.apa.org/fulltext/2026-79270-001.pdf

This is very nice.

I never understood why math is such a divisive topic and gets certain people all defensive. Why is the reaction so different from literacy? For example, I have never heard anyone say "I am not a reading person; I don't like to read." but I have head the "I am not a math person" so many times (I'd estimate 30%+ of population around me).

I think bad math teachers (educated in the education department) and bad textbooks are to blame for this collective trauma inflicted on the general populaiton... grown up adults swerving away aggressively at first mention of an formula or algebraic equation.

Chill, y'all. Some of this stuff[1] was know thousands of years ago... it would take you a few months to (re)learn all of high school math and solve all your math phobia issues. You're an adult now, you can totally handle that shit.

[1] https://minireference.com/static/conceptmaps/math_concepts.p...

Good question. I had to think about it.

In general, the No Bullshit Guide textbooks (of which there are four) differ from other textbooks by being intensely focussed on learner needs: they are written in a conversational stile, get to the point, explain the WHY? behind concepts, and focus on applications rather than theory and formal proofs

This book specifically is special because it uses a computational approach to explain the core ideas of statistics like sampling distributions. Think for loop that generates random samples, computes their means, and plots the results. This allows readers to understand what's going on directly rather than rely on formulas or predefined procedures (recipes) for statistical analysis. See here for more about the UVP of this book: https://minireference.com/blog/nobsstats-sales-pitch/#:~:tex...

See also announce blog post: https://news.ycombinator.com/item?id=45543069

Direct links to previews PDFs: - Part 1: https://minireference.com/static/excerpts/noBSstats_part1_pr... [223pp, 9MB] - Part 2: https://minireference.com/static/excerpts/noBSstats_part2_pr... [191pp, 16MB]

Concept maps: https://minireference.com/static/conceptmaps/statistics_conc...

Would love to hear your feedback.

For a more thorough block list, see: https://gist.github.com/robleh/583165b8e3da40ad0f04154aefa75...

drops the hopepage recommendatios completely, and also suggestions from the right sidebar

make youtube pull-only ; no push

Direct links to the PDF previews:

- https://minireference.com/static/excerpts/noBSstats_part1_pr... [223pp, 9MB]

- https://minireference.com/static/excerpts/noBSstats_part2_pr... [191pp, 16MB]

The concept maps from the book might also be useful to check out: https://minireference.com/static/conceptmaps/statistics_conc...

> Oroko Radio / Accra, Ghana / https://radio.garden/listen/oro-radio/pmCVm8Ts

This is solid!

It's 7pm EDT here = 11pm GMT in Accra, and they are playing some amazing beats. Thanks a lot for the recommendation.

If anyone is looking for an introduction to Python for absolute beginners, you can check out the Python tutorial I just prepared: https://nobsstats.com/tutorials/python_tutorial.html

It includes a section on "punctuation" symbols, which I haven't seen explained concisely anywhere else and might be helpful, even for non-beginners: https://nobsstats.com/tutorials/python_tutorial.html#python-...

code: https://github.com/HyperPotatoNeo/RSA

We could simplify it even further: it's all about intellectual stamina.

From my personal experience and from my observations of others, the limiting factor to learning is the availability of mental energy to persevere/endure in the learning process.

You have a problem. Either an academic homework problem or a real-world problem of some sort (think error message in the console). You try one thing but it doesn't work... then another, and still you're not making progress. You're starting to realize solving this problem is not obvious and will require some "brain sweat" and you have to make a choice whether to invest the hour (or day!) of your time to learn enough to solve your problem.

Learners willingness to persevere in their efforts to solve the problem depends on lots of factors, but most importantly they are conditional on the level of learners' interest/motivation (intrinsic or extrinsic). Ideally, learners will be motivated purely by the "knowledge buzz" of learning new concepts and seeing the connections between them, but any source of motivation is OK: as long as it pushes/pulls you enough to go through the necessary learning and solve the problem.

I recognize that learning to learn is a useful multiplier, but learners need to have base rate of (1) intellectual stamina (brain muscles?) and (2) interest/motivation to push through. Otherwise learning is not going to be happening.

There is no grand theory perhaps, but educational research is not totally useless, people have accumulated a lot of insights over the years and some of these "intuitions" are also backed by studies.

Here are some book-length reviews of currently known things about learning/teaching that I found to have a very high signal-to-noise ratio:

https://www.routledge.com/How-Learning-Happens-Seminal-Works...

https://www.routledge.com/How-Teaching-Happens-Seminal-Works...

Oh yes, I found this in my `~/.gitconfig`:

   [color "diff"]
     plain = normal
     meta = bold
     frag = cyan
     old = red
     new = blue

I guess I changed to match the red-blue color convention I'm used to from latexdiff.

The article mentions the git diffing command `git diff --word-diff`, which is cool, but I find an even better version to be:

   git diff --color-words

which shows words removed in red, and words added in blue. The output produced is similar to `latexdiff` in case you're familiar.

Update: using Sigil to look inside the .epub, I saw it was produced by `pandoc` and the math is rendered as MathML.

Wow very nice. Lots of content in here, with no lengthy explanations but useful point-form intuition.

The .epub has very clean math done in HTML (no images), which is a cool way to do things. I've never seen this before. I wonder what the author used to produce the .epub from the .tex?

+1 for lists. I used to procrastinate and suffer every year to do my taxes. Always submit last minute (stress) or late (stress + penalties), but ever since I created a checklist for all the steps (data extractions, data transform in Excel, data loading into tax software, etc.) the tax season is no longer this bad. It still takes me a day, but no stress... just grind though the checkboxes!

update: perhaps it is the verbose flag, e.g. `claude --verbose` you were talking about?

I like this a lot...

Can you describe how to access this feature? I can't find anything about this online, and I asked Claude on command line and nothing comes up for "learning mode"

Seems very cool.

I love the simulations. They are such a good way to learn STATS... you can still look at the theorem using math notation after, but if you've seen it work first using simulated random samples, then the math will make a lot more sense.

Here is a notebook with some more graphs and visualizations of the CLT: https://nobsstats.com/site/notebooks/28_random_samples/#samp...

runnable link: https://mybinder.org/v2/gh/minireference/noBSstats/main?labp...

Or with and .assign:

    (
        purchases.loc[
            lambda x: x["amount"] < 10 * x.groupby("country")["amount"].transform("median")
        ]
        .assign(total=lambda df: df["amount"] - df["discount"])
        .groupby("country")["total"]
        .sum()
        .reset_index()  # to produce a DataFrame result
    )

Hi Matt! Big fan. The first that I thought when I read this article was "How would Matt use his .assign-chain-kung-fu on this one" and I tried

   import pandas as pd
   purchases = pd.read_csv("https://www.sumsar.net/blog/pandas-feels-clunky-when-coming-from-r/purchases.csv")
   (purchases
     .assign(country_median=lambda df:
         df.groupby("country")["amount"].transform("median")
     )
     .query("amount <= country_median * 10")
     .groupby("country")
     .assign(total=lambda df: (df["amount"] - df["discount"]).sum())
   )

but it seems 'DataFrameGroupBy' object has no attribute 'assign' so its not that simple, though with a slight re-ordering of the chain operations, it works cf. https://news.ycombinator.com/item?id=44236487

Hahahah... Yes, contributing to the frequency tables. At least we're not including the symbols in question in our comments, so as not add to the imbalance!

> do you still intend to publish a guide to statistics?

Yes! Very much so. I'm in the final stages of cooking it right now, hoping to make it available for sale in the coming months.

Statistics is by far the hardest subject I've had to write about, as I had to (re)learn pretty much everything. I did several rewrites (working on tis book for the past seven years!), but now it's finally ready to ship. You can check out the notebooks from the book here: https://nobsstats.com/ and a detailed outline here https://docs.google.com/document/d/1fwep23-95U-w1QMPU31nOvUn...

ETA July for digital download, then one round of copy-editing, and September for first print version.