psyteachr / quant-fun-v2 Goto Github PK

As part of the Nature reviews psychology paper, we're citing each book and I never realised they had DOIs through Zenodo. Particularly for users outside Glasgow who want to cite the books, it might be worth adding a "here's how to cite this book" entry on each overview page.

On line 387, there should be a link to the Shiny app for playing around with a normal disitribution:
access it here

However, it comes up as page not found:
Page not found
Sorry, but the page you requested doesn't exist.

Not a terminal problem, but it seems Hmisc is not needed in chapter 10, but the instructions tell students to load Hmisc. Not a problem itself, but we had some install issues which means this could be avoided if they were not prompted to load it.

The below should talk about performance not qqplotr

First, we can use check_model() to produce a range of assumption test visualisations. Helpfully, this function also provides a brief explanation of what you should be looking for in each plot - if only all functions in R were so helpful!

If you get the error message Failed with error: ‘there is no package called ‘qqplotr’’, install the package qqplotr, you don't need to load it using library(), but check_model() uses it in the background.

If your check_model() plots are not showing, try maximising your plot window.

On line 223 there is the sentence: "Type and run the below code in a new code chunk. When you have run the code have a look at sort_desc and note that the data is not sorted by descending year!".

This should be "Type and run the below code in a new code chunk. When you have run the code have a look at sort_desc and note that the data is now sorted by descending year!" as we're demonstrating the data should be in descending order now.

The assumptions related to the homogeneity of variance as differentiating Welchand Student's t-test is a bit awkward. in the bullet list for Welch's assumptions add a fourth bullet point:
Not important for Welch's test, but worthwhile knowing if you use students' t-test is that the two groups need to have similar variance and normally this means they have to have a very similar number of participants. This is not an assumption of Welch's t-test. Why is it useful for the curent anaqlysis? MCQ choices
1 (correct) Men and women are very different and also there are different numbers of men and women in this study
2 (incorrect) My creativity is sapped...

Line 231 is currently: 7.7 Activity 7: Daving plots (#introviz-a7)

I think it should be Saving plots {#introviz-a7} to render properly

t-test formula is shown as welch's but uses student's df. Needs resolved.

A student in one of my support sessions came to me with problems on activity 9 of chapter 16. She was entering "0.22" for the question on minimum effect size, but the box was telling her it was wrong.

I did some digging and found that the fitb function accepts ".22", even though "0.22" is listed as an argument. Sorry, I'm not familiar with the function, so I don't know how to fix it!

On line chapter 10 line 180, there is supposed to be a link to towards data science to explain boxplot, but looks like the article has been taken down. Looks like it should be something like this just to explain what a boxplot does: https://builtin.com/data-science/boxplot