library(tadaatoolbox) # remotes::install_github("tadaadata/tadaatoolbox")

As you can see, there's quite a nice distinction. Especially the second plot shows an obvious difference which is so big a statistical test for significance would be entirely pointless. But who am I to judge, here's a t-test.

```{r}

tadaa_t.test(xfiles, rating, plotarc, print = "markdown")

```

If you're not familiar with t-tests and power analysis: That's ridiculous. The p-value indicates significance by it's own right, and the effect size of $d = 1.421$ tells use the effect is *huge*, with effect sizes greater than 0.8 commonly referred to as large. Also, the power is 1 (rounded value, so *very close to 1*), which basically means that with that data it's theoretically next to impossible no not spot that difference if our data was a sample of a larger "population" of episodes.