## ── Attaching packages ─────────────────────────────────────── tidyverse 1.3.2 ──
## ✔ ggplot2 3.4.0 ✔ purrr 0.3.4
## ✔ tibble 3.1.7 ✔ dplyr 1.0.9
## ✔ tidyr 1.2.0 ✔ stringr 1.5.0
## ✔ readr 2.1.2 ✔ forcats 0.5.1
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag() masks stats::lag()
## here() starts at /home/vboyce/Research/mia23
These visualizations are based on 139 submissions.
dat <- semantic |> select(-Timestamp, -`Email Address`, -Name, -`SUNet ID`) |>
pivot_longer(everything()) |> separate(name, c("type", "instance"))
ggplot(dat, aes(x=reorder(instance, value), y=value, color=type))+
geom_jitter(, height=.1, width=.2, alpha=.1)+
stat_summary(fun.data = "mean_cl_boot", color="black")+
facet_wrap(~type, scales="free_x")+
theme(legend.position = "none")+
labs(y="Goodness as example of class", x="")
ggsave(here(images, "dist.png"), dev="png")
There looks to be a correlation between the average ratings and the spread of the ratings. The prototypical examples like apple get almost uniformly rated as a 1 (“good example of the class”) whereas less prototypical examples have less agreement – some people think wresting is a good example of a sport, but others don’t.
Might be something to do with how people are using the scales? Like, maybe some people are thinking “well, an ostrich isn’t that great of a bird, but at least it’s still a bird” and would only use 6 or 7 on the scale for say lizard (it’s such a bad example of a bird, it’s not even a bird). I’m surprised by the ostrich / wren ordering.