bins — sets the number of bars. R figures out the width automatically based on your data range.

binwidth — sets the width of each bar in the actual units of your variable. R figures out how many bars are needed.

So if your data ranges from 0–100:

bins = 10 → 10 bars, each automatically ~10 units wide binwidth = 10 → bars 10 units wide, automatically creates 10 bars Same result here, but they diverge when your data range is irregular. binwidth is generally preferred because it's more interpretable — saying "each bar represents 5 years" means more than "I want 20 bars.

color there is making the OUTLINE of the bars red. fill makes the entire thing red. use fill, then.

it would be horizontal instead- more or less different yeah

bill depth with one continuous line- natually the color is mapped to the geom level, mapped to geompoint (aes(color=bill_dep))

no, they are applied on both local levels and on the global level- yields same result, but global is just easier.

method=lm not added, a regression curve is used making everything more curvy. you have three separate trend lines since the color is specified on mapping, and the graph is stratified as per the island here.

made- it looks almost liek there is no/rnegative relationship, but if you look at each individual species, they have a positive relationship but the trend line is negative. This is an example of confounding as per Simpson's Paradox, where the overall trend reverses as each species is on its own distinct point on the plot, but we see here that each has a positive relationship.. The overall negative trend here is being confounded by species- here species-specific stratificaiton is most important, as the crude trend overall can lie.

na.rm controls whether R warns you about missing (NA) values when it plots. By default it's FALSE, which means R will plot everything it can but show you a warning message like "Removed 2 rows containing missing values" — you've already seen this in your console earlier.

Setting it to TRUE just silences that warning. It still removes the NA rows from the plot either way — the only difference is whether R tells you about it or not.

r ggplot(penguins, aes(x=flipper_length_mm, y=body_mass_g)) + geom_point(na.rm=TRUE) Run that, then run it without na.rm=TRUE and compare your console output. The plot looks identical — but without it you get the warning message, with it you don't.

For your epi work, I'd actually recommend leaving it as FALSE (the default). You want to know when data is missing — missing data is a big deal in epidemiology and silencing warnings about it is a bad habit. But the exercise just wants you to know the argument exists.

in labs(), we then add a line with a comma that says: caption = "......"

in R studio, put them in same line, taking care to add the plus after the function., so:

ggplot (data=penguins) + geom_point()

without even needing to try, species has 3 distinct classes- this would not be properly represrnted on a scatter. instead, try a bar graph., i.e, geom_bar()

numeric variable indicating the bill depth of a penguin in millimeters(mm)

344 rows, 8 column.

When you put something inside ggplot(aes(...)), it applies to everything you add after it — every geom_point(), geom_line(), geom_smooth(), all of them. That's the "global" level. In other words, you will see that there is GLOBAL APPLICATION OF COLOR=SPECIES HERE, such that color is applied to the LINE POINTS AND SMOOTH CURVE, creating 3 different lines per color/species.

When you put something inside a specific geom like geom_point(aes(...)), it only applies to that one layer. That's the "local" level. As a result, putting color=species here maintains the different colors for the various points, since this is the points layer, BUT only applies "local color" to the points in question.