Stats help!

[IamAporcupine]

Could someone please help me with this - which statistic I can use to compare something like:

group 1 - 18 bananas, 3 are green
group 2 - 46 bananas, 42 are green
group 3 - 78 bananas, 70 are green

thanks!

Found it as soon as I posted

The most basic comparison you could do here is looking at the proportion of green bananas in each group.

The groups are different sizes, so I would start with getting mean, mode, medium, minimum and maximum and quartile values for the % green bananas in a group. I'd plot these in a box and whisker plot to get a sense of the spread of the data (e.g. are the proportions consistent, do they cluster, are they spread widely?).

You could also create a histogram to understand frequency distribution (e.g. how many groups have 10-20% green bananas, how many have 20-30% and so forth).

I'd then want to understand whether the size of the group is correlated with the proportion of green bananas. The easiest way to do this is to create a scatter plot with one variable on each axis. That'll give you a visual indication. But you can also do a correlation analysis to get the R^2 correlation coefficient which will tell you the strength of the relationship, and the p-value which gives you the statistical significance of the result.

Going deeper, Chi-Squared can be used to tell you whether the % green bananas occurred by chance. If there is other data available you could create a regression model which predicts the % of green bananas based on the other variables.

Are you on the same course as me op? At Cardiff? Curious. Good luck btw. It's tough this stats lark :/

@SpaceCadet4000 - thank you so much!

@domton - nope, not a course, just need to add some stats to a presentations I am giving soon