10.9 Meeting Videos
10.9.1 Cohort 5
Meeting chat log
00:13:43 Njoki Njuki Lucy: Is it best to visualize the variation in a categorical variable with only two levels using a bar chart? If not, what's the chart to use if I may ask?
00:16:00 Ryan Metcalf: Great question Njoki, Categorical, by definition is a set that a variable can have. Say, Male / Female / Other. This example indicates a variable can have three states. It depends on your data set.
00:16:51 Eileen: bar or pie chart?
00:16:51 Ryan Metcalf: There are other forms of presentation other than a bar chart. I.E “quantifying” each category.
00:18:37 Eileen: box chart
00:18:46 Njoki Njuki Lucy: thank you so much everyone :)
00:24:31 lucus w: This website is excellent in determining geom to use: www.data-to-viz.com
00:25:22 Njoki Njuki Lucy: awesome, thanks
00:25:44 Eileen: Box charts are great for showing outliers
00:26:31 Federica Gazzelloni: other interesting resources:
00:26:34 Federica Gazzelloni: https://www.r-graph-gallery.com/ggplot2-package.html
00:26:51 Federica Gazzelloni: http://www.cookbook-r.com/Graphs/
00:34:19 Amitrajit: what is the difference in putting aes() inside geom_count() rather than main ggplot() call?
00:35:38 Ryan Metcalf: Like maybe Supply vs Demand curves?
00:41:16 Federica Gazzelloni: what about the factor() that we add to a variable when we apply a color?
00:42:33 Susie Neilson: I do aes your way Jon!
00:43:07 Federica Gazzelloni: and grouping inside the aes
00:49:27 Amitrajit: thanks!
00:49:32 Federica Gazzelloni: thanks
00:49:35 Njoki Njuki Lucy: thank you, bye
00:49:45 Eileen: Thank you!10.9.2 Cohort 6
Meeting chat log
00:06:21 Matthew Efoli: good evening Daniel and Esmeralda
00:07:39 Matthew Efoli: hello everyone
00:08:08 Daniel Adereti: Hello Matthew!
00:08:44 Daniel Adereti: I guess we can start? so we can finish the 2 chapters as Exploratory Data Analysis is quite long and involved
00:09:04 Freya Watkins: Sounds good! Hi all :)
00:10:55 Freya Watkins: yes can see
00:23:14 Daniel Adereti: na > Not available
00:23:32 Maria Eleni Soilemezidi: rm = remove
00:25:49 Esmeralda Cruz: yes
00:26:29 Esmeralda Cruz: to remove the outliers maybe?
00:29:20 Adeyemi Olusola: No
00:29:22 Freya Watkins: we can't see it no
00:29:27 Maria Eleni Soilemezidi: no we can't see it!
00:29:38 Maria Eleni Soilemezidi: thank you! Yes
00:32:57 Daniel Adereti: Cedric's article is a nice one! Helpful to understand descriptive use case of different plot ideas
00:43:19 Daniel Adereti: we can do the exercises
00:43:27 Esmeralda Cruz: ok
00:43:28 Maria Eleni Soilemezidi: yes, sure!
00:45:20 Adeyemi Olusola: we can try reorder
00:45:28 Adeyemi Olusola: from the previous example
00:51:44 Maria Eleni Soilemezidi: that's a good idea
00:52:28 Daniel Adereti: Thanks!
00:52:42 Daniel Adereti: cut_in_color_graph <- diamonds %>%
group_by(color, cut) %>%
summarise(n = n()) %>%
mutate(proportion_cut_in_color = n/sum(n)) %>%
ggplot(aes(x = color, y = cut))+
geom_tile(aes(fill = proportion_cut_in_color))+
labs(fill = "proportion\ncut in color")
00:53:32 Esmeralda Cruz: 😮
00:53:47 Adeyemi Olusola: smiles
00:54:13 Adeyemi Olusola: but lets try reorder...I think we should be able to pull something from it, though not sure about the heatmap thingy
00:54:26 Adeyemi Olusola: on our own though*
01:05:38 Maria Eleni Soilemezidi: no worries! Thank you for the presentation, Matthew! :)
01:05:39 Freya Watkins: Thanks Matthew!
01:06:44 Maria Eleni Soilemezidi: bye everyone, see you next week!