ggplot plot averages of groups of varibales

Question

I have the following code and data:

library(dplyr)
library(ggplot2)

d %>% 
    ggplot() +
    coord_flip() +
    ggforce::geom_sina(aes(
      x = variable,
      y = value,
      color = stdfvalue),
      method = "counts", maxwidth = 0.7, alpha = 0.7)

How can I draw I "vertical" average line through each of the groups of variables?

That is, start at the average value for "wind" and plot a line down to the average point of "Temp" then down to the average point of "scattered clouds" etc. So plot vertically a line plot of average points of each of the variables.

Data:

d <- structure(list(model = c("2011-10-12 to 2017-10-10", "2010-01-01 to 2015-12-31", 
"2013-07-24 to 2019-07-22", "2010-01-01 to 2015-12-31", "2011-10-12 to 2017-10-10", 
"2013-07-24 to 2019-07-22", "2013-07-24 to 2019-07-22", "2011-10-12 to 2017-10-10", 
"2010-01-01 to 2015-12-31", "2013-07-24 to 2019-07-22", "2013-07-24 to 2019-07-22", 
"2010-01-01 to 2015-12-31", "2013-07-24 to 2019-07-22", "2013-07-24 to 2019-07-22", 
"2013-07-24 to 2019-07-22", "2010-01-01 to 2015-12-31", "2013-07-24 to 2019-07-22", 
"2010-01-01 to 2015-12-31", "2011-10-12 to 2017-10-10", "2011-10-12 to 2017-10-10", 
"2013-07-24 to 2019-07-22", "2010-01-01 to 2015-12-31", "2013-07-24 to 2019-07-22", 
"2013-07-24 to 2019-07-22", "2013-07-24 to 2019-07-22", "2011-10-12 to 2017-10-10", 
"2010-01-01 to 2015-12-31", "2010-01-01 to 2015-12-31", "2010-01-01 to 2015-12-31", 
"2010-01-01 to 2015-12-31", "2010-01-01 to 2015-12-31", "2011-10-12 to 2017-10-10", 
"2011-10-12 to 2017-10-10", "2013-07-24 to 2019-07-22", "2010-01-01 to 2015-12-31", 
"2011-10-12 to 2017-10-10", "2010-01-01 to 2015-12-31", "2010-01-01 to 2015-12-31", 
"2013-07-24 to 2019-07-22", "2011-10-12 to 2017-10-10", "2010-01-01 to 2015-12-31", 
"2010-01-01 to 2015-12-31", "2010-01-01 to 2015-12-31", "2011-10-12 to 2017-10-10", 
"2013-07-24 to 2019-07-22", "2010-01-01 to 2015-12-31", "2013-07-24 to 2019-07-22", 
"2010-01-01 to 2015-12-31", "2010-01-01 to 2015-12-31", "2013-07-24 to 2019-07-22", 
"2010-01-01 to 2015-12-31", "2010-01-01 to 2015-12-31", "2013-07-24 to 2019-07-22", 
"2011-10-12 to 2017-10-10", "2010-01-01 to 2015-12-31", "2011-10-12 to 2017-10-10", 
"2010-01-01 to 2015-12-31", "2010-01-01 to 2015-12-31", "2010-01-01 to 2015-12-31", 
"2011-10-12 to 2017-10-10", "2011-10-12 to 2017-10-10", "2011-10-12 to 2017-10-10", 
"2011-10-12 to 2017-10-10", "2010-01-01 to 2015-12-31", "2013-07-24 to 2019-07-22", 
"2010-01-01 to 2015-12-31", "2013-07-24 to 2019-07-22", "2010-01-01 to 2015-12-31", 
"2010-01-01 to 2015-12-31", "2013-07-24 to 2019-07-22", "2013-07-24 to 2019-07-22", 
"2010-01-01 to 2015-12-31", "2011-10-12 to 2017-10-10", "2011-10-12 to 2017-10-10", 
"2013-07-24 to 2019-07-22", "2011-10-12 to 2017-10-10", "2011-10-12 to 2017-10-10", 
"2013-07-24 to 2019-07-22", "2010-01-01 to 2015-12-31", "2011-10-12 to 2017-10-10", 
"2011-10-12 to 2017-10-10", "2011-10-12 to 2017-10-10", "2010-01-01 to 2015-12-31", 
"2010-01-01 to 2015-12-31", "2010-01-01 to 2015-12-31", "2010-01-01 to 2015-12-31", 
"2010-01-01 to 2015-12-31", "2010-01-01 to 2015-12-31", "2013-07-24 to 2019-07-22", 
"2011-10-12 to 2017-10-10", "2010-01-01 to 2015-12-31", "2011-10-12 to 2017-10-10", 
"2010-01-01 to 2015-12-31", "2013-07-24 to 2019-07-22", "2013-07-24 to 2019-07-22", 
"2011-10-12 to 2017-10-10", "2010-01-01 to 2015-12-31", "2011-10-12 to 2017-10-10", 
"2013-07-24 to 2019-07-22", "2013-07-24 to 2019-07-22"), variable = c("Temp", 
"Holiday", "Humidity", "Temp", "Humidity", "Temp", "Scattered Clouds", 
"Scattered Clouds", "Scattered Clouds", "Humidity", "Holiday", 
"Protocol Active", "Holiday", "Temp", "Wind", "Humidity", "Scattered Clouds", 
"Temp", "Holiday", "Holiday", "Humidity", "Wind", "Protocol Active", 
"Wind", "Protocol Active", "Temp", "Humidity", "Temp", "Temp", 
"Wind", "Holiday", "Wind", "Humidity", "Wind", "Holiday", "Scattered Clouds", 
"Wind", "Humidity", "Humidity", "Scattered Clouds", "Holiday", 
"Holiday", "Wind", "Wind", "Wind", "Holiday", "Scattered Clouds", 
"Wind", "Wind", "Scattered Clouds", "Wind", "Protocol Active", 
"Humidity", "Humidity", "Wind", "Protocol Active", "Protocol Active", 
"Holiday", "Wind", "Humidity", "Protocol Active", "Humidity", 
"Humidity", "Holiday", "Scattered Clouds", "Holiday", "Scattered Clouds", 
"Scattered Clouds", "Temp", "Temp", "Humidity", "Scattered Clouds", 
"Humidity", "Wind", "Holiday", "Temp", "Wind", "Humidity", "Holiday", 
"Scattered Clouds", "Holiday", "Wind", "Scattered Clouds", "Protocol Active", 
"Scattered Clouds", "Humidity", "Humidity", "Scattered Clouds", 
"Wind", "Scattered Clouds", "Protocol Active", "Humidity", "Wind", 
"Scattered Clouds", "Temp", "Temp", "Humidity", "Temp", "Wind", 
"Holiday"), value = c(0.000459374103229493, -0.253675699234009, 
-0.0786094292998314, -0.196785241365433, -0.0280073583126068, 
-0.1349718272686, 0.0337677411735058, -0.199854329228401, -0.151771619915962, 
-0.033388290554285, -0.197295814752579, -0.0150858527049422, 
-0.158438324928284, -0.0510015822947025, -0.256258130073547, 
-0.0824626013636589, -0.0339591130614281, -0.163890063762665, 
-0.0855430290102959, -0.119788505136967, 0.0866491049528122, 
-0.119025081396103, 0.188910558819771, 0.366333365440369, 0.199591532349586, 
0.149334877729416, 0.0670526176691055, 0.0176683943718672, -0.0365814939141273, 
-0.154049754142761, -0.228669106960297, -0.196197375655174, -0.0200027860701084, 
-0.12459571659565, -0.296628773212433, -0.08418869972229, -0.0166760385036469, 
-0.0674063041806221, -0.0115637453272939, -0.0849187523126602, 
-0.309920370578766, -0.255993634462357, -0.122082889080048, -0.100378297269344, 
-0.10309936106205, -0.282891929149628, -0.017831839621067, 0.995135962963104, 
-0.0860754400491714, -0.0530533157289028, 0.690496385097504, 
-0.00516092125326395, 0.122894659638405, -0.0707448422908783, 
-0.207068875432014, 0.00719616562128067, -0.0254188310354948, 
-0.152992576360703, -0.532763719558716, -0.0513307563960552, 
0.000162361699040048, -0.042680311948061, -0.00825674924999475, 
-0.31216362118721, -0.0360889285802841, -0.257501393556595, -0.0178098864853382, 
-0.248683765530586, 0.0639149844646454, 0.16026109457016, 0.0928365215659142, 
-0.198139742016792, 0.10240363329649, 0.0677793100476265, -0.183281898498535, 
-0.158823877573013, 0.795968234539032, 0.057108536362648, -0.227293521165848, 
-0.101821482181549, -0.139844477176666, 0.466279178857803, -0.140976637601852, 
-0.0496001504361629, -0.117273397743702, 0.0588472411036491, 
0.0477366708219051, -0.13063570857048, -0.0677084028720856, -0.124116607010365, 
-0.0330690667033195, 0.13056093454361, -0.0848147794604301, -0.0219597667455673, 
-0.305300086736679, -0.149264678359032, 0.113734841346741, 0.00306061143055558, 
0.376322239637375, -0.139279842376709), rfvalue = c(86.258064516129, 
0, 49.7096774193548, 70.75, 35.3225806451613, 60.1034482758621, 
1, 1, 1, 71.0357142857143, 0, 0, 0, 44.3571428571429, 6.42857142857143, 
61.4838709677419, 0, 62.7407407407407, 0, 0, 34, 4.29032258064516, 
0, 3.06451612903226, 0, 45.3846153846154, 33.5806451612903, 77.9032258064516, 
78.7741935483871, 5.35714285714286, 0, 11.3225806451613, 57.2692307692308, 
6.83870967741935, 0, 1, 3.89473684210526, 59.4838709677419, 55.5714285714286, 
0, 0, 0, 4.88461538461539, 6.14285714285714, 4.52941176470588, 
0, 1, 0.730769230769231, 5.45161290322581, 1, 1.3, 0, 26.1935483870968, 
31.1612903225806, 9.46875, 0, 0, 0, 21.5555555555556, 56.625, 
0, 49.1481481481481, 59.1111111111111, 0, 1, 0, 1, 0, 79.0967741935484, 
44.8518518518519, 34.1290322580645, 1, 25.6774193548387, 3.53571428571429, 
1, 56.6, 1.22222222222222, 72.1481481481482, 0, 1, 0, 0.935483870967742, 
1, 0, 1, 56.258064516129, 64.258064516129, 0, 4.14814814814815, 
0, 0, 15.5666666666667, 4.51612903225806, 1, 77.625, 56.8, 23.6896551724138, 
86.1034482758621, 2.70967741935484, 0), stdfvalue = c(0.864929959091151, 
0, 0.42152133580705, 0.628319713993871, 0.256029684601113, 0.443066319122509, 
1, 1, 1, 0.666830108666843, 0, 0, 0, 0.195424862735886, 0.257195445075949, 
0.5569573283859, 0, 0.501513259940226, 0, 0, 0.240816326530612, 
0.172937293729373, 0, 0.120975734518367, 0, 0.214191815233131, 
0.235992578849722, 0.741573033707865, 0.755362614913177, 0.216595945308817, 
0, 0.455640614800309, 0.508477237048666, 0.27380305362338, 0, 
1, 0.156748306409588, 0.533951762523191, 0.488947786906971, 0, 
0, 0, 0.197258187357197, 0.246004900090711, 0.180293326448518, 
0, 1, 0.0272658035034273, 0.22046204620462, 1, 0.0505610561056106, 
0, 0.151020408163265, 0.208163265306122, 0.384859735973597, 0, 
0, 0, 0.87950128346168, 0.501066790352505, 0, 0.415062186490758, 
0.529663986806844, 0, 1, 0, 1, 0, 0.76046986721144, 0.203205129451128, 
0.242300556586271, 1, 0.145083487940631, 0.140487635324737, 1, 
0.392749747640215, 0.046855115753104, 0.679626193911908, 0, 1, 
0, 0.035250133523233, 1, 0, 1, 0.496846011131725, 0.58886827458256, 
0, 0.164854929968904, 0, 0, 0.0287816944959802, 0.182178217821782, 
1, 0.71862699648374, 0.395933908300423, 0.122218668031476, 0.862468344342713, 
0.10660735408969, 0), mean_value = c(0.0849713237573971, 0.258623607625513, 
0.087893476006846, 0.116846546852014, 0.0762156088812836, 0.108219837388825, 
0.0428427606529557, 0.1136661705191, 0.130059023540346, 0.087893476006846, 
0.174296025264175, 0.0261154186427083, 0.174296025264175, 0.108219837388825, 
0.325793222173917, 0.0778882477085785, 0.0428427606529557, 0.116846546852014, 
0.15248118933341, 0.15248118933341, 0.087893476006846, 0.287538362248933, 
0.21894815838024, 0.325793222173917, 0.21894815838024, 0.0849713237573971, 
0.0778882477085785, 0.116846546852014, 0.116846546852014, 0.287538362248933, 
0.258623607625513, 0.245334523414615, 0.0762156088812836, 0.325793222173917, 
0.258623607625513, 0.1136661705191, 0.287538362248933, 0.0778882477085785, 
0.087893476006846, 0.1136661705191, 0.258623607625513, 0.258623607625513, 
0.287538362248933, 0.245334523414615, 0.325793222173917, 0.258623607625513, 
0.0428427606529557, 0.287538362248933, 0.287538362248933, 0.0428427606529557, 
0.287538362248933, 0.0261154186427083, 0.087893476006846, 0.0762156088812836, 
0.287538362248933, 0.00663847356772084, 0.0261154186427083, 0.258623607625513, 
0.287538362248933, 0.0762156088812836, 0.00663847356772084, 0.0762156088812836, 
0.0762156088812836, 0.258623607625513, 0.0428427606529557, 0.258623607625513, 
0.0428427606529557, 0.130059023540346, 0.116846546852014, 0.108219837388825, 
0.087893476006846, 0.130059023540346, 0.0762156088812836, 0.245334523414615, 
0.174296025264175, 0.0849713237573971, 0.245334523414615, 0.087893476006846, 
0.258623607625513, 0.1136661705191, 0.15248118933341, 0.245334523414615, 
0.130059023540346, 0.0261154186427083, 0.130059023540346, 0.0778882477085785, 
0.0778882477085785, 0.130059023540346, 0.325793222173917, 0.1136661705191, 
0.0261154186427083, 0.0762156088812836, 0.287538362248933, 0.0428427606529557, 
0.108219837388825, 0.0849713237573971, 0.0778882477085785, 0.0849713237573971, 
0.325793222173917, 0.174296025264175)), row.names = c(NA, -100L
), class = "data.frame")

I get an error when trying to run your code. Is ggforce a separate package from ggplot2? — M Waz, Jan 09 '20 at 22:46

dc37 · Answer 1 · 2020-01-09T22:53:51.093

2

You can calculate first the mean for each conditions using dplyr:

library(dplyr)
D_mean <- d %>% group_by(variable) %>% summarise(Mean = mean(value, na.rm = TRUE))

# A tibble: 6 x 2
  variable            Mean
  <chr>              <dbl>
1 Holiday          -0.212 
2 Humidity          0.0184
3 Protocol Active   0.0297
4 Scattered Clouds -0.0961
5 Temp             -0.0573
6 Wind              0.0756

Then, you can use geom_errorbar to draw a line at the mean point of each variable (basically, it will draw an error bar but as you are giving ymin and ymax really close to the mean, you have the illusion of a line):

library(dplyr)
library(ggplot2)
d %>% 
  ggplot() +
  coord_flip() +
  ggforce::geom_sina(aes(
    x = variable,
    y = value,
    color = stdfvalue),
    method = "counts", maxwidth = 0.7, alpha = 0.7)+
  geom_errorbar(inherit.aes = FALSE, data = D_mean, 
                aes(x = variable, y = Mean, 
                    ymin = Mean-0.0001, ymax  = Mean+0.0001), 
                size = 1)

edited Jan 09 '20 at 22:53

answered Jan 09 '20 at 22:50

dc37

15,840
4
15
32

Oh this is great, thanks! Is it possible to link the lines up into a smooth line - all joining with each other? – user8959427 Jan 09 '20 at 22:53
Maybe you can try with `geom_step(inherit.aes = FALSE, data = D_mean, aes(x = variable, y = Mean, group =1))` but I'm not sure that it is a good representation of your data because you will suggest that Holiday and Wind can be associated ... looks weird ? – dc37 Jan 09 '20 at 23:00
To connect lines each other, you can see this post: https://stackoverflow.com/questions/16350720/using-geom-line-with-x-axis-being-factors. Try `geom_step(inherit.aes = FALSE, data = D_mean, aes(x = variable, y = Mean, group =1))` – dc37 Jan 09 '20 at 23:03
Thanks, I Will take a look at it. The variables / data I want to join are not the same as in this plot. I have variables for "Month 1", … "Month 12". So the final plot will have a much clearer relationship than the variables presented here. – user8959427 Jan 09 '20 at 23:04
1

Make senses, so check the post that I sent you the link, the example they used are actually dates (days of the week). It will help you I think. – dc37 Jan 09 '20 at 23:17

ggplot plot averages of groups of varibales

1 Answers1