r - reorder barchart by sum of bar segments with ggplot/plyr -

I need 11 times in the following vertical barplos re-ordered with the sum of the first two segments Each time, i.e. (red + green) sorted into conspiracies.

  & gt; Dput (q1m.bl) structure (list (alcohol = structure (C (1L, 2L, 3L, 4L, 1L, 2L, 3L, 4L, 1L, 2L, 3L, 4 L, 2L, 3L, 4L, 1L, 2L, 3L, 4L, 1L, 2L, 3L, 4L, 1L, 2L, 3L, 4L, 2L, 3L, 4L, 1L, 2L , 3L, 4L, 1L, 2L, 3L, 4L, 1L, 2L, 3L, 4L), .Q = structure (C (1L, 1L, 1L, 1L, 2L, 2L, LLP, 2L, 2L, 3L, 3L, 3L, 3L, 4L, 4L, 4L, 5L, 5L, 5L, 5L, 6L, 6L, 6L, 6L, 7L, 7L, 7L, 7L, 8L, 8L, 8L, 9L, 9L, 9L, 9L, 10L, 10L, 10L, 10L, 11L, 11L, 11L, 11L), Label = a ( "Pushers_a", "Pushers_com", "Pushers_biad" "PUSHERS_SEC", "PUSHERS_STAB", "PUSHERS_COST" "PUSHERS_INNO", "PUSHERS_VAL", "PUSHERS_INDEP", "PUSHERS_STDS", "PUSHERS_SRC"), class = "factor"), Counts = c (1L, 3 L, 4L, 1L, 3L, 3 L, 2L, 1L, 4L, 2L, 2L, 1L, 3 L, 5 L, 1L, 1L, 1L, 6L, 1L, 5 L, 1L, 2L, 1L, 1L, 1L, 6L, 1L, 2L, 6L , 1L, 2L, 4L, 2L, 1L, 3 L, 3 L, 2L, 1L, 2L, 1L, 5 L, 1L), blpos = c (0.111111111111111, .444444444444444, .888888888888889, 1, .333333333333333, 0. 66666666666 6667 .888888888888889, .666666666666667, .888888888888889, 1, .333333333333333, .888888888888889, 1, .111111111111111, .2222222222222, .888888888888889, 1, .555555555555556, .666666666666667, .888888888888889, 1,. 111111111111111, .22222222222222, .888888888888889, 1, .222222222222222, .888888888888889, 1, 0.222226622222222, 0.666666666666667, 0.888888888888889, 1, 0.333333333333333, 0.666666666666667, 0.888888888888889, 1, 0.222222222222222, 0.333333333333333, 0.888888888888889, 1)). Name = C ("item", "mdQ", "calculation", "blosso"), line name = c (na, -42l), class = "data.frame")  

< P> plot ...

  ggplot (q1m.bl, ais (x = item q, y = count, fill = iasa)) + gom_bar (state = "identity", condition = " the ") + Jiom_tekst (AES (Y = Blopo, = label to count), Acjl = 1) + Theme (Axisktextkx = Element_text (angle = 90, hjust = 0), text = Element_text (size = 10)) + Coord_flip ()  

Uh, not enough representative points to embed, ima ges Sorry for the inconvenience. Here's the plot:

I played around with the arrangement of the system and after checking the data frame I thought the following sorting should be trick. (Note: Blops mean "position of bar label" and position of different numbers in conspiracy.) But this "sorted" data leads to a similar conspiracy as above by plotting the frame. I do not know what information to change to change the conspiracy of the item column.

  q1m.bl.s  

What is the best way anyway? Do I DF (ddply / arrange / reorder / etc) before plotting because I think this is a presentation issue and should be done in inside ggplot. Does not matter? Still no one was mentioning the stacked bar segments and using the factor data ... I A new question.

Thank you very much for explaining me!

ItemQ is about to rearrange the factor level of the variable.

Overall (Calculate the D $, list (MDUQ = D $ mDQ), < code> D & lt; - subset (q1m.bl, ItemA %%% c in ( "sehr wichtig", "wichtig")) Yoga & amp; amount) Mdkyurder & lt; - as.c Helitr (total [order ( -Notes $ x),] $ mDQ) q1m.bl $ itemQ & lt; - factor (q1m.bl $ itemcu, level = mdqu Order)

Then you should actually be able to run the code as you provide it and it will create it:

Edit (Conas): Convocation, I am adding my first answer again to the use of DDP Showing, because I do not feel comfortable with it / I do not get it completely, I'm sure others can benefit It is. :-) So, with your permission, I hereby redirect it to:

  Library (PLIR) mDQ. Order & lt; - q1m.bl% & gt;% group_by (mdQ)%>% Filters% C (% s) (%) (%) (%) (%) List% & gt;% as.character q1m.bl $ ItemQ & lt; - factor (q1m.bl $ itemQ, level = itemQ.order)