r - Tufte tables: convert quartile plots into standard error plots hacking qTable function from NMOF package -

If you remember that Tuff is a good version of the table created in which the small quadrant running next to the inserted data rows The plot includes:

This One such implementation is the use of the solution and the work qTable , which basically creates the above table and outputs it as a latex code:

  Vshykta is (Anmof) x & lt; - rnorm (- 100, mean = 1, sd = 0.5) X <- cbind (x, 0, 0, 0, 0, 0, 0, 0, 0, 0, sd = 2) y & lt; - Rnorm (100, mean = 1, sd = 2) z   
 This method of visualization is particularly useful if you have a little information to present, and you do not want to waste a space for the full graph but I  qTable  Would like to hack a bit of quantity Instead of publishing, I mean like to display the standard errors. I am not great at hacking in such works, and I used animal force to do it. I replaced the line with the  qTable  function that measures the amount:  
  a < - Applicable (X, 2L, quantity, C (0.25, 0.5, 0.75))   
 Some very cruel, which calculates standard errors:  
  to calculate columns (mental) #got 'se' function column standard deviation M = callman (x) means SE = SD (x) / SKRTR (nO (x)) # standard error SELo = M -SE # bound bound SEHi = M + SE # upper bound A = t (data.frame (SELo, M, I know it, I know that this is possibly obsolete approach, but it actually does have some vista Works for - It imposes standard errors, but keeps this difference in this plot:   
 P> 
  
 To overcome the interval, you can insert these two rows: b [3,] b [4, ]  [<3],   
 right before  for  loop that starts with  
  For (CC1L: Nd (X) [2L]) This will for {...   
? Reading the graph from left to right, with the five rows of  B   
  1) the left segments begin 2) the left segments are finished 3) the dots are 4) Correct section ends with 5)   
 by doing this  B [2,]  and  b [4],  From  b [3,]  you are getting rid of the gap.