Viewing 3 posts - 1 through 3 (of 3 total)
  • Author
    Posts
  • Rune Lomholt
    Participant
    Post count: 19

    Hello

    I have a dataset with variables that were asked twice, but with different references. I want to compilate a table of summary statistics comparing the following statistics between two versions of the variable: count mean sd var cv p50 skewness. Also I want to do an paired ttest between the two versions of the variable and report the two-tailed p-value in a column added to the table.

    questions:
    Is there some way I can write a code using docx comparing the two variables directly in the table by ex putting the statistics side by side?

    How can I add a column with the p-value from the paired ttest between the two versions of the variable?

    Dr. Attaullah Shah
    Keymaster
    Post count: 106

    Hello Rune
    Welcome to asdocx forum. I think you need some sort of a customized table, for which flexmat is a good choice. flexmat is part of the asdocx package and it is already installed if you have asdocx. There are several examples where I show how to make tables with flexmat, or modify tables that are already made by asdocx. References to these resources are given below. Please note the option transpose, it changes the table layout from wide to long.

    1. Introduction to flexmat : [ view post here ]

    2. Add a Stata matrix to existing asdocx table [view post here ]

    3. Create tables by cells, rows, or columns [ YouTube Video ]

    To give a quick start for your desired solution, let me present the following code:

    * Load some example data
    sysuse bpwide
    
    * Create statistics with asdocx
    asdocx sum bp_before bp_after, stat(count mean sd var p50 skewness) replace transpose
    
      0 |1                                    2                   3 
    ----+-----------------------------------------------------------------------------
      1 |Variables                    bp before            bp after 
    ----+-----------------------------------------------------------------------------
      2 |N                                  120                 120 
      3 |Mean                            156.45           151.35833 
      4 |Std. Dev.                     11.38985            14.17762 
      5 |variance                     129.72857           201.00497 
      6 |Median                           154.5               149.5 
      7 |skewness                       0.54729             0.38843 
    ----------------------------------------------------------------------------------
    
    * Conduct a two sample t-test
    ttest  bp_before == bp_after
    
    * The p-value from the t-test can now be appended to row 8 of the table created by asdoc
    flexmat addrow, data(p-value, `r(p)')  row(8) dec(4)
                                
                              Table: Summary Statistics
      0 |1                                    2                   3 
    ----+-----------------------------------------------------------------------------
      1 |Variables                    bp before            bp after 
    ----+-----------------------------------------------------------------------------
      2 |N                                  120                 120 
      3 |Mean                            156.45           151.35833 
      4 |Std. Dev.                     11.38985            14.17762 
      5 |variance                     129.72857           201.00497 
      6 |Median                           154.5               149.5 
      7 |skewness                       0.54729             0.38843 
      8 |p-value                         0.0011                     
    ----------------------------------------------------------------------------------
    
    * Export the new table to MS Word
    asdocx export
    Rune Lomholt
    Participant
    Post count: 19

    Thank you Dr. Shah!

Viewing 3 posts - 1 through 3 (of 3 total)
  • You must be logged in to reply to this topic.