search

cproctor / qualitative-coding

Qualitative coding for computer scientists
Other
12 stars 3 forks source link

Crosstab counting inconsistency? #55

Closed cmaimone closed 3 months ago

cmaimone commented 3 months ago

Working through the vignette, I ran a crosstab with two different units (default of line, and paragraph). Why are the counts along the diagonal not the same?

(base) /Users/christina/Desktop/qc_demo % qc codes crosstab equity participation committee_participation -a
                   code    equity    participation    committee_participation
-----------------------  --------  ---------------  -------------------------
                 equity        15                7                          0
          participation                         15                          0
committee_participation                                                     9
(base) /Users/christina/Desktop/qc_demo % qc codes crosstab equity participation committee_participation -a --unit paragraph
                   code    equity    participation    committee_participation
-----------------------  --------  ---------------  -------------------------
                 equity        15                7                          0
          participation                         13                          0
committee_participation                                                     6

re: https://github.com/openjournals/joss-reviews/issues/7031

cproctor commented 3 months ago

This is the intended behavior. If the same code is applied to multiple lines within a paragraph, these will be counted separately with a line-based unit of analysis, but will only count once when counting by paragraph. If this behavior is surprising, please let me know and I'll extend the docs to explain. Thanks!