Exclude zero values from stats

Below we have a question, Q1, where 5% of respondents answered "0". For analysis, you may want to exclude zero values in summary statistics like "Mean". One way to exclude zero values is to recode them to [NA]. 

Question "Thinking about your Condition X patients, how many of those are currently receiving a GA (gamma antagonist)?" displaying blue bars with percentages: 0 (5.0%), 1 to 10 (67.0%), 11 to 20 (16.0%), and other ranges. Red box highlights "Mean 11.2" and "N 100" at bottom.

From the element's context menu select "Recode..." to bring up the recode dialog:

Q1 question header with dropdown context menu displaying options: "Select to copy as text", "Copy as image", "Round by..." (E B), "Statistics..." (E S), "Color..." (E O), "Format..." (E F), "Recode..." (E R) highlighted with red border, "Filters..." (E I), "Chart type..." (E H), "More properties..." (E M), "Edit JSON..." (E J). Right side shows partial data with percentages 5.0%, 67.0%, 16.0%, 5.0%, 1.0%, 3.0%, 1.0%, 2.0%, and values 11.2, 100.

From here you can filter, select and drag responses to existing or new categories. For this example, we only want to code zeros to "[NA]". 

Recode dialog with "Codes" section on left containing blue "(New code)" button, "[NA]" entry with count 1, and "0" entry. Right side shows "Uncoded values" section with "(uncoded)" showing count 7, followed by numeric values 10, 20, 30, 50, 80, 40, 70 in separate rows. Cancel and Apply buttons in top-right.

Q1 now doesn't appear to have any "0"s, and "Mean" has increased. You can press the [NA] button to toggle whether missing values are included in the basis or not.

Question "Thinking about your Condition X patients, how many of those are currently receiving a GA (gamma antagonist)?" now displaying "[NA]" at 5.0% as first category, followed by "1 to 10" (67.0%), "11 to 20" (16.0%), and other ranges. Red box highlights updated "Mean 11.8" (increased from 11.2) and "N 100" at bottom.