[EpiData-list] New analysis

EpiData development and support epidata-list at lists.umanitoba.ca
Sat Feb 6 12:25:23 CST 2021


We are gradually adding all functionality of Analysis classic.

The options cannot have a blank from ! to the actual option××:
Freq v1 v19 !ci; 
Not
Freq v1 v19 ! Ci

For select in analysis two approaches :

Read datafile !c ;
Select v19 <> 1 do drop data;
// now we do not need select ;

Tab v2 v4 !pr !d1!t;

// the same analysis for the other part;
Read datafile !c ;
Select v19 = 1 do drop data;
// now we do not need select ;

Tab v2 v4 !pr !d1!t;

Approach 2
Read datafile !c ;
Select v19 = 1 do
Begin
Tab v2 v4 !pr !d1!t;
Tab v2 v6 !pr !d1!t;
.... Any other command within this select

End;
Select v19 <> 1 do
Begin
Tab v2 v4 !pr !d1!t;
Tab v2 v6 !pr !d1!t;
.... Any other command within this select

End;

Best wishes
Jens Lauritsen
EpiData Association 

Den 6. februar 2021 11.06.00 CET, EpiData development and support <epidata-list at lists.umanitoba.ca> skrev:
>Hello, 
>
>For training, I'm trying to understand the new version of "Analysis". 
>
> 
>
>A few comments : 
>
> 
>
>- The command "freq var ! r;" (as shown in the guide) doesn't work, but
>"freq var ! pr" is OK.
>
>Strictly speaking and on the educational plan, the percentages
>presented
>with "freq" are in columns : "freq var ! c;" or "freq var ! pc;" would
>be
>more logical.
>
> 
>
>- To select a part of the file, I noted, (as an example for a
>frequency) the
>command "select (var1 <3) do freq var2 ! pr;". 
>
>   Should we repeat with each command "select (var1 <3) do"? 
>
>  Is there a way, like the old "select", to select part of the file for
>further commands? 
>
> 
>
>- The command (example) "select (var1 = 3) do var1: = 2;" is quite
>complicated compared to that "if ... then ...." 
>
> 
>
>- How to create a new numeric variable with 3 digits and 2 decimal
>places?
>As in the old "define var ###. ##"? 
>
> 
>
>- When will the "regress" commands and the graph commands (including
>scatter
>with a regression line as in the old "Epi-Info" ...? 
>
> 
>
>Thank you for your help.
>
> 
>
>Bernard BRANGER - 44000 NANTES - France
>
>_______________________________________________
>EpiData-list mailing list
>EpiData-list at lists.umanitoba.ca
>http://lists.umanitoba.ca/mailman/listinfo/epidata-list

-- 
Sent from my Android device with K-9 Mail. Please excuse my brevity.


More information about the EpiData-list mailing list