nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

354532_en-US

liliconcom

Missing values

Hello,

I want to make analysis (Chi2 test) but I would make it without missing values.

I don't find the formule in order to doesn't take in account this missing values.

Moreover, I would make a Chi2 to compare the proportion with a EVA pain level ≥4 (cf. variable in yellow) according to the Intervention (salmon color) = 0 between the 2 randomization groups (variable in red).

You will also find my data base in file attach.

Thank you +++

Vocab : "Données manquantes" = missing values ; EVA < 4 (code = 1 for EVA variable (in yellow) ; EVA > 4 code = 1 for EVA variable (in yellow) ; "Bras prise en charge habituelle" --> cf. Bras randomisation (in red) ;

Analytics

Re: Missing values

I may have misunderstood your question, but what I think you want is to

1. ignore the missing values in the EVA column (I just selected those rows and hid and excluded them), and

2. look at the Fit Y (yellow) by X (salmon) separately for the 2 randomization groups (red), which I did by putting the randomization groups in the By box.

Attached is the file, and I'm sorry if I did not understand you correctly.

Re: Missing values

Hello,

I didn't understand which column is used to indicate missing data.

In the script below you can substitute that column for Age, and then use the local data filter in the report to filter out the desired rows.

Categorical(
 Structured( :BRAS RANDOMISATON * :Intervention, :EVA ),
 Share Chart( 0 ),
 Legend( 0 ),
 Test Response Homogeneity( 1 ),
 Local Data Filter(
  Add Filter( columns( :Age ), Where( :Age >= 11.94 & :Age <= 17 ) )
 )
)