anova error : Error in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...) : 0 (non-NA) cases

[silvio_lefevre]

I get an error when I execute the anova although i haven't N/A values.

Thank you in advance for your help

[startz]

all being FALSE means that there is at least one non-NA case. It is possible that one row of SUPPL_group is not NA while a different row in training is not NA. At least that's a possible explanation.

[silvio_lefevre]

this seems strange to me because I have no N/A in each variable

[startz]

You're right, that proves there are no NAs in the data.

I haven't used the anova_test function--let me ask, is it okay to not specify a formula?

[silvio_lefevre]

I'm not obliged to use the anova_test function. however, i need to use a function to perform a two-factor anova with repeated measurements.

[startz]

According to the help information, one of the arguments of anova_test is formula. You don't seem to have one, but maybe it's optional?

[silvio_lefevre]

I tried but it still doesn't work ...