--residualized_controls

Switch

--residualized_controls

Description

For --nfold_test_regression, additionally reports how well language predicts the residual of the outcome after regressing out the controls — i.e. whether language explains variance beyond what the controls alone explain. Aliased as --res_controls.

Argument and Default Value

No argument. Boolean switch, default: False.

Details

Verified: with --outcome_controls set, output included extra resid_r, resid_rho, resid_train_mean_mae, etc. keys alongside the normal metrics — the residualized-controls comparison.

Other Switches

Required Switches:

• --nfold_test_regression

• --outcome_controls

Example Commands

Using the packaged dla_tutorial dataset (see Differential Language Analysis (DLA) Tutorial):

dlatkInterface.py -d dla_tutorial -t msgs -c user_id -f 'feat$1to3gram$msgs$user_id$0_05' \
--outcome_table blog_outcomes --outcomes age --outcome_controls is_student --combo_test_regression \
--residualized_controls