Composition — tbl_merge() & tbl_stack()

Side-by-side (tbl_merge)

Two tables glued horizontally, sharing the first label column when identical:

t_female = ps.tbl_one(df[df.sex == 'F'], by='arm').add_p()
t_male   = ps.tbl_one(df[df.sex == 'M'], by='arm').add_p()

ps.tbl_merge([t_female, t_male], tab_spanners=['Female', 'Male'])

Requirements:

• All inputs must have the same number of body rows.
• If the first column of every row is identical across tables, it's collapsed to a single label column (share_first_column=True, default).

Vertical (tbl_stack)

Multiple tables concatenated under one header:

ps.tbl_stack(
    [t_cohort_a, t_cohort_b],
    group_labels=['Cohort A', 'Cohort B'],
)

Requirements:

• All inputs must have the same column count and header structure.
• Optional group_labels insert a group-header row between blocks.