“…We adopt this approach to examine the relationship between different nursing staff inputs and patient satisfaction measures in hospitals. We posit a general production function of the form: Y = F ( N 1 , N 1 2 , N 1 3 , N 2 , N 2 2 , N 2 3 , D , W , K : H , S ), where Y , the “output” being produced, here, is the level of patient satisfaction; N is the vector of nursing inputs, measured in hours (e.g., N 1 = NAs, N 2 = RNs); D = physician inputs; W = all other hospital staff (a residual category); K measures the hospital’s capital stock—typically given by number of beds (Carey, 2000; Carey & Dor, 2008; Dor & Farley, 1996). All effects are conditioned on hospital characteristics ( H ), including for-profit/nonprofit status, case mix index (CMI), and teaching status and state-level characteristics ( S ) such as minimum nurse staffing laws and state fixed effects.…”