Skip to main content
. Author manuscript; available in PMC: 2015 Feb 1.
Published in final edited form as: Health Econ. 2014 Apr 17;24(2):224–237. doi: 10.1002/hec.3012

Table 6.

IV Estimates and Specification Tests for the Effect of Caregiving Hours

Outcome Obs. OLS or Probit IV-2SLS or IV-Probit IV specification tests

(1) (2) Coeff. (3) (S.E.) (4) Coeff. (5) (S.E.) (6) Overid (7) Exo.
Daughter-in-law sample
Prob(Pain affecting daily activities) 2527 0.18** (0.08) 0.33** (0.13) NS NS
Prob(Self-reported health fair to poor) 2527 0.08 (0.08) 0.25** (0.13) NS *
Prob(Any outpatient care use) 2524 0.03 (0.09) −0.10 (0.12) NS NS
 Log(Out-of-pocket costs) if any use 1548 0.10 (0.10) 0.24* (0.14) NS NS
Prob(Any regular prescription drug use) 2527 0.14* (0.08) 0.16 (0.12) NS NS
 Log(Out-of-pocket costs) if any use 720 0.13 (0.10) 0.15 (0.15) NS NS

Notes: NS denotes ‘not significant’ at the 10% level. Logarithm of weekly care hours was used. F-statistics for first-stage regressions were 26 in the four IV-Probit models and 23 and 12 in the two conditional IV-2SLS models. The same set of control variables were used as in Table 5. One observation was dropped because of missing value for caregiving hours.

Robust standard errors in parentheses.

***

p<0.01,

**

p<0.05,

*

p<0.1.