Skip to main content
. Author manuscript; available in PMC: 2014 Feb 26.
Published in final edited form as: Ann Appl Stat. 2012 Dec;7(3):1362–1385. doi: 10.1214/12-AOAS624

Table 3.

Percentage increase in mortality associated with a unit increase in the state unemployment rate, using different model specifications.

Model B1 L1 D1 HP16.25 HP1100
Total -0.52 *** -0.31 *** -0.16 * -0.24 *** -0.33 ***
** *** * *** ***
** ** * ***
20-44 -1.15 *** 0.14 -0.54 * -0.73 *** -0.47 **
** * ** *
* + *
45-64 -0.72 *** -0.01 -0.13 -0.14 -0.22 **
** *
* *
65+ -0.43 *** -0.16 *** -0.03 -0.16 * -0.25 ***
** + + ***
* *
Cardiovascular disease -0.38 *** -0.20 ** -0.06 -0.14 -0.24 ***
+ **
+ *
Ischemic heart disease -0.33 + -0.35 ** -0.14 -0.28 + -0.58 ***
+ + **
*
Cancer -0.20 * 0.21 *** 0.13 0.05 0.04
*
*
Respiratory disease -1.04 *** -0.39 ** -0.37 -0.69 ** -0.71 ***
*** ** ** ***
** * * **
Other infectious disease -0.35 0.37 -1.14 * -1.72 *** -0.89 *
* ** +
* **
Traffic injury -3.76 *** -3.48 *** -1.48 *** -1.44 *** -2.11 ***
*** *** *** *** ***
*** *** *** *** ***
Suicide 0.25 1.06 *** 0.94 * 0.80 * 0.77 **
* + + +
* * + +
Homicide -1.71 *** -1.20 * -1.02 -0.74 -0.41
* +
* +

Columns represent models, as described in equation (1) and Table 1. Rows represent mortality categories. Table entries are estimates of 100α, using OLS with states weighted by the square root of the state population. Statistical significance is shown using standard OLS errors (black symbols, top row), error estimates clustered by state (gray symbols, middle row) and error estimates of Cameron et al. (2011, Section 2.2) with two-way clustering by state and year (gray symbols, bottom row; red in electronic version).

***

P < 0.001,

**

P < 0.01,

*

P < 0.05,

+

P < 0.1.