Table 7.

Logistic regression estimates for effect of home/work stratum on commute disruptions.

dep. variable: $\mathbf{l}\mathbf{o}\mathbf{g}\left(\frac{\mathbf{P}\mathbf{r}(\mathit{d}isrup{t}_i\mathbf{=}\mathbf{1})}{\mathbf{1}-\mathbf{P}\mathbf{r}(\mathit{d}isrup{t}_i\mathbf{=}\mathbf{1})}\right)$
ind. variable	(i)	(ii)	(iii)	(iv)	(v)	(vi)	(vii)	(viii)
(a) all commuters
stratc(i,home)	0.187***		0.079***	0.092***	0.068***		0.031	0.046*
	(0.021)		(0.023)	(0.023)	(0.025)		(0.025)	(0.026)
stratc(i,work)		0.311***	0.282***	0.289***		0.227***	0.222***	0.232***
		(0.021)	(0.023)	(0.023)		(0.025)	(0.025)	(0.026)
stratc(i,home) × stratc(i,work)				−0.060***				−0.043**
				(0.018)				(0.022)
const	−0.817***	−1.209***	−1.332***	−1.365***
	(0.059)	(0.064)	(0.074)	(.075)
FE	none	none	none	none	muni.	muni.	muni.	muni.
average marginal effects
stratc(i,home)	4.5		1.9	2.2	1.6		0.7	1.1
stratc(i,work)		7.4	6.7	6.8		5.2	5.1	5.3
stratc(i,home) × stratc(i,work)				−1.4				−1.0
N	9069	9069	9069	9069	9069	9069	9069	9069
LL	−6126	−6054	−6048	−6042	−5953	−5915	−5914	−5912

dep. variable: $\mathbf{l}\mathbf{o}\mathbf{g}\left(\frac{\mathbf{P}\mathbf{r}({\mathit{d}isrupt}_{\mathit{i}}\mathbf{=}\mathbf{1})}{\mathbf{1}-\mathbf{P}\mathbf{r}({\mathit{d}isrupt}_{\mathit{i}}\mathbf{=}\mathbf{1})}\right)$
ind. variable	(i)	(ii)
(b) low-stratum commuters
stratc(i,work)	0.326***	0.271***
	(0.036)	(0.042)
FE	none	muni.
const	−1.255**
	(0.106)
average marginal effects
stratc(i,work)	7.7	6.1
N	3360	3360
LL	−2232	−2153