Table 1 Target of initial outbound calls.

Dependent variable = p(first outbound call is to nonfamily plan member)	Coef.	SE	z	p > \|z\|
(1) Ego and family all unembedded	0.572	0.067	8.50	<0.001^***
(2) Ego and family all embedded	−0.267	0.027	−9.97	<0.001^***
Earthquake intensity group (1 = severe)	0.075	0.034	2.18	0.029^*
(1) X Earthquake intensity group	0.227	0.104	2.19	0.029^*
(2) X Earthquake intensity group	−0.003	0.041	−0.06	0.950
ln(Degree centrality of ego)	0.166	0.023	7.14	<0.001^***
ln(Total call frequency of ego)	−0.074	0.018	−4.22	<0.001^***
ln(Total text frequency of ego)	0.009	0.007	1.40	0.162
ln(Internet usage frequency of ego)	−0.008	0.006	−1.35	0.176
ln(Phone retail price in Yuan)	−0.013	0.013	−1.03	0.304
ln(Total WeChat frequency)	<0.001	0.007	0.06	0.952
ln(Total frequency of other instant messaging)	0.002	0.006	0.34	0.734
Roaming dummy (1 = traveling outside of prefecture)	−0.343	0.026	−13.00	<0.001^***
Rural dummy (1 = rural)	−0.087	0.021	−4.20	<0.001^***
Damage dummy (1 = cell towers damaged in neighborhood)	0.267	0.068	3.90	<0.001^***
Prior intra-family outbound call share	−1.847	0.072	−25.65	<0.001^***
(Intercept)	0.920	0.091	10.14	<0.001^***

Notes: number of obs. = 26,394; model controls for random effects of families; the p values are obtained from a two-sided z-test.
Model 1 tests whether higher earthquake intensity affects the relationship between family embeddedness structure and likelihood that an ego’s first outbound call after the earthquake was to a nonfamily plan member (i.e., friend). We find that, if a family is completely unembedded, its members are more likely to call a friend first after the earthquake at higher earthquake intensities. However, fully embedded family structures only had a main effect; egos in fully embedded (unembedded) families were more (less) likely to first call a family member regardless of earthquake intensity.
^***p < 0.001, ^**p < 0.01, ^*p < 0.05.

Quick links

Search