Social Cohesion and Collective Violence: Latent Variable Approach to Explaining Riots in East Jerusalem

Figures & data

Figure 1. Residential patterns in Jerusalem.

Figure 2. Riots in Jerusalem, 2013–2015.

Table 1. Summary statistics of the main variables.

Variable	Mean	Std. dev.	Min.	Max.
Riots	0.009	0.101	0	3
Cohesion	0	0.787	−1.482	1.648
Economic disadvantage	30.633	5.235	20	42
Ethnic friction	0.4	0.49	0	1
Population	15765.5	9479.791	2260	38810
Major events	0.601	0.664	0	2

Table 2. Summary statistics of proxies for cohesion.

Variable	Mean	Std. dev.	Min.	Max.
Crime	62.964	47.022	2	180
Trust	2.051	0.600	0.688	3.278
Hamula	9.307	9.809	0	36
Negative migration	0.047	0.047	0.009	0.3284
Local organizations	5.470	2.637	3	11

Table 3. Factor analysis on social cohesion indicators.

Indicator	Factor loading
Crime	−0.6610
Trust	0.3921
Hamula	−0.2048
Negative migration	−0.4423
Local organizations	0.5768
Cronbach’s alpha	0.51
Eigenvalue	1.16090

Figure 3. Spatial autoregressive model (police data). In all models, thick lines denote significance levels of p < .05 and the thin extensions denote significance levels of p < .1.

Table 4. Random effects with clustered errors model (police data).

	(1)	(2)
	Model A	Model B
Variables	Riots	Riots
Cohesion	−0.201***	−0.230**
	(0.0520)	(0.109)
Economic motivation	0.0491	0.0692***
	(0.0315)	(0.0187)
Ethnic friction	−0.0193	−0.0264
	(0.0254)	(0.0207)
Riots lagged		0.0678*** (0.0224)
Population		−0.0105
		(0.0348)
Major event		0.0238
		(0.0211)
Constant	0.116***	0.126**
	(0.0296)	(0.0590)
Observations	14,235	14,235
Number of neighborhoods	13	13

Note. Robust standard errors in parentheses.

***p < .01, **p < .05.

Figure 4. Interaction terms on all datasets.

Figure 5. Riots in Jerusalem, 2013–2015 (United Nations data).

Figure 6. SAR model—United Nations stone throwing.

Figure 7. Riots in Jerusalem, 2013–2015 (crowd sourced).

Figure 8. SAR model—Crowd-sourced riots.

Figure 9. SAR model—Crowd-sourced stone throwing.