Table 2.

Multilevel Ordered Logistic Models of Individual and Manager Characteristics on Employees’ Job Insecurity, in Logged Odds

VARIABLES	Model 1	Model 2	Model 3
Merger knowledge
Employee interviewed post-merger announcement	0.932***	1.695***	−0.130
	(0.172)	(0.316)	(0.519)
Employee characteristics
Female	−0.262	−0.296	−0.250
	(0.179)	(0.178)	(0.178)
Age	0.223*	0.221*	0.231*
	(0.091)	(0.091)	(0.091)
Age squared	−0.002	−0.002	−0.002*
	(0.001)	(0.001)	(0.001)
Married/Partnered	0.003	0.011	0.013
	(0.210)	(0.210)	(0.210)
Parent	−0.013	−0.015	−0.023
	(0.176)	(0.176)	(0.176)
Has adult care responsibility	0.163	0.172	0.169
	(0.187)	(0.187)	(0.187)
Race/ethnicity
Asian	−0.056	−0.107	−0.066
	(0.225)	(0.224)	(0.224)
Hispanic, Black or African American	0.649*	0.667*	0.607*
	(0.272)	(0.272)	(0.271)
College graduate	−0.496*	−0.492*	−0.514*
	(0.228)	(0.228)	(0.228)
Logged income	−0.889	−0.969*	−0.944*
	(0.479)	(0.479)	(0.477)
Tenure (years)	−0.012	−0.014	−0.012
	(0.011)	(0.011)	(0.011)
Core function	−0.140	−0.159	−0.176
	(0.190)	(0.185)	(0.187)
Decision authority	−0.410***	−0.386***	−0.413***
	(0.116)	(0.116)	(0.115)
Job demands	0.163	0.191	0.146
	(0.119)	(0.118)	(0.117)
Manager characteristics
Female	0.283	0.200	0.275
	(0.183)	(0.183)	(0.180)
Race/ethnicity
Asian	−0.540*	−0.508*	−0.509*
	(0.232)	(0.229)	(0.230)
Hispanic, Black or African American	−0.030	0.024	−0.034
	(0.312)	(0.308)	(0.308)
Manager’s tenure	−0.011	0.011	−0.012
	(0.008)	(0.011)	(0.008)
Manager’s job insecurity	0.332**	0.355**	0.065
	(0.118)	(0.118)	(0.171)
Interactions
Employee interviewed post-merger announcement* Manager’s tenure		−0.045**
		(0.016)
Employee interviewed post-merger announcement* Manager’s job insecurity			0.494*
			(0.229)
Thresholds (see Note 1)
Cutpoint 1	−7.139	−7.512	−8.201
	(5.301)	(5.288)	(5.288)
Cutpoint 2	−3.878	−4.232	−4.948
	(5.297)	(5.279)	(5.280)
Cutpoint 3	−1.695	−2.041	−2.750
	(5.302)	(5.281)	(5.280)
Level 2 Pseudo R Squared (see Note 2)	0.967	1.000	1.000
Observations	666	666	666

Notes:

¹These are the thresholds used to differentiate the adjacent levels of the response variable, job insecurity. For example, “cutpoint 1” is the estimated cutpoint used to differentiate those who report job insecurity to be 1 “not at all likely” to be laid off from those who report job insecurity to be 2 “not too likely” to be laid off. “Cutpoint 2” differentiates those who report job insecurity to be 2 “not too likely” to be laid off from those who report 3 “fairly likely” to be laid off, etc. Ordered logistic models assume that the categorical outcomes we observe in the data come from an underlying latent variable that is continuous. When values of the independent variables are evaluated at zero, the cutpoints are the estimated thresholds on the latent variable used to make the four groups that we observe with different levels of job insecurity (1, 2, 3, and 4). For example, the estimate of “cutpoint 1” in Model 1 indicates that respondents who had a value of −7.139 or less on the underlying latent variable that gives rise to the job insecurity variable would constitute the group who reported “not at all likely to lose jobs” (i.e., valued 1 for the outcome).

²Pseudo R squared is defined as the ratio of two numbers. The numerator is the level-2 random effect difference between the current model and an unconditional model, and the denominator is the level-2 random effects of the unconditional model. The unconditional model is a null model including no covariate. See Raudenbush and Bryk (2002).

³Standard errors in parentheses.

^***p<0.001,

^**p<0.01,

^*p<0.05