Who wants to be the next James Bond or Anna Chapman? exploring the correlates of a willingness to enter the Centro Nacional de Inteligencia (CNI) among Spanish university students

Figures & data

Table 1. Descriptive statistics for sociodemographic variables based on willingness to work for the CNI (n = 2,616).

		No (n = 1361)		Yes (n = 1255)		Total
		n	%	n	%	n	%
Sex
	Male	557	49.5	569	50.5	1126	100
	Female	769	54.1	674	45.9	1470	100
Academic discipline
	Applied Sciences	313	55.5	251	44.5	564	100
	Formal Sciences	8	44.4	10	55.6	18	100
	Humanities	167	72	65	28	232	100
	Natural Sciences	59	59	41	41	100	100
	Social Sciences	814	47.8	888	52.2	1702	100
Relatives in security professions
	Yes	144	45.4	173	54.6	317	100
	No	1155	53.3	1012	46.7	2167	100

Table 2. Factor loadings and communalities based on Principal Component Analysis (PCA) with direct oblimin Kaiser normalisation rotation for 18 items (N = 2616), ‘The CNI …’ .

	Components
	1	2	3	4	Communalities
Compiles private and public confidential data	.776				.641
Monitors activity of social media	.769				.625
Conducts telephone tapping without court orders	.691				.626
Violates individual rights and liberties	.559				.575
Enters private homes without court orders	.535				.445
Is effective at fighting against organised crime		.719			.625
Is effective at fighting against terrorism		.769			.663
Is effective at protecting communication networks from cyber-attacks (hacking)		.606			.503
Knows the risks and threats which the States confronts		.724			.618
Works for the interests of the community		.672			.560
Works to combat threats against the State		.764			.678
Is effective at advising the government on national policy			.918		.837
Is effective at advising the government on international policy			.897		.820
Is effective at defending the economic and industrial interests of Spain			.793		.624
Has its own objectives not set by the Government				.691	.461
Hides behind secrecy so as not to provide explanations of its work				.608	.570
Jumps procedures when security is at risk				.600	.545
Completes missions attentive to the interests of the political party in government and not the State				.546	.465

Rotation converged in 20 iterations. Factor loadings <0.4 are suppressed. Percentage of variance explained by each factor: Factor 1 = 22.66 per cent, Factor 2 = 20.36 per cent, Factor 3 = 11.27 per cent, Factor 4 = 6.17 per cent. Total percentage of variance explained = 60.45 per cent, KMO =.823.

Table 3. Binary logistic regression models predicting willingness to work for the CNI.

Variables	Model 1 (Education)	Model 2 (Political)	Model 3 (Knowledge)	Model 4 (Confidence)	Model 5 (Combined)
Education (ref. = Social Sciences)
Applied Sciences	−0.06 (0.95)				0.05 (1.05)
Formal Sciences	−0.47 (0.63)				−0.18 (0.84)
Humanities	−1.06*** (0.35)				−0.85*** (0.43)
Natural Sciences	−0.63*** (0.53)				−0.69*** (0.49)
Political opinions
Ideology		0.15*** (1.16)			0.09*** (1.09)
Democracy satisfaction		0.18** (1.19)			0.03 (1.03)
Trust in institutions		0.31** (1.36)			0.07 (1.07)
Knowledge
Missions			0.09*** (1.10)		0.08*** (1.08)
Powers			0.01 (1.01)		−0.02 (0.98)
Reputation			0.37*** (1.44)		0.21*** (1.23)
Image			0.38*** (1.47)		0.28*** (1.32)
Confidence in CNI
Procedural unlawfulness				−0.12** (0.89)	−0.07 (0.94)
Police effectiveness				0.59*** (1.81)	0.44*** (1.55)
Intelligence effectiveness				0.15** (1.16)	0.04 (1.04)
Non-neutrality				0.00 (1.00)	−0.00 (0.99)
Control variables
Gender (ref. = male)	−0.17 (0.85)	−0.13 (0.87)	−0.22* (0.80)	−0.29** (0.75)	−0.20* (0.82)
College year	−0.15** (0.86)	−0.10* (0.90)	−0.12** (0.89)	−0.13** (0.88)	−0.17** (0.84)
Relatives (ref. = no)	0.35** (1.41)	0.23 (1.26)	0.27* (1.31)	0.21 (1.24)	0.19 (1.21)
Constant	0.44*** (1.55)	−1.44*** (0.24)	−1.85 (0.16)	0.31** (1.36)	−1.34*** (0.26)
Hosmer and Lemeshow test^a	13.535	17.742*	6.008	10.539	10.429
Nagelkerke R^2	0.043	0.077	0.101	0.125	0.193
n	2464	2376	2420	2319	2234

N = 2616 students from Spanish Universities. The table displays the standardised coefficient and odds ratio in brackets. *p < .05; **p < .005; ***p < .001. ^aHosmer – Lemeshow statistics indicate a good fit of the model if the significance value is higher than .05.

Figure 1. Percentage of students per university undergraduate degree who would be willing to work for the CNI.

Table 4. Results of the pairwise comparisons of proportions (N = 2616).

	Willing to work for the CNI
	No	Yes
Architecture	53.3 per centa,b,e,j	46.7 per centa,b,e,j
Business	43.2 per cente,g,h,i	56.8 per cente,g,h,i
Chemistry	59.0 per centa,e,i,j	41.0 per centa,e,i,j
Communication	57.1 per centa,d,e,i	42.9 per centa,d,e,i
Computer Science	37.1 per centb,i	62.9 per centb,i
Criminology	23.8 per centb	76.2 per centb
Education	66.9 per centa,j,k	33.1 per centa,j,k
Engineering	57.3 per centa,c,e	42.7 per centa,c,e
History	72.9 per centa,e,j	27.1 per centa,e,j
International Relations	18.6 per centb,g	81.4 per centb,g
Law	47.3 per cente,f,i	52.7 per cente,f,i
Mathematics	44.4 per centa,b,e,j	55.6 per centa,b,e,j
Medicine	70.6 per centa,j,k	29.4 per centa,j,k
Philology	78.8 per centj	21.2 per centj
Philosophy	47.1 per centa,b,e,j	52.9 per centa,b,e,j
Political Science	62.5 per centa,e,j	37.5 per centa,e,j
Psychology	46.7 per centb,c,d,e,k	53.3 per centb,c,d,e,k
Translation	67.6 per centa,e,j	32.4 per centa,e,j
Tourism	57.4 per centa,f,h,i,j,k	42.6 per centa,f,h,i,j,k

Values in the same column that do not share the same subscript are significantly different at p < 0.05 in the bilateral test of equality for proportions.

Table

	Components
	1	2	3	Communalities
To what extent do you trust in each of the following political institutions?
The political parties	.850			.729
The Government	.829			.748
The Parliament	.685			.680
The Monarchy		.873		.780
The Judicial power		.719		.635
Ombudsman			.948	.946

Rotation converged in 4 iterations. Factor loadings <0.4 are suppressed. Percentage of variance explained by each factor: Factor 1 = 46.87 per cent, Factor 2 = 15.50 per cent, Factor 3 = 12.93 per cent, Total percentage of variance explained = 75.29 per cent, KMO =.803.