The impact of group psychological abuse on distress: the mediating role of social functioning and resilience

Figures & data

Figure 1. Proposed model concerning the relationship between group psychological abuse and distress: social functioning and resilience as mediators

Table 1. Descriptive data of the samples of victims and non-victims

	Total sample	Victims	Non-victims	Comparisons
				Statistics	Effect size
Sex
Men	31.5%	31.9%	30.8%	χ^2 = 6.85	V = .09
Women	67.1%	65.9%	69.2%
Other	1.4%	2.2%	0%
Age
Mean (SD)	49.5 (15.8)	47.1 (14.5)	53.7 (17)	t = 5.64	d = .43
Educational level
Primary education	7%	7.2%	6.5%	χ^2 = 2.80	V = .06
Secondary education	22.6%	20.7%	25.9%
University	70.4%	72%	67.7%
Job status
Student	5.6%	6.8%	3.4%	χ^2 = 71.15	V = .30
Unemployed	12.8%	11.7%	14.6%
Full time work	41.1%	48.1%	29.3%
Part time work	16.1%	18.1%	12.6%
Pensioner	24.5%	15.3%	40.1%
Age joining the group
Mean (SD)	19 (16.4)	13.9 (13.8)	27.8 (16.8)	t = 12.03	d = .93
Years inside the group
Mean (SD)	14.8 (12.5)	17.3 (12.5)	10.6 (11.3)	t = −7.79	d = −.55
Years outside the group
Mean (SD)	15.7 (13.5)	15.9 (12.9)	15.4 (14.5)	t = −.49	d = −.03
Group nature
Religious	61.3%	77%	34.9%	χ^2 = 138.18	V = .42
Non-religious	38.7%	33%	65.1%
Method of leaving
Personal reflection	56.9%	63.9%	45.1%	χ^2 = 101.14	V = .36
Counselled	7.2%	9.8%	2.7%
Expelled/Dissolution	21.4%	20.8%	22.4%
Other (e.g. life change)	14.5%	5.4%	29.8%
Group psychological abuse
Mean (SD)	62.7 (45.9)	95 (21.9)	8.2 (11.1)	t = −73.85	d = −4.65

Total sample n = 794. Victims: n = 499. Non-victims: n = 295. χ² = Pearson chi-square test. V = Cramer’s V. t = Student’s t test. d = Cohen’s d. All chi-square tests and t tests were significant at p < .05, except in ‘educational level’ (p = .24) and ‘years outside the group’ (p = .62). Group psychological abuse = measured through the Psychological Abuse Experienced in Groups Scale.

Table 2. Descriptive data and correlations between measures

	Victims			Non-Victims		Comparisons
	M (SD)			M (SD)		t	d
Social functioning	36.68 (8.57)			40.04 (8.55)		−5.34***	−.39
Resilience	25.72 (7.98)			27.69 (7.94)		−3.36**	−.25
Psychosocial difficulties	6.57 (28.48)			9.08 (16.23)		42.53***	2.74
Psychopathological symptoms	1.54 (.95)			.38 (.59)		21.21***	1.39
	1			2	3	4	5
1. Group psychological abuse	-			-	-	-	-
2. Social functioning	−.19***			-	-	-	-
3. Resilience	−.09**			.57***	-	-	-
4. Psychosocial difficulties	.84***			−.32***	−.25***	-	-
5. Psychopathological symptoms	.59***			−.42***	−.37***	.79***	-

Total sample: n = 794. Victims: n = 499. Non-victims: n = 295.

*p < .05; **p < .01; ***p < .001

Table 3. Model coefficients of serial mediation analyses

		Consequent
		M1	M2		Y1	Y2
Antecedent		Coefficient (SE)	Coefficient (SE)		Coefficient (SE)	Coefficient (SE)
X	a	−.027 (.007) ***	−.001 (.005)	c’	.728 (.018) ***	.012 (.001) ***
M1	d	-	.523 (.027) ***	b1	−.429 (.111) ***	−.023 (.003) ***
M2		-	-	b2	−.587 (.121) ***	−.025 (.004) ***
C1	f1	2.07 (.653) **	.914 (.518)	g1	−3.85 (1.70) *	.043 (.059)
C2	f2	−.069 (3.32)	4.25 (1.70) *	g2	2.58 (5.01)	−.115 (.193)
C3	f3	.059 (.021) ***	−.018 (.015)	g3	.021 (.049)	−.001 (.001)
constant	i1	37.85 (.77) ***	6.67 (1.16) ***	i2	38.40 (4.31) ***	1.93 (.161) ***
		R^2 = .06	R^2 = .33		R^2 = .74	R^2 = .48
		F4, 789 = 13.07***	F5, 788 = 79.89***		F6, 787 = 657.43***	F6, 787 = 208.90***

n = 794. Coefficient = unstandardized regression coefficients. X = Group psychological abuse. M₁ = Social functioning. M₂ = Resilience. C₁ = Sex: Men vs Women and Other. C₂ = Sex: Other vs Women and Men. C₃ = Age joining the group. Y₁ = Psychosocial difficulties. Y₂ = Psychopathological symptoms.

*p < .05; **p < .01; ***p < .001

Figure 2. Serial mediation models: social functioning and resilience mediating the association between group psychological abuse and distress

Values shown are standardized coefficients. Total effect of Group Psychological Abuse is shown in parenthesis. Covariates were sex and age joining the group but are not represented here. Panel A: Serial mediation model predicting Psychosocial Difficulties. Panel B: Serial mediation model predicting Psychopathological Symptoms. *p < .05; **p < .01; ***p < .001

Data availability statement

The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due that it contains information that could compromise the privacy of research participants.