Validation of MAP-NET—A network analysis tool

Figures & data

Table

Section A: Data on FP
Module 1: Basic data on FP	Demographics and social background.
Module 2: Problem profile and motivation of FP	Problem severity in eight domains (work & finances, psychiatric and physical health, relations, alcohol, drugs, gambling and criminality). For each problem: change in past 3 years, and motivation to handle it.
Section B: The composition of the PSN
Module 3: Name-generating form of NM	Persons of importance in FP’s life, including those who play positive as well as negative roles.
Module 4: Network map	Structure and proximity of network relationships.
Module 5: Family and relations tree	Specific type of relation, e.g. partner, parent, grand-parent, sibling, child, other relative (same of partner), friend, acquaintance, co-member in support group, colleague, neighbour, various types of professional helper, etc.
Section C: Interactional and functional aspects of the PSN
Module 6: Interactional aspects of FP’s relations	For each NM: Frequency, form and duration of contacts, and geographic distance. If and how relation depends on FP’s substance misuse. Quality of relation, positive (rich) and negative (problematic).
Module 7: Functional aspects of FP’s relations	For each NM: Mutual provision of support (emotional, cognitive, practical & economic). Impact on various problems (negative or positive). Emotive functions. Sense of belonging. Links to new positive social contacts.
Section D: Data on the NMs and interrelations among and between NMs
Module 8: Basic data on the NMs—their resources and problems	For each NM: Social situation and resources. Own problems (alcohol, drugs, physical and emotional health, gambling and criminality). Active part in FP’s problems.
Module 9: Contact map—interrelations between NMs	For each NM: Contacts with other NMs in the PSN. For each relation: Frequency and quality.

Table 1. Problem severity distribution, and correlations (Spearman Rho) between problem severity in eight areas and relevant other scales, and internal consistency (α) of these scales

Problem area	Problem severity Median (min–max)	Emotional distress HAD	Cynical distrust CD	Internalised shame ISS	Social integration SS13	Social desirability MCSDS
nα	112–113 (a)	99–100 (b) 0.73	100–101 (b) 0.81	96–98 (b) 0.96	98–99 (b) 0.82	97–99 (b) 0.80
Work and finances	1 (0–3)	0.29**	0.17§	0.19§	−0.18§	−0.13
Physical health	1 (0–3)	0.22*	0.26**	0.07	−0.11	0.14
Emotional health	1 (0–3)	0.44***	0.28**	0.47***	−0.19§	−0.09
Social relations	1 (0–3)	0.35***	0.19§	0.46***	−0.28**	−0.16
Alcohol	3 (1–3)	0.20*	0.09	0.22*	−0.05	−0.00
Drugs	1 (0–3)	0.09	0.09	0.00	−0.01	−0.19§
Gambling	0 (0–3)	0.02	−0.04	−0.08	−0.02	−0.16
Criminality	0 (0–3)	0.07	0.17^§	0.08	−0.14	−0.21*

Notes: ^§p < 0.10; *p < 0.05; **p < 0.01; ***p < 0.001.

(a) One interview had internal missing data on gambling and criminality; (b) some of the 101 questionnaires on tests for convergent and predictive validity were incomplete.

Table 2. Comparisons between focal and collateral interviews (n = 51) on severity in eight problem areas, and for the relevant cases (a), on change of severity in the past 3 years, using rank order correlation (Spearman Rho) and symmetric correlation (Gamma, γ)

Problem area	Level of severity (0–4)					Change (−3 to +3)
	n	Rho	p(Rho)	γ	p(γ)	n (a)	Rho	p(Rho)	γ	p(γ)
Work and finances	51	0.61	< 0.001	0.69	< 0.001	38	0.57	< 0.001	0.58	< 0.001
Physical health	51	0.59	< 0.001	0.67	< 0.001	32	0.26	0.156	0.26	0.178
Emotional health	51	0.61	<0.001	0.68	< 0.001	36	0.42	0.012	0.43	0.011
Social relations	51	0.65	< 0.001	0.77	<0.001	25	0.45	0.024	0.44	0.035
Alcohol	51	0.52	< 0.001	0.78	0.002	43	0.38	0.011	0.38	0.011
Drugs (c)	49	0.84	< 0.001	0.95	< 0.001	15	0.57	0.026	0.57	0.012
Gambling (c)	46	0.67	< 0.001	0.95	0.131	7	b	–	b	–
Criminality (c)	48	0.79	< 0.001	0.98	0.009	8	b	–	b	–

Notes: (a) Number of relevant cases, i.e. those with the problem (severity > 0); (b) n(a) < 10; no correlations calculated; (c) Two collaterals could not provide information on drugs, three on criminality and five on gambling.

Table 3. Factor analysis (a, b) in relevant cases (c) with KMO (d), communalities, factor loadings and explained variance (R²); and internal consistency (α) of scales concerning motivation to handle problems in eight problem areas

Motivation to handle…	n (c)	KMO (d)	Communalities			Loadings			R^2	α
			TTA	RTA	STA	TTA	RTA	STA
Work and finances (a)	79	0.64	0.26	0.75	0.52	0.51	0.87	0.72	0.65	0.73
Physical health (a)	66	0.68	0.72	0.87	0.43	0.85	0.93	0.65	0.67	0.84
Emotional health (a)	80	0.72	0.54	0.80	0.77	0.73	0.89	0.88	0.72	0.87
Social relations (a)	59	0.70	0.60	0.93	0.66	0.77	0.96	0.81	0.73	0.88
Alcohol (b)	85	0.66	0.84	0.95	0.82	0.94	0.98	0.91	0.89	0.93
Drugs (a)	56	0.72	0.90	0.96	0.71	0.95	0.99	0.84	0.86	0.94
Gambling (e)	11									0.90
Criminality (e)	26									0.96

Notes: Based on three in each scale: Thought of taking action (TTA), Ready to take action (RTA), Started to take action (STA).

(a) The extraction method was PAF or (b) PCA (when PAF produced no solution due to high communality for one item); (c) Relevant cases were those with the problem (severity > 0); (d) KMO = Kaiser–Meyer–Olkin Measure of Sampling Adequacy; (e) Factor analysis not performed due to few cases with the problem (<50).

Table 4. Correlations (Rho) between phases of change (RTCQ) and alcohol problem level (n = 100), change (n = 82), as well as motivation (n = 77–79)

	Pre-contemplation	Contemplation	Action
Alcohol problem severity	−0.51***	+0.54***	+0.25*
Change in last 3 years	+0.24*	−0.26*	−0.13
Thought of taking action	−0.30**	+0.12	+0.46***
Ready to take action	−0.30**	+0.12	+0.45***
Started to take action	−0.23*	+0.07	+0.42***
Scale: Motivation to handle alcohol problems	−0.24*	+0.08	+0.44***

Notes: *p < 0.05; **p < 0.01; ***p < 0.001.

Table 5. Factor analysis^a on reduced model of six items on quality of relation, resulting in two factors: rich vs. problematic features (n = 1,252 after pairwise deletion)

Quality of the relation, six items:	Communalities	Loadings
		Rich	Problematic
I feel that this person really cares about me	0.59	0.87	−0.38
It feels like this person wants to connect with me	0.54	0.81	−0.32
I really care about this person	0.51	0.76	−0.37
I can probably act as threatening against this person	0.24	−0.19	0.57
I feel that the person can be threatening	0.31	−0.33	0.80
I feel unpleasant in this relationship	0.34	−0.52	0.56

Notes: ^a PAF with Promax rotation.

Table 6. Rank correlations (Rho) between quality of the relation—positive (rich), negative (problematic)—and proximity, order in name generation, and with scales presented below, i.e. mutual support and emotive function

	n	Rich relation	Problematic relation
Order in name generation (reversed) (a)	1,255	0.31***	−0.06*
Proximity of relation (b)	1,251	0.59***	−0.31***
Support from NM to FP (c)	1,235	0.51***	−0.24***
Support from FP to NM (c)	1,226	0.57***	−0.28***
Emotive function of the relation (d)	1,248	0.64***	−0.58***

Notes: (a) Presented in Module 3; (b) in Module 4; (c) in Module 6; (d) in Module 7.

*p < 0.05; ***p < 0.001.

Table 7. Agreement (systematic correlation, γ) between focal and collateral interviews concerning exchange of four types of support in both directions

	n	γ	p(γ)
Support from network person to focal person
Emotional	252	0.64	< 0.001
Cognitive	243	0.59	< 0.001
Practical	242	0.53	< 0.001
Economic	234	0.73	< 0.001
Support from focal person to network person
Emotional	239	0.40	< 0.001
Cognitive	241	0.39	< 0.001
Practical	238	0.57	< 0.001
Economic	220	0.77	< 0.001

Table 8. Absolute agreement and systematic correlation (γ) between focal and collateral interviews concerning the network persons’ impact on FPs’ substance use problems, when relevant

	n	Absolute agreement	γ	p(γ)
Negative influence
Alcohol	179	0.83	0.89	< 0.001
Drugs	44	0.84	0.90	0.026
Positive influence
Alcohol	187	0.40	0.40	< 0.001
Drugs	47	0.32	0.33	0.068

Table 9. Agreement (systematic correlation, γ) between focal and collateral interviews concerning FPs’ emotional relations to the NMs

	n	γ	p(γ)
Conflict vs. harmony	264	0.59	< 0.001
Disgust vs. respect	261	0.49	< 0.001
Opposing vs. shared values	248	0.49	< 0.001
Fear/threat vs. safety	263	0.64	< 0.001
Lowers vs. strengthens self-sense	255	0.66	< 0.001
Distrust vs. trust	248	0.55	< 0.001

Table 10. Systematic agreement (к), systematic correlation (γ), and Pearson correlation (R) between focal and collateral interviews on socio-demographic information concerning NM

	n	к	p(к)	R	p(R)	γ	p(γ)
Age (a)	210			0.98	< 0.001
Education (b)	162	0.68	< 0.001			0.92	< 0.001
Marital status (c)	202	0.84	< 0.001
Work situation (d)	185	0.82	< 0.001			0.89	< 0.001
Housing (e)	174	0.72	< 0.001			0.91	< 0.001

Notes: (a) continuous variable: No. of years; (b) ordinal variable: Had not completed elementary school/completed elementary school/completed high school/academic education (≥2 years); (c) Dichotomous variable: married or cohabiting vs. not (reversed); (d) Ordinal variable in relation to labour market situation: out of job/long-term sick leave or retired/In vocational training or education/Employed or self-employed; (e) Ordinal variable based on housing tenure security: no home/low threshold/lodging/second hand contract/own rental contract or ownership of apartment.

Table 11. Absolute agreement and systematic agreement (к) between focal and collateral interviews on NMs’ problems

	n	Absolute agreement	к	p(к)
Alcohol problem	175	0.86	0.64	< 0.001
Drug problem	175	0.93	0.62	< 0.001
Emotional problem	150	0.82	0.54	< 0.001
Physical problem	171	0.85	0.59	< 0.001
Criminality	165	0.95	0.61	< 0.001