Do Norwegian Sami and non-indigenous individuals understand questions about mental health similarly? A SAMINOR 2 study

Figures & data

Table 1. Invariance testing of the responses to the HSCL-10 symptoms questionnaire.

	Norw	Norw-Kven M diff (t)	Sami Core	Sami Affilitation	Sami Background
Model fit HSCL-10 All 10 items
χ^2(df)	950.4^a (35)	334.2^a (35)	213.4^a (35)	313.3^a (35)	306.8^a (35)
RMSEA/SRMR/TLI	.069/.043/.891	.067/.040/.912	.062/.039/.923	.074/.040/.908	.085/.045/.880
Model fit HSCL-6 Depr items only
χ^2(df)	140.3^a (9)	41.8^a (9)	49.6^a (9)	47.1^a (9)	36.5^a (9)
RMSEA/SRMR/TLI	.053/.022/.962	.045/.017/.978	.059/.027/.958	.056/.022/.966	.055/.021/.970
Invariance testing of HSCL-10
M0: all equal ^1		Unadj fit: χ^2 = 9835.6^a, df = 318, RMSEA = .118, SRMR = .284, TLI = .730
		Adj fit: χ^2 = 11,583.2^a, df = 478, RMSEA = .106, SRMR = .200, TLI = .700
Latent mean, unadj ^2	0 (reference)	.057 (4.31^a)	.055 (3.69^a)	.098 (6.32^a)	.068 (4.02^a)
Latent mean, adj ^3		.028 (2.16^c)	.025 (1.66)	.074 (4.93^a)	.042 (2.56^b)
M1: free residuals ^4		Unadj fit: χ^2 = 2375.3^a, df = 254, RMSEA = .063, SRMR = .062, TLI = .925
		Adj fit: χ^2 = 2921.5^a, df = 414, RMSEA = .054, SRMR = .047, TLI = .922
Latent mean, unadj ^2	0 (reference)	.029 (4.16^a)	.034 (4.28^a)	.054 (6.39^a)	.034 (3.77^a)
Latent mean, adj ^3		.013 (1.86)	.018 (2.20^c)	.041 (5.00^a)	.020 (2.25^c)
M2: free loadings ^5		Unadj fit: χ^2 = 2338.7^a, df = 242, RMSEA = .064, SRMR = .062, TLI = .922
		Adj fit: χ^2 = 2895.2^a, df = 402, RMSEA = .055, SRMR = .047, TLI = .920
Latent mean ^2	0 (reference)	.031 (4.44^a)	.037 (4.57^a)	.056 (6.59^a)	.036 (3.99^a)
Latent mean, adj ^3		.016 (2.23^c)	.021 (2.59^b)	.044 (5.29^a)	.022 (2.54^c)
M3: free intercepts ^6		Unadj fit: χ^2 = 2089.0^a, df = 206, RMSEA = .065, SRMR = .061, TLI = .918
		Adj fit: χ^2 = 2715.5^a, df = 366, RMSEA = .056, SRMR = .046, TLI = .917
Latent mean ^2	0 (reference)	.028 (3.23^a)	.035 (3.59^a)	.053 (5.23^a)	.027 (2.55^c)
Latent mean, adj ^3		.013 (1.51)	.019 (1.93)	.039 (3.97^a)	.017 (1.63)
M4: free variances ^7		Unadj fit: χ^2 = 2086.8^a, df = 202, RMSEA = .066, SRMR = .046, TLI = .916
		Adj fit: χ^2 = 2710.5^a, df = 362, RMSEA = .056, SRMR = .037, TLI = .916
Latent mean ^2	0 (reference)	.028 (3.22^a)	.035 (3.58^a)	.054 (5.21^a)	.028 (2.54^c)
Latent mean, adj ^3		.014 (1.54)	.019 (1.94)	.040 (3.96^a)	.018 (1.65)

^a p < .001, ^b p < .01, ^c p < .05. ¹ All loadings specified as 1, all intercepts and all residuals constrained equal across all items and all groups. ² The zero latent mean score (M = 0) equals a HSCL sum score of 1.321 in the Norwegian sample. ³ HSCL item scores in addition adjusted for the covariates age, gender, income and education. ⁴ Nine error variances in addition set free (except item 8), loadings and intercepts also set free across items but constrained equal across groups. ⁵ Same as model M1, in addition 3 loadings set free across groups (item 4, 6 and 9). ⁶ Same as model M2, in addition 9 intercepts set free across groups (except for item 7 to achieve convergence). ⁷ Same as model M3, but factor variances in addition set free.

Table 2. Estimated HSCL item scores using the adjusted measurement model parameters.

	Estimated raw score (range 1–4)
HSCL item	Norw	Norw-Kven	Sami core	Sami affiliation	Sami backgr	diff
Sudden fears	1.40	1.44	1.46	1.46	1.44	.06
Frightened or anxious	1.65	1.71	1.75	1.74	1.69	.10
Faintness or dizziness	1.70	1.77	1.72	1.78	1.76	.08
Tense or upset	1.97	2.02	2.04	2.06	2.02	.09
Blame yourself	1.99	2.05	2.03	2.10	2.02	.11
Insomnia/sleeplessness	2.06	2.20	1.86	2.14	2.18	.34
Dejected or melancholic	1.94	2.02	2.02	2.05	2.00	.11
Useless, or little value	1.85	1.90	1.90	1.94	1.90	.09
Everything is a struggle	2.06	2.15	2.14	2.14	2.14	.09
Hopeless about future	1.88	1.95	1.99	1.96	1.96	.11
M	1.850	1.921	1.891	1.937	1.911	.087

Diff = Largest difference.