Measurement of diabetes-related emotional distress using the Problem Areas in Diabetes scale: psychometric evaluations show that the short form is better than the full form

Table 3 CFA: Goodness-of-fit statistics for the K-PAID and K-PAID-5 models and standardized loadings

Underlying factor model	X²(p)	df	CMIN/DF	GFI	RMR	RMSEA	CFI	NFI	Standardized loadings
Full form (K-PAID)
One-factor model^a	634.567 (p < 0.001)	169	3.755	0.864	0.074	0.079	0.903	0.873	0.29-0.83
Two-factor model^b	615.042 (p < 0.001)	168	3.661	0.907	0.074	0.078	0.907	0.877	0.31-0.84
Two-factor model^c	580.633 (p < 0.001)	168	3.456	0.876	0.072	0.075	0.914	0.884	0.31-0.84
Three-factor model^d	567.558 (p < 0.001)	166	3.419	0.878	0.072	0.074	0.917	0.887	0.29-0.83
Four-factor model^e	561.753 (p < 0.001)	164	3.425	0.880	0.072	0.074	0.918	0.888	0.37-0.88
Short form (K-PAID-5)
One-factor model^f	11.536 (p = 0.021)	4	2.884	0.990	0.039	0.066	0.993	0.989	0.65-0.84

df, Degrees of freedom; CMIN/DF, Ratio of chi-square value to the degrees of freedom; GFI, Goodness-of-fit index; RMR, Root-mean-square residual; RMSEA, Root-mean-square error of approximation; CFI, Comparative-fit index; NFI, Normed fit index.
^aAfter modification with the covariance of error terms between items 1 and 2 [modification index (MI) = 52.59, ∆X²(1) = 55.989, p < 0.01], based on the factor model of Welch et al. [8].
^bAfter modification with the covariance of error terms between items 1 and 2 [MI = 51.79, ∆X²(1) = 54.691, p < 0.01], based on the factor model of Miller et al. [13].
^cAfter modification with the covariance of error terms between items 1 and 2 [MI = 55.56, ∆X²(1) = 55.718, p < 0.01], based on the factor model of Huis In ‘t Veld et al. [15].
^dAfter modification with the covariance of error terms between items 1 and 2 [MI = 53.24, ∆X²(1) = 56.501, p < 0.01], based on the factor model of Papathanasiou et al. [12].
^eBased on the factor model of Snoek et al. [9].
^fAfter modification with the covariance of error terms between items 3 and 6 [MI = 31.50, ∆X²(1) = 42.026, p < 0.01], based on the factor model by McGuire et al. [21].

ISSN: 1477-7525