Skip to main content

Table 5 Known-groups validity of the K-PAID and K-PAID-5 by gender

From: 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

Underlying construct Scale/subscales and their names (number of items) Males (n = 226) Females (n = 214) t(p) Cohen’s effect size (d)
Mean±SD Mean±SD
Full form (K-PAID)      
One-factor constructa Single scale (20 items) 33.21±22.23 36.99±21.96 -1.84 (p = 0.07)  
Two-factor constructb Subscale 1: Lack of confidence (7 items) 8.65±6.21 8.88±6.44 -0.38 (p = 0.71)  
  Subscale 2: Negative emotional consequences (13 items) 24.56±16.01 28.11±16.57 -2.29 (p = 0.02) 0.22
Two-factor constructc Subscale 1: Diabetes distress (15 items) 28.05±17.40 30.70±17.42 -1.60 (p = 0.11)  
  Subscale 2: Support-related issues (5 items) 5.16±4.74 6.29±5.43 -2.32 (p = 0.02)g 0.22
Three-factor constructd Subscale 1: Diabetes-related emotional problems (15 items) 24.65±16.29 27.64±16.84 -1.89 (p = 0.06)  
  Subscale 2: Food-related problems (3 items) 6.60±4.23 6.94±4.16 -0.84 (p = 0.40)  
  Subscale 3: Social-support-related problems (2 items) 1.95±2.28 2.41±2.69 -1.93 (p = 0.05)g  
Four-factor constructe Subscale 1: Diabetes-related emotional problems (12 items) 20.87±14.56 24.40±15.57 -2.46 (p = 0.01) 0.23
  Subscale 2: Treatment-related problems (3 items) 3.78±2.94 3.24±2.78 2.00 (p = 0.05)  
  Subscale 3: Food-related problems (3 items) 6.60±4.23 6.94±4.16 -0.84 (p = 0.40)  
  Subscale 4: Social-support-related problems (2 items) 1.95±2.29 2.41±2.69 -1.93 (p = 0.05)g  
Short form (K-PAID-5)      
One-factor constructf Single scale (5 items) 7.97±5.35 9.71±5.64 -3.30 (p < 0.01) 0.31
  1. aBased on the factor construct of Welch et al. [8]; bBased on the factor construct of Miller et al. [13]; cBased on the factor construct of Huis In’t Veld et al. [15]; dBased on the factor construct of Papathanasiou et al. [12]; eBased on the factor construct of Snoek et al. [9]; fBased on the factor construct of McGuire et al. [21]; gEqual variances not assumed.