Skip to main content

Table 7 Analysis of the depressive symptoms severity levels for the validation group (N = 355)

From: Comparison of unweighted and item response theory-based weighted sum scoring for the Nine-Questions Depression-Rating Scale in the Northern Thai Dialect

Scoring Model

Depressive Symptoms Severity

Statistic

Overall Comparisona

Pairwise Comparisonb

None(n = 302)

Mild(n = 31)

Moderate(n = 14)

Severe(n = 8)

0

1

2

3

0 vs. 1

0 vs. 2

0 vs. 3

1 vs. 2

1 vs. 3

2 vs. 3

9Q unweighted sum score

1.37

6.52

15.76

32.25

Mean difference

 

5.15

14.39

30.88

9.24

25.73

16.49

(0–81 points)

(0.16)

(0.93)

(4.41)

(3.74)

F-statistic

164.82

35.64

133.24

356.25

80.35

190.88

237.65

(Reference)

    

RP

1.000

1.000

1.000

1.000

1.000

1.000

1.000

9Q frequency sum score

0.95

4.00

8.43

14.00

Mean difference

 

3.05

7.48

13.05

4.43

10

5.57

(0–27 points)

(0.10)

(0.43)

(1.55)

(1.55)

F-statistic

162.04

58.47

167.32

296.78

106.98

171.58

224.77

RP

0.983

1.641

1.256

0.833

1.331

0.899

0.946

9Q-GRM score

3.00

14.33

30.63

48.09

Mean difference

 

11.33

27.63

45.09

16.3

33.76

17.46

(0–81 points)

(0.32)

(1.69)

(4.54)

(5.04)

F-statistic

187.95

73.92

209.17

324.53

134.11

192.09

258.34

RP

1.140

2.074

1.570

0.911

1.669

1.006

1.087

9Q-GRM-DIF score

3.07

14.58

31.50

48.32

Mean difference

 

11.51

28.43

45.25

16.92

33.74

16.82

(0–81 points)

(0.32)

(1.67)

(4.59)

(4.79)

F-statistic

192.35

75.94

220.57

325.57

140.52

193.51

264.31

RP

1.167

2.131

1.655

0.914

1.749

1.014

1.112

9QSF-NRM score

2.22

10.35

21.20

35.09

Mean difference

 

8.13

18.98

32.87

10.85

24.74

13.89

(0–81 points)

(0.22)

(1.29)

(3.51)

(4.51)

F-statistic

172.23

68.40

177.61

310.47

116.41

182.73

236.37

RP

1.045

1.919

1.333

0.871

1.449

0.957

0.995

9QSF-NRM-DIF score

2.42

11.07

22.57

38.24

Mean difference

 

8.65

20.15

35.82

11.5

27.17

15.67

(0–81 points)

(0.25)

(1.35)

(3.99)

(4.94)

F-statistics

164.76

63.82

164.73

303.34

108.14

177.18

226.74

RP

1.000

1.791

1.236

0.851

1.346

0.928

0.954

  1. n frequency, GRM Graded-response model, DIF Differential item functioning, RP Relative precision
  2. a The overall comparison results obtained by using ANOVA with Bonferroni adjustment were significant for all of the models
  3. b The pairwise test results for the models using independent t-tests were significant
Back to article page

BMC Medical Research Methodology

ISSN: 1471-2288

Contact us