Skip to main content

Table 3 Test of goodness for fit of under-reporting field survey data and DSP dataset

From: Propensity score weighting for addressing under-reporting in mortality surveillance: a proof-of-concept study using the nationally representative mortality data in China

Variable

Proportions in DSP dataset (Pi)

Proportions in under-reporting survey dataset (Si)

(Si-Pi)^2/Pi

χ2

P

Age

     

 0-5

0.6

1.3

0.817

0.312

>0.05

 6-14

0.8

0.5

0.113

 15-44

8.1

8.0

0.001

 45-64

22.5

22.8

0.004

 65+

67.3

66.9

0.003

Cause of death

 Cancer

23.8

24.1

0.004

1.239

>0.05

 Cardiovascular disease

43.9

42.4

0.051

 Respiratory disease

9.0

11.1

0.490

 Nervous system disease

1.4

1.3

0.007

 Digestive system disease

2.4

2.3

0.004

 Urinary system disease

1.3

1.2

0.008

 Congenital anomalies

0.4

0.4

0.000

 Injury

8.2

9.0

0.078

 Infectious diseases

3.2

3.5

0.028

 Pregnancy, childbirth and the puerperium

0.5

0.6

0.020

 Other disease

5.9

4.1

0.549

Highest level of hospital where disease was diagnosed

 Provincial level

14.2

12.6

0.180

1.876

>0.05

 City level

28.0

25.8

0.173

 County level

34.2

40. 5

1.124

 Township level

13.1

12.3

0.049

 Village level

3.9

3.8

0.003

 Other

5.7

4. 5

0.257

 No treatment

1.0

0.7

0.090

Diagnostic criteria

 Symptoms + physio-biochemistry

55.8

57.6

0.058

0.360

>0.05

 Pathology

7.9

6.9

0.127

 Symptoms/signs

25.1

25.6

0.010

 Autopsy

0.6

0.6

0.000

 Surgery

1.7

1.4

0.053

 Inference

7.8

6.9

0.104

 Other

1.1

1.0

0.009

  
  1. ^2=square