Table 3

Distribution of the 'quality of statistical analysis' variables and median number of citations received by the publishing journal.

AGP

AJP

BJP

NJP

Total

P-value of chi square tests






N

(%)

Md

N

(%)

Md

N

(%)

Md

N

(%)

Md

N

(%)


Inappropriate analysis

0.044

▪yes

10

(11.1)

64

20

(14.8)

33.5

33

(17.9)

25

12

(30.8)

0.5

75

(16.7)

▪no

80

(88.9)

64

115

(85.2)

33

151

(82.1)

19

27

(69.2)

1

373

(83.3)

Incomplete analysis

0.018

▪yes

18

(20.0)

68

30

(22.2)

31.5

39

(21.2)

20

17

(43.6)

0

104

(23.2)

▪no

72

(80.0)

64

105

(77.8)

34

145

(78.8)

20

22

(56.4)

1

344

(76.8)

Overuse of tests

0.195

▪yes

29

(32.2)

52

38

(28.1)

34

56

(30.4)

19

18

(46.2)

1

141

(31.5)

▪no

61

(67.8)

66

97

(71.9)

32

128

(69.6)

22.5

21

(53.8)

1

307

(68.5)

All

90

135

184

39

448


The statistical significance of differences between variable categories was tested with Mann-Whitney U test. No statistically significant (p < 0.05) differences were found.

Nieminen et al. BMC Medical Research Methodology 2006 6:42   doi:10.1186/1471-2288-6-42

Open Data