Skip to main content

Table 3 Repeated measure analysis with the linear mixed model for Patient Falls and Patient Injury Falls for 2007 NDNQI 3rd quarter

From: The Box-Cox power transformation on nursing sensitive indicators: Does it matter if structural effects are omitted during the estimation of the transformation parameter?

Source of Variation distribution

Degree of Freedom

F-Value

Prob > F

Residual goodness-of-fit test for normal

Indicator: Total Falls (transformation with additive effect)

Teaching Status

2

2.36

0.0945

 

Magnet Status

1

3.83

0.0505

 

Teaching × Magnet

2

5.15

0.0058

0.071

Indicator: Total Falls (transformation without additive effect)

Teaching Status

2

2.73

0.065

 

Magnet Status

1

5.91

0.0151

 

Teaching × Magnet

2

7.14

0.0008

0.023

Indicator: Total Falls (no transformation)

 

Teaching Status

2

2.14

0.1178

 

Magnet Status

1

5.90

0.0151

 

Teaching × Magnet

2

6.94

0.001

0.097

Indicator: Total Injury Falls (transformation with additive effect)

Teaching Status

2

1.83

0.1603

 

Magnet Status

1

9.37

0.0022

 

Teaching × Magnet

2

4.14

0.016

0.117

Indicator: Total Injury Falls (transformation without additive effect)

Teaching Status

2

1.37

0.2536

 

Magnet Status

1

9.78

0.0018

 

Teaching × Magnet

2

4.45

0.0118

0.029

Indicator: Total Injury Falls (no transformation)

Teaching Status

2

6.55

0.0014

 

Magnet Status

1

3.63

0.0569

 

Teaching × Magnet

2

3.29

0.0373

0.146

  1. Repeated measure analysis for Patient Falls and Patient Injury Falls for a) No transformation, b) transformed without structural effects), and c) transformed with structural effects. Residual goodness-of-fit tests for normal distribution is based on Kolmogorov-Smirnov statistic.