Skip to main content

Table 4 Linear and mixed effect models to estimate the differences in GPS-based and self-reported travel time (self-reported – GPS) for home to work and work to home trips

From: Travel patterns during pregnancy: comparison between Global Positioning System (GPS) tracking and questionnaire data

 

General linear regression

Mixed effect model (fixed effects)

 

Beta

Standard error

p-value

Partial R2

VIF

Beta

Standard error

p-value

Home to work trips (93 tripsa from 31 subjects)

(Model R2 = 0.33; cross validation R2 = 0.29)

 

Intercept

-17.62

5.11

0.0009

 

0

-17.16

8.13

0.0439

Self-reported distance for home to work trips

-0.27

0.08

0.0011

0.09

1.53

-0.28

0.12

0.0196

Vehicle make (Japanese car: 1; other: 0)

7.81

2.24

0.0008

0.09

1.07

7.24

3.44

0.0397

Rush hour (yes/no)

-7.42

2.52

0.0041

0.07

1.07

-5.97

2.29

0.0114

More than two persons in the household (yes/no)

6.52

2.29

0.0055

0.04

1.06

5.72

3.48

0.1054

Percentage of travel time on freeways

16.12

4.43

0.0005

0.04

1.56

19.49

5.78

0.0013

Work to home trips (89 trips from 31 subjects)

(Model R2 = 0.36; cross validation R2 = 0.31)

 

Intercept

0.71

4.50

0.8752

 

0

0.45

4.91

0.9272

Self-reported duration for work to home trips

0.48

0.07

<.0001

0.32

1.00

0.46

0.10

<.0001

Weekday (yes/no)

-10.80

4.43

0.0165

0.04

1.00

-10.57

3.99

0.0105

  1. aExcluding four trips with missing self-reported distance.