Everyday Cycling in Urban Environments: Understanding Behaviors and Constraints in Space-Time - Computational Approaches for Urban Environments - page 189

Fig. 8.4 Reported travel mode by participants from travel diary

Tabl e 8. 2 Summary of reported trips and logged GPS points

All GPS points
(Raw)

Only cycle trips
(OCT) Item Participants All diary trips OCT GPS points

Tot al 79 2,432 2,415,666 941 205,987

Out of the 2,432 reported trips, 43 % of participants' trips were actually reported
to be cycle trips which are slightly less than combined trips for walking and car
use (Fig. 8.4 ). The other trips' mode comprised mainly motor bike, scooter, and
running/jogging.
The total GPS logged points for the sample for the 7-day period were almost
two and half million points using a log interval of 5 s. Thus, the smaller the
logging interval, the higher the likelihood of more point data which, if no careful
consideration is given to data handling, can prove difficult in when it comes to data
management and editing. For example, given a total 2,415,666 points in the case of
this research, all merged data was impossible to be stored in MS Excel workbook
given its limitation of 1,048,576 rows suggesting that without awareness, almost
half of the point data will be lost in the course of editing or storage (Table 8.2 ).
Given the fact that 941 trips were identified from the revealed route choice
preferences (i.e., the GPS-measured trips), 9.6 % of the cycle trips reported in the
travel diary were not recorded by the GPS device. The trend for reasons in data
loss for cycle trips follows similar observations made by Dill ( 2009 ), suggesting
that there are still challenges in the use of GPS technologies in capturing movement
behaviors.