Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
8A convenience sample of 620 participants provided cell phone records for the CPRS, comprising less than 20% of the total number of participants in the SHRP 2 NDS. Antin, Stulce, Eichelberger, and Hankey (2015) described how the NDS sam- ple was collected and how it compared along a variety of dimen- sions to the overall driving population in the United States. It became apparent during the video verification process that even when the team believed the cell phone records matched to the correct participant and that the assignment of time zone was correct, this did not ensure that the participant was the one using his or her phone for a given event. For example, research staff may view video footage corresponding to the date and time of a cell phone call and document that the participant (driver) is not on the phone, but it appears that the front seat passenger may be using the phone (e.g., a phone is in the cup holder and just before the onset of the call the front seat passenger reaches over and picks up the phone). There were also scenarios in which no interaction with a cell phone could be detected for a given event, even though viewing other driving trips for that participant clearly indicated that the cell phone records were properly matched. Researchers who use the cell phone data in the future must keep in mind that there are a number of possi- ble scenarios that would account for a participant not having any involvement related to a call or text documented in the cell phone records. It is possible that a driver may not have heard his or her cell phone ring or felt it vibrate, or that a driver may have simply chosen to ignore an incoming cell phone call or text. Day-to-day situations where cell phone users loan their phone to family or friends to use for a day or accidentally leave their phone at home where it may be answered and used by others may also occur. It is also important to note that there may be months in which a given participant had few, if any, calls or texts. We may have participant data for the time period and know the number of texts or calls to be small, or even zero. At the onset of the study, it was deemed important to include data from a wide demographic of cell phone users. Therefore, the fact that there may be no calls made or received in a month may still be important data, and a month of zero calls or texts would still be included in the total months of data. However, it is also crucial that future users of the data under- stand that there are some months in which cell phone data may not be available for a given participant. In these cases, one must carefully distinguish between the aforementioned scenario of having data and the data reflecting zero calls, as opposed to the completely different scenario in which the data are simply unavailable and the number of calls or texts is unknown. In an effort to ensure these scenarios are clearly reflected in the data- base, a data availability table has been created for each partici- pant. This table will be part of the SHRP 2 NDS database and can be used by future researchers to ensure that any future analyses only take into account the time frame when data were actually available, regardless of the number of calls made dur- ing that period. A sample of this cell phone data availability table can be found in Appendix B. The data availability table also provides clear details regard- ing the types of data available for different time periods. For example, a participant may have provided 2 years of call data but only 3 months of text data. Also, Verizon and AT&T did not provide any information regarding the number of pic- tures or videos texted to and from participants, so the data table for records provided by the carriers will indicate that the number of pictures and videos sent and received for these participants is unavailable and unknown. Future Research Implications The data collected in the SHRP 2 NDS give researchers more than a million hours of naturalistic driving data that can be used to investigate a wide variety of topics, including the role of cell phone use while driving, especially just before and during safety-critical events. However, it must be noted that the cell phone data specifically referenced in this report were collected from the cell phone records of only 620 of the more than 3,100 C h a p t e R 4 Data Use Considerations
9SHRP 2 NDS participants. Still, these data will enable research- ers to quickly locate specific points of interest within the million hours of driving data in which a cell phone event might have occurred for one of these 620 participants. The cell phone data not only provide information about calls that took place while driving but also provide data about cell phone use in general. These data afford the opportunity for future researchers to gain insight into metrics of exposure and how different drivers may choose to restrict their cell phone use while driving. The cell phone data for a given primary participant are also tied to the driver assessment data provided by that participant; driver assessment data are much more limited for the 34 sec- ondary drivers enrolled in the CPRS. These primary partici- pant assessments were sampled from the following dimensions of functional capacity thought to be important to driving skill and safety: health, physical ability, perception, cognition, and psychomotor ability, as well as attitudes and perspectives on a variety of topics collected via a questionnaire (Antin et al. 2011). This information will allow future researchers to further investigate the roles that such factors play in a driverâs willing- ness to engage with the cell phone while driving and the rela- tive risks associated thereto.
