THE ROLE OF TECHNOLOGY IN MEASUREMENT
As discussed in Chapter 7, developments in technology are affording new methods for data collection. For example, monitoring and sensing technologies are increasingly being used to monitor activities and track behaviors such as movement patterns or sleep behaviors. Wearable technologies such as smart watches that incorporate sensing and computing technologies are aimed at unobtrusively monitoring health indicators and providing feedback to the user. Advantages of these technologies are that they are for the most part unobtrusive and provide objective measurement of behavior in real time. However, there are also issues with privacy, data security and access, and data integration. For example, decisions need to be made about the schedule for data sampling (e.g., 24 hours/day; random times during a day). This has important implications for capturing relevant behavior patterns as well as for data management and analysis. Other important considerations are related to data coding and data integration. For example, in the PRISM trial, we collected real-time data on use of the PRISM software and had to make decisions about what constitutes actual use.
Other developments include the use of online data collection protocols and computer-assisted interviewing methods such as CATI. Use of CAT methods is also increasing, which allow assessments to be tailored to the level of the individual. The software program selects items that are relevant to the individual; the last response of the individual determines the next question that is asked. Items are selected from item banks that are assumed to represent the universe of potential items and levels of items for a particular construct or domain (e.g., health literacy). An example is the NIH-sponsored Patient-Reported Outcomes Measurement Information System (PROMIS) item bank for patient-reported health status for physical, mental, and social well-being (www.nihpromis.org). Potential benefits of these types of systems are efficiency (individuals are not asked questions that are irrelevant or beyond their ability level) and flexibility. Of course, use of these types of assessment techniques requires access to computer equipment, which can be costly. Development of item banks is also complex and time-consuming. Other concerns center around the psychometric properties of the assessment tool, determining branching protocols, and the content validity and currency of the item bank.