Date of Graduation


Document Type


Degree Name

Master of Science in Kinesiology (MS)

Degree Level



Health, Human Performance and Recreation


Erin Howie

Committee Member

Kaitlin Gallagher

Second Committee Member

Michelle Gray


Accelerometers, Physical Activity, Preferred vs Non-preferred Wrist


Introduction. People who participate in regular physical activity have a decreased risk of chronic diseases and premature death. A dramatic decrease of physical activity occurs from adolescence to young adulthood. With important implications to health, physical activity is an important behavior to measure. However, inconsistencies exist on how to measure physical activity. When using accelerometers, differences between the preferred or non-preferred wrist may result in different estimates of physical activity. Purpose. The purpose of this study was to compare the preferred and non-preferred wrist accelerometry measured physical activity using commonly used research accelerometers during structured daily college activities (Actigraph GT3x-bt and GT9X Link) and free-living conditions of college students (Actigraph GT9X Link). Methods: 30 college students (15 females and 15 males) completed 7 laboratory tasks including shooting a basketball (BB), relaxing on a couch (Relax), hitting a racquetball (RB), walking up and down stairs (WUS), walking on an inclined surface (WUI), walking while using a smart phone (WSP), and using a laptop (COM). An Actigraph GT3x-bt and Actigraph Link on each wrist and the right hip. After the tasks, the students completed one week of free-living conditions wearing an Actigraph Link on each wrist. Accelerometer counts from the preferred and non-preferred wrists were compared using Wilcoxon signed-rank tests for the lab activities and a paired t tests for the free-living conditions with α at .05. Results: Preferred and non-preferred total counts per minute from the Actigraph Link were significantly different for BB (p= <.001), COM (p=.004), RB (p= <.001), Relax (p=.027), WSP (p=.001), and WUS (p=.043). The free-living conditions showed no significant differences between the preferred and non-preferred wrist. Conclusion. Researchers should be aware when measuring physical activity in structured activities that the preferred and non-preferred wrist can affect the measurement. Though for free living conditions, less concern should be placed on the preferred or preferred wrist.