Use of SOPARC in a Large-Scale, Multisite Research Project

Background/Purpose: Faculty are increasingly being asked to seek and participate in large scale, multi-site research funded by private and federal agencies. We describe a collaborative effort among five universities and a private corporation in which researchers from different academic fields are participating in the 4-year NIH-funded project titled, Validation of SOPARC (System for Observation of Parks and Recreation in Communities) and use in Diverse Populations. Method: The study's aims are to: (1) assess SOPARC in different populations, geographic settings, and seasons to determine its utility nationally; (2) validate SOPARC relative to park use measurement and population level energy expenditure; and (3) test the reliability and validity of self reported physical activity in parks. SOPARC and other data have been collected in 30 neighborhood parks in Los Angeles, Philadelphia, Chapel Hill, Albuquerque and Columbus during three different seasons. We describe the training of assessors, how data were collected and managed, and how SOPARC data are incorporated into analyses with those from GPS, accelerometry, dietary logs, park-use surveys, and neighborhood audits on physical incivilities. Analysis/Results: SOPARC data has been collected on over 76,000 individuals in the five cities. Interobserver reliabilities using SOPARC are high (>91%). Annual meetings and monthly conference calls are sufficient to solve immediate problems. Use of the internet to send data to a central location for processing has worked superbly. Conclusions: SOPARC is generalizable and usable within five diverse US regions. We conclude by discussing management issues related to communication, institutional review, training, data collection, technology, and analysis.