Development and Validation of a Weather-Based Physical Activity Index

Purpose: The purposes of this study were to develop weather-based index to predict physical activity, and to validate the Physical Activity Index by examining relationships between participants' actual physical activity level and the index score. Methods: Participants were 200 healthy order women (age: 64.7±2.3 yrs., height: 156.1±3.5 cm, weight: 57.8±2.4 kg, BMI: 23.8±1.4 kg/m2) from five different Centers for the Welfare of the Aged in Seodaemun-gu, Eunpyeong-gu, Songpa-gu, Kangbuk-gu, and Kangnam-gu in Seoul, Korea. Their daily steps were counted and recorded for 365 consecutive days using Omron HJ-113 pedometer. Weather variables, including daily average temperature, high temperature, low temperature, cloudiness, and precipitation, were obtained from the Korea Meteorological Administration for a one-year period, from January 1st through December 31st in 2007. Analysis/Results: First, the five weather variables were used to calculate the factor score, the Physical Activity Index. Principal Axis Factoring (the number of factors was fixed at one) and regression method were used to obtain factor scores. The factor scores were standardized, such that the distribution has a mean of zero and a standard deviation of one. To establish validity evidence, comparisons were made between the number of participants' steps averaged over the year period and the Physical Activity Index. Two hundred participants' step counts were averaged each day, and were compared to the factor score for that day. The participants' step counts were positively correlated with the Physical Activity Index (r = .40), which supported correlational evidence of validity. Conclusions: Therefore, the Physical Activity Index can significantly predict physical activities of order women. Using meteorological data to develop the Physical Activity Index for other age groups could be beneficial to promote physical activity. A better index could be obtained if more variables (e.g., daily ozone level and humidity) were included in the formation of factor scores. Because increase/decrease in temperature or weather variables do not always mean better/worse weather for participation for physical activities, further development of the Physical Activity Index is necessary.