Related Data for: Overall preference of running shoes can be predicted by suitable perception factors using a multiple regression model

Objective: This study examined (a) the strength of four individual footwear perception factors to influence the overall preference of running shoes and (b) whether these perception factors satisfied the nonmulticollinear assumption in a regression model. Background: Running footwear must fulfill multiple functional criteria to satisfy its potential users. Footwear perception factors, such as fit and cushioning, are commonly used to guide shoe design and development, but it is unclear whether running-footwear users are able to differentiate one factor from another. Methods: One hundred casual runners assessed four running shoes on a 15-cm visual analogue scale for four footwear perception factors (fit, cushioning, arch support, and stability) as well as for overall preference during a treadmill running protocol. Results: Diagnostic tests showed an absence of multicollinearity between factors, where values for tolerance ranged from .36 to .72, corresponding to variance inflation factors of 2.8 to 1.4. The multiple regression model of these four footwear perception variables accounted for 77.7% to 81.6% of variance in overall preference, with each factor explaining a unique part of the total variance. Conclusion: Casual runners were able to rate each footwear perception factor separately, thus assigning each factor a true potential to improve overall preference for the users. The results also support the use of a multiple regression model of footwear perception factors to predict overall running shoe preference. Application: Regression modeling is a useful tool for running-shoe manufacturers to more precisely evaluate how individual factors contribute to the subjective assessment of running footwear.