Development and reliability testing of a fast-food restaurant observation form

Am J Health Promot. 2015 Sep-Oct;30(1):9-18.
Authors: Rimkus L, Ohri-Vachaspati P, Powell LM, Zenk SN, Quinn CM, Barker DC, Pugach O, Resnick EA, Chaloupka FJ.

Purpose: To develop a reliable observational data collection instrument to measure characteristics of the fast-food restaurant environment likely to influence consumer behaviors, including product availability, pricing, and promotion. Design: The study used observational data collection. Setting: Restaurants were in the Chicago Metropolitan Statistical Area. Subjects: A total of 131 chain fast-food restaurant outlets were included. Measures: Interrater reliability was measured for product availability, pricing, and promotion measures on a fast-food restaurant observational data collection instrument. Analysis: Analysis was done with Cohen's coefficient and proportion of overall agreement for categorical variables and intraclass correlation coefficient (ICC) for continuous variables. Results: Interrater reliability, as measured by average coefficient, was .79 for menu characteristics, .84 for kids' menu characteristics, .92 for food availability and sizes, .85 for beverage availability and sizes, .78 for measures on the availability of nutrition information, .75 for characteristics of exterior advertisements, and .62 and .90 for exterior and interior characteristics measures, respectively. For continuous measures, average ICC was .88 for food pricing measures, .83 for beverage prices, and .65 for counts of exterior advertisements. Conclusion: Over 85% of measures demonstrated substantial or almost perfect agreement. Although some measures required revision or protocol clarification, results from this study suggest that the instrument may be used to reliably measure the fast-food restaurant environment.