Development of a Pesticide Log: Design and Testing of a Three-Month Diary of Infrequent Behavior (303548)

Keywords: survey diary, design, cognitive testing

Survey diaries require respondents to record certain behaviors over a period of time, whether a day, a week, or even longer. Researchers use diaries to collect self-report data for a wide range of behaviors, including daily activities, travel, diet, exercise, and purchases. Theoretically, a diary provides accurate data because behavior is recorded as it happens. In reality, the burden of real-time tracking and recording may lead some respondents to under-report behaviors to minimize burden, or to postpone completing the diary until the end of the time period, possibly leading to under- or over-reporting. Given this, survey diary design, if successful in measuring behavior, must seek to minimize respondent burden and encourage timely, accurate recording of the behavior of interest. This is particularly true when complex details must be recorded about a behavior or the behavior is infrequent.

This paper will discuss the design and testing of a diary to capture household use of pesticides over a three-month period. Household pesticide use is particularly challenging to capture, since it covers a wide variety of product types, including ant baits, citronella candles, roach bombs, fly strips, termite treatments, lice shampoo, bug spray applied to the skin, and pet flea collars. Frequency of use varies widely by season and household. The diary also captures a high level of detail, including the pesticide form (e.g., spray, shampoo, powder), quantity, dates and frequency of application, reason for application, who applied it (and when applicable, on whom) and the EPA Registration Number. We will discuss our findings, including lessons learned from two rounds of cognitive testing during the design process. We will also discuss how elements such as terminology, layout, visual and creative organization can help minimize burden and encourage accurate reporting.