Recall Bias in a Sportfishing Mail Survey

Abstract

Information on sportfishing activity is often collected through mail and telephone surveys. These surveys may rely on a respondent's memories of a previous year's activity, which are subject to recall errors. Limited research has been conducted to quantify the possible bias caused by recall periods less than 1 year. Data from a 1982 mail survey of registered Texas boat owners were used to determine whether estimated saltwater sport boat fishing effort differed based on recall periods of 1 month and 1 year. The mean annual number of days of saltwater fishing per registered boat owner was statistically similar for the two recall periods although variances were statistically different. However, the number of days of saltwater fishing was about 7.5 times higher for coastal county residents than for inland county residents. The pattern of use among boat access types and fishing areas was generally similar between coastal and inland county residents within each recall period. However, the estimated number of days each access type was used and the reported number of days each area was fished were significantly different between the two recall periods. Saltwater fishing participation estimates based on a 1-year recall could be adjusted using the relationship between the two recall periods to improve accuracy and precision of estimates and to reduce sampling costs. Fisheries managers could more effectively monitor changes in fishing activity for possible development of forecasting models.