STP1055

    On the Use of the R50

    Published: Jan 1989


      Format Pages Price  
    PDF Version (112K) 6 $25   ADD TO CART
    Complete Source PDF (1.5M) 6 $55   ADD TO CART


    Abstract

    R50 is a statistic used to describe the repellency of a substance in the manner that an LC50 is used to describe toxicity. Although the R50 is calculated using the same methods as for estimating an LC50, the data initially collected on individual animals are proportional, not binary. Transforming nonbinary data into binary data is subjective, can result in substantial losses of information, and, consequently, can result in a greater chance for erroneous inferences concerning the repellency of a substance. The subjective nature of R50 estimation arises because a test animal is defined as repelled if less than 50% of treated food items are consumed. The R50 is calculated as the concentration at which 50% of the population consumes less than 50% of the food times. It is more appropriate and of more direct interest to estimate the concentration of a substance at which an average of 50% of the food items are consumed rather than to estimate the R50.

    Problematic aspects of R50 estimation are demonstrated in this paper, and alternative analyses are recommended for studying the repellency of a substance. Example data sets are used to compare inferences from estimation of an R50 versus use of inverse regression methods to estimate concentrations at which various proportions of food items are consumed.

    Keywords:

    R, 50, repellency, inverse regression


    Author Information:

    Engeman, RM
    Statistician, chief of bird control research section, statistician, and chief of research support section, U.S.D.A.-APHIS, Science and Technology, Denver Wildlife Research Center, Bldg. 16, Denver Federal Center, Denver, CO

    Otis, DL
    Statistician, chief of bird control research section, statistician, and chief of research support section, U.S.D.A.-APHIS, Science and Technology, Denver Wildlife Research Center, Bldg. 16, Denver Federal Center, Denver, CO

    Bromaghin, JF
    Statistician, chief of bird control research section, statistician, and chief of research support section, U.S.D.A.-APHIS, Science and Technology, Denver Wildlife Research Center, Bldg. 16, Denver Federal Center, Denver, CO

    Dusenberry, WE
    Statistician, chief of bird control research section, statistician, and chief of research support section, U.S.D.A.-APHIS, Science and Technology, Denver Wildlife Research Center, Bldg. 16, Denver Federal Center, Denver, CO


    Paper ID: STP25256S

    Committee/Subcommittee: E35.17

    DOI: 10.1520/STP25256S


    CrossRef ASTM International is a member of CrossRef.