You are not currently logged in.
Access your personal account or get JSTOR access through your library or other institution:
Fitting Straight Lines when One Variable is Controlled
Journal of the American Statistical Association
Vol. 53, No. 281 (Mar., 1958), pp. 106-117
Stable URL: http://www.jstor.org/stable/2282571
Page Count: 12
Preview not available
Experiments are analyzed in which "runs" are made on the apparatus, the data are bivariate, observations on one variable are taken for a predetermined set of values for the observations on the other variable, and straight lines are fitted on each run to the observed points. The possibility is considered that the "true" line may vary from run to run because of uncontrolled factors causing differences between runs. The mathematical model generalizes one introduced by Berkson.
Journal of the American Statistical Association © 1958 American Statistical Association