This is the authors' abstract. We don't add key points for this paper.
We have developed three programs for data processing for radioimmunoassays RIA implemented on a small, inexpensive, programmable calculator (Texas Instrument Co., Model 59): 1) The first program performs a weighted logit-log regression for analysis of the RIA standard curve, and provides several descriptive statistics: slope, intercept, ED50 (and their corresponding standard errors), residual variance, and correlation coefficient. This program also provides: dose interpolation for unknowns; an estimate of the precision (percent coefficient of variation) for each result; corrections for variable sample volumes or recoveries; and the mean and standard error of the mean for samples analyzed in replicate. 2) The second program uses the "four parameter logistic model" to describe the dose responses curve and perform dose interpolation. This is the equivalent to the use of the logit-log method, with provision for adjustment for the position of the 0% and 100% response. As such, it is more flexible and versatile than the logit-log method. 3) A third program is used for routine Within-Lab Quality Control: it calculates within-assay and between-assay precision, utilizing an analysis of variance (ANOVA) with a components of variance estimation. These programs incorporate the most important features of programs previously developed in this laboratory for an IBM 370 or a DECsystem-10 computer, and demonstrate the availability of adequate statistical analyses to laboratories without access to large centralized computer facilities.
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Davis et al. (1980) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: