Premium
Commission on Crystallographic Apparatus. Microdensitometer Project report I. Inter‐Experimental Agreement
Author(s) -
S. Abrahamsson,
P. Kierkegaar,
Eva Andersson,
Oliver Lindqvist,
Georg Lundgren,
L. SJOLm
Publication year - 1980
Publication title -
journal of applied crystallography
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.429
H-Index - 162
ISSN - 1600-5767
DOI - 10.1107/s0021889880012174
Subject(s) - microdensitometer , crystal (programming language) , analytical chemistry (journal) , symmetry (geometry) , consistency (knowledge bases) , intensity (physics) , crystallography , mathematics , materials science , physics , chemistry , optics , geometry , computer science , chromatography , programming language
The aim of the Microdensitometer Project was to investigate the agreement between intensity measurements performed by different laboratories. Each of fifteen participants was provided with four precession films prepared from two different crystals of sodium tartrate dihydrate: two films (A and B) of different exposure times from a small crystal and two similar films (C and D) from a larger crystal. A total of 33000 measured intensities and, in addition, 17000 scaled intensities were submitted for analysis. The inter-film factor between films A and B and C and D was timed to be 3.0 and the average values obtained from the different data sets were 2.90 and 3.00, respectively. The processing of these data sets included an analysis of the spread of intensities for symmetry-related reflections both within any one experiment and between experiments. In addition, a calculation using the analysis of variance technique has been made based on the weighted deviations of intensities from the set of mean values in order to locate errors from various sources. By using an P~m value defined later in the text as ZZllhk.,--lnul Rsym ~ hk s ZSIh k hk the internal consistency within each experiment was evaluated from the mm symmetry in the film plane. The values of R~ym were found to lie in the ranges 0.055-0.102 and 0.043-0.073 for the A and C films respectively. R~ym values obtained from the average data sets constructed from the participants' scaled data, B to A and D to C, were 0.057 and 0.050 respectively. R~u, values were obtained by scaling the data sets in pairs using the scaling procedure of Hamilton, Rollett & Sparks (1965). The /~o, value was then defined as Rm.,-- ~ II,- Ijl/~ Ili+ IjI hk hk