Rae, A David2015-12-100889-311Xhttp://hdl.handle.net/1885/62678Some common misconceptions in least-squares crystal structure refinement can be resolved by recasting the usual equations in terms of partial observations, partial models and partial residuals. An observation has components that are determined by an initiKeywords: Background; Choice of variables; Comparative refinement; Competitive refinement; Constraints; Data merging; Duplicated and semi-duplicated observations; Global phase; Hierarchical model; Independent variables; Initiation steps; Least Square; Look-ahead op (3 + 1) dimensional crystallography.; Background; Choice of variables; Comparative refinement; Competitive refinement; Composite structures; Constraints; Data merging; Data merging with feedback; Dependent and independent variables; Detection of systematiThe use of partial observations, partial models and partial residuals to improve the least-squares refinement of crystal structures201310.1080/0889311X.2013.8168462016-02-24