作者: W. T. WILLIAMS , J. M. LAMBERT
DOI: 10.1038/191202A0
关键词:
摘要: A METHOD of subdividing an associated population the individuals which are specified by presence or absence a number attributes has been described previously1; process is ultimately derived from R-techniques factor analysis, and groups so obtained will be called R-groups. The may similarly subdivided into Q-groups corresponding Q-techniques. There results 2-way table in every attribute-individual record referred to one each kind group. requirement extract this most concentrated records, noda around regarded as varying. problem essentially that establishing coincidences between R Q factors; even normal analysis impaired formation correlation matrix2,3, subdivision involving successive matrices divergence well considerable.