FACTOR ANALYSIS BY LAWLEY'S METHOD OF MAXIMUM LIKELIHOOD
FACTOR ANALYSIS BY LAWLEY'S METHOD OF MAXIMUM LIKELIHOOD
复制标题
劳利最大似然法因子分析
DOI：
10.1111/j.2044-8317.1949.tb00270.x
复制
发表时间：
1949
期刊：
影响因子：
--
通讯作者：
W. G. Emmett
中科院分区：
文献类型：
--
作者：
W. G. Emmett
在最近的一次调查中，作者 (I) 使用 DN Lawley 的因子分析最大似然法 (2,3,4) 来确定质心分析未能检测到的空间因子的显着性。这是可能的，因为劳利方法提供了对残差矩阵显着性的令人满意的检验；通过其他分析方法获得的重变量的显着性检验只是粗略的，并且可能导致错误的结论。劳利的分析方法很少被使用，可能是因为计算繁琐。然而，与收集测试分数和计算相关性相比，这种劳动是值得的，而且通常很小。劳利对其方法（4）的描述涉及对两个因素的分析，并且当涉及两个以上因素时需要放大。因此，在劳利博士的同意下，我们在此给出了三个或更多因素的分析方案，显示了可以应用的算术检查以及如何节省时间。到处都是我们的主题。内容可能与劳利博士在其他地方给出的内容重叠；这样做是为了连续性和完整性。
In a recent enquiry the writer (I) used DN Lawley’s Maximum Likelihood Method of Factor Analysis (2, 3, 4) to establish the significance of a spatial factor which a centroid analysis had failed to detect. This was possible in that Lawley’s method provides a satisfactory test of the significance of the residual matrix; tests of significance of reduals obtained by other methods of analysis are only crude and may lead to erroneous conclusions. Lawley’s method of analysis has been little used, perhaps on account of the laborious calculations involved; the labour, however, is well spent and is usually small compared with that of collecting test scores and ccimputing correlations.Lawley’s description of his method (4) relates to an analysis into two factors, and requires amplification when more than two factors are in question. We therefore give here, with Dr. Lawley’s assent, a scheme of analysis for three or more factors, showing arithifietical checks that may be applied and how time may be saved. Here and there our subject. matter may overlap that given elsewhere by Dr. Lawley; this has been done for the sake of continuity and completeness.