文献里面的描述：
Statistical analyses were performed by using statistical software (SPSS Statistics version 20, IBM, Armonk, NY; and R version 3.5.1, the R Foundation for Statistical Computing, Vienna, Austria). For a single test, P values less than .05 were considered to indicate statistical significance. We used Bonferroni correction in all post hoc comparisons throughout the study to control for multiple comparisons and we show the corrected P value except when comparing multiple brain regions, wherein we indicated the resulting new significance level. Group differences in Table 1 were tested by using Kruskal-Wallis test, χ2 test, or analysis of covariance, as appropriate. Shapiro-Wilk test (P < .05) showed deviations from the normality distribution in MH, tau, and amyloid β PET. Nonparametric Spearman correlation adjusted for sex, age, and education was subsequently used.
请问这里是怎么adjusted for sex, age的呢，SPSS里面的偏相关好像可以，但是是参数方法。
我自己是因为先在SPM里面用GLM在adjusted for age sex找到了与评分显著相关的脑区，之后我想提取脑区内的数据，来重现这种相关的结果，但是我的评分的分布是非正态的，不符合Pearson相关的假设，可能要用非参数的方法。那我这时候是直接把脑区里面的均值提取出来以后，用一个只有age sex 的模型的残差来和评分做非参数的相关可以吗。