Statistics applied to chemistry: ten common doubts

Quimica nova

Bruns, Roy Edward¹  Passari, Livia Maria Zambrozi Garcia¹  Soares, Patricia Kaori¹  Scarminio, Ieda Spacino² 

[1] Universidade Estadual de Campinas, Campinas, Brasil;Universidade Estadual de Londrina, Londrina, Brasil

关键词: linear regression;    median;    non-parametric statistics.;   

DOI  :  10.1590/S0100-40422011000500028

学科分类：化学（综合）

来源: Sociedade Brasileira de Quimica

【 摘 要 】

Ten common doubtsof chemistry students and professionals about their statistical applicationsare discussed. The use of the N-1 denominator instead of N is described forthe standard deviation. The statistical meaning of the denominators of the rootmean square error of calibration (RMSEC) and root mean square error of validation(RMSEV) are given for researchers using multivariate calibration methods. Thereason why scientists and engineers use the average instead of the median isexplained. Several problematic aspects about regression and correlation aretreated. The popular use of triplicate experiments in teaching and researchlaboratories is seen to have its origin in statistical confidence intervals.Nonparametric statistics and bootstrapping methods round out the discussion.

【 授权许可】

CC BY