A comparison of recent approaches to the analysis of repeated measurements

被引:46
作者
Keselman, HJ
Algina, J
Kowalchuk, BK
Wolfinger, RD
机构
[1] Univ Manitoba, Dept Psychol, Winnipeg, MB R3T 2N2, Canada
[2] Univ Florida, Gainesville, FL 32611 USA
[3] SAS Inst, Cary, NC USA
关键词
D O I
10.1348/000711099158964
中图分类号
O1 [数学];
学科分类号
0701 ; 070101 ;
摘要
Looney & Stanley's (1989) recommendations regarding analysis strategies for repeated measures designs containing between-subjects grouping variables and within-subjects repeated measures variables were re-examined and compared to recent analysis strategies. That is, corrected degrees of freedom univariate tests, multivariate tests, mixed model tests, and tests due to Keselman, Carriere & Lix (1993) and to Algina (1994), Huynh (1978) and Lecoutre (1991) were compared for rates of Type I error in unbalanced non-spherical repeated measures designs having varied covariance structures and no missing data on the within-subjects variable. Heterogeneous within-subjects and heterogeneous within- and between-subjects structures were investigated along with multivariate non-normality. Results indicated that the tests due to Keselman et al. and Algina, Huynh and Lecoutre provided effective Type I error control whereas the default mixed model approach computed with PROC MIXED (SAS Institute, 1995) generally did not. Based on power differences, we recommend that applied researchers adopt the Welch-James type test described by Keselman et al.
引用
收藏
页码:63 / 78
页数:16
相关论文
共 51 条
[1]   NEW LOOK AT STATISTICAL-MODEL IDENTIFICATION [J].
AKAIKE, H .
IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 1974, AC19 (06) :716-723
[2]   SOME ALTERNATIVE APPROXIMATE TESTS FOR A SPLIT-PLOT DESIGN [J].
ALGINA, J .
MULTIVARIATE BEHAVIORAL RESEARCH, 1994, 29 (04) :365-384
[3]  
Algina J, 1997, BRIT J MATH STAT PSY, V50, P243
[4]   A power comparison of the Welch-James and improved general approximation tests in the split-plot design [J].
Algina, J ;
Keselman, HJ .
JOURNAL OF EDUCATIONAL AND BEHAVIORAL STATISTICS, 1998, 23 (02) :152-169
[5]   Testing repeated measures hypotheses when covariance matrices are heterogeneous: Revisiting the robustness of the Welch-James test [J].
Algina, J ;
Keselman, HJ .
MULTIVARIATE BEHAVIORAL RESEARCH, 1997, 32 (03) :255-274
[6]   TYPE-I ERROR RATES FOR HUYNH GENERAL APPROXIMATION AND IMPROVED GENERAL APPROXIMATION TESTS [J].
ALGINA, J ;
OSHIMA, TC .
BRITISH JOURNAL OF MATHEMATICAL & STATISTICAL PSYCHOLOGY, 1994, 47 :151-165
[7]   AN IMPROVED GENERAL APPROXIMATION TEST FOR THE MAIN EFFECT IN A SPLIT-PLOT DESIGN [J].
ALGINA, J ;
OSHIMA, TC .
BRITISH JOURNAL OF MATHEMATICAL & STATISTICAL PSYCHOLOGY, 1995, 48 :149-160
[8]  
[Anonymous], 1996, Journal of Agricultural Biological and Environmental Statistics
[10]   ROBUSTNESS [J].
BRADLEY, JV .
BRITISH JOURNAL OF MATHEMATICAL & STATISTICAL PSYCHOLOGY, 1978, 31 (NOV) :144-152