First choose a compare command

Number

Test

Welch Anova testing Means Equal, allowing Std's Not Equal

Warning: Small sample sizes. Use Caution.

MeanAbsDif to Median

MeanAbsDif to Mean

Std Dev

Bartlett

Levene

Brown-Forsythe

O'Brien[.5]

Tests that the Variances are Equal

Van der Waerden Test (Normal Quantiles)

Median Test (Number of Points Above Median)

Wilcoxon / Kruskal-Wallis Tests (Rank Sums)

1-way Test, Chi-Square Approximation

Prob>|Z|

Z

S

2-Sample Test, Normal Approximation

(Mean-Mean0)/Std0

Score Mean

Score Sum

Count

Van der Waerden Test (Normal Quantiles)

Median Test (Number of Points Above Median)

Wilcoxon / Kruskal-Wallis Tests (Rank Sums)



t–Test

1-way Anova

Std Error uses a pooled estimate of error variance

 Display

Analysis

Assuming equal variances

Means and Std Deviations

Oneway Anova

Quantiles

Means for Oneway Anova

Std Err Mean

If a column has any negative values, the mean is significantly greater than the min.

If a column has any positive values, the mean is significantly less than the max.

Positive values show pairs of means that are significantly different.

Mean[i]-Mean[j]+LSD

Abs(Dif)-LSD

Mean[i]-Mean[j]-LSD

Abs(Dif)-LSD

Abs(Dif)-LSD

Comparisons with a control using Dunnett's Method

Comparisons with the best using Hsu's MCB

Comparisons for all pairs using Tukey-Kramer HSD

Comparisons for each pair using Student's t

|d|

d

q*

t

Dunnett's

With Control

Hsu's MCB

With Best

Tukey-Kramer

All Pairs

Student's t

Each Pair

Alpha=

Dif=Mean[i]-Mean[j]

Means Comparisons