TMean2Estimate Command

TTest2[<List of Sample Data 1>,<List of Sample Data 2>,<Level>,<Boolean Pooled>]

Calculates a T confidence interval estimate of the difference between two population means using the given sample data sets and confidence Level.

If Pooled = true, then population variances are assumed equal and sample standard deviations are combined in calculation.

If Pooled = false, then population variances are not assumed equal and sample standard deviations are not combined.

Results are returned in list form as {lower confidence limit, upper confidence limit}.

TMean2Estimate[<Sample Mean 1 >,<Sample Standard Deviation 1 >, <Sample Size 1>, <Sample Mean 2 >,<Sample Standard Deviation 2 >, <Sample Size 2>, <Level>,<Boolean Pooled>]

Calculates a T confidence interval estimate of the difference between two population means using the given sample statistics and confidence Level. Results are returned in list form as {lower confidence limit, upper confidence limit}.

Pooled is defined as above.

Note: This interval has definition: (sample mean1 - sample mean2) ± t* SE, where

n1, n2 = sample sizes

s1, s2 = sample standard deviations

If pooled = true

sp = pooled standard deviation

SE = sp * sqrt(1/n1 + 1/n2)

t* = critical value for given confidence Level and T distribution with degrees of freedom = n1 + n2 - 2.

If pooled = false

SE = sqrt(s1^2/n1 + s2^2/n2)

t* = critical value for given confidence Level and T distribution with degrees of freedom given by the Welch-Satterthwaite approximation .