General Vocab
The entire group that a researcher wants to learn about
Population
A method that choses a random starting point and then selects every kth individual
Systemic Sample
This ANOVA sum of squares measures variability between the group means
Sum of Squares Treatment (SSTr)
The alternative hypothesis for an ANOVA test comparing 4 means
At least one mean is different
You use the sample mean (x bar) to estimate this
The population mean (mu)
A specific value, category, or setting of a factor
Level
A method that divides the population into relatively homogenous (similar) groups, then takes an SRS from each group
Stratified Random Sample
The mean square error when an ANOVA compares 5 groups with 45 total observations, and the error sum of squares is 352
MSE=8.80.
This is how Tukey Kramer test determines WHICH mean(s) is/are significantly different from the others
Creating confidence intervals
Grouping similar subjects by a characteristic that we cannot control
Blocking
A design in which one or both groups - either subjects and/or evaluators - does not know which treatment was received
Blinding
Bias that occurs when part of the population is absent or underrepresented in the sampling frame
Undercoverage
A study compares four sleep routines with 36 total participants. SSTotal=510 and SSTr=153. Find SSE, then find the F statistic
SSE= 357 F=4.57
The estimated p-value and decision of our hypothesis test when the researcher’s ANOVA test statistic is F=5.6
P-value <0.01; reject H0; there is significant evidence that at least one of the means differs
Interpretation of a 95% CI
95% Confident that the true mean population is within this interval
How much mean of y is expected to vary from sample to sample due to random chance
Standard Error
4 Principles of Experimental Design
Control, Randomize, Replicate, Block
The conclusion when an ANOVA result has p=0.078 at α=0.05
Fail to reject H0; there is not sufficient evidence that at least one population mean differs.
The F statistic when you have 3 groups and 30 total subjects. SSTr=40 and SSE=162
F=3.33
Your decision (reject or fail to reject) when a null hypothesis is assuming there is no difference between two means and CI is reported as (-1.5, 0.5)
Fail to reject
How far the true population mean is likely to be from your sample mean
Margin of Error
A bigger sample makes your SE ___ (bigger or smaller) and a CI ___ (narrower or wider)
Smaller, Narrower
MSTr is variance ___ (explained or unexplained) MSE is variance ___ (explained or unexplained)
MSTr is explained; MSE is unexplained
JMP output for a Tukey-Kramer test reports that the CI for the difference in mean responses from levels Aspirin and Motrin is (0.51, 1.5). This shows what about the mean reponses?
They are significantly different
The kind of variables that correlation uses
Quantitative