aStat manual Open aStat

Choosing an analysis

The right analysis depends on two things: what kind of outcome you measured, and how the experiment was laid out. aStat suggests one from your columns, but you know how the trial was run.

First: what is the experimental unit?

The experimental unit is the thing a treatment was applied to independently. If a diet was fed to a whole pen, the pen is the unit, even if you weighed every pig. Analyzing each pig as if it got the diet on its own overstates the evidence (pseudoreplication). Use pen averages, or a mixed model that includes the pen.

Measurements (weights, gains, yields)

Your questionLayoutUse
What do the numbers look like?AnySummary statistics
Is the mean different from a target?One groupOne-sample t-test
Do two groups differ?Different animals in each groupTwo-sample t-test
Did the same animals change?Before and after columnsPaired t-test
Do three or more groups differ?Completely randomizedOne-way ANOVA
Do treatments differ, allowing for barns, litters, or fields?Every treatment in every blockRCBD
Same, but blocks too small for every treatmentIncomplete blocksBIBD
Same, blocking in two directions (row and column)Each treatment once per row and columnLatin square
Do two factors act together?Every combination testedTwo-factor ANOVA
Do groups differ after allowing for starting weight?Group, plus a measurement taken before treatmentANCOVA
How does one measurement change with another?Two or more measurementsRegression or correlation
Do several measurements differ together?Several measurements per animalMANOVA

The same animal or plot measured more than once

Your questionLayoutUse
One factor on whole units, another within themRation per steer, aging time per cutSplit-plot
Do treatments differ over time?Each animal on one treatment, weighed repeatedlyRepeated measures
Each animal gets every treatment in turnPeriods with a washoutCrossover
How repeatable is a trait?Several records per animal, no treatmentRepeatability
Average growth curve, and how much animals differ in slopeEach animal measured at several timesRandom coefficients
Heritability and the best siresProgeny records by sire (and dam)Sire evaluation (BLUP)

Outcomes that are not measurements

OutcomeExampleUse
Yes or no, overall rateDystocia rate in a herdOne proportion
Yes or no, by group, no modelDystocia by sire breedTwo-way table
Yes or no, with treatments, blocks, or a dosePregnancy by protocol and herdLogistic or probit model
CountsFecal egg counts, lesions, piglets bornCount model
Ordered scoresLocomotion score 1 to 5Ordinal scores

Before the study

To find how many animals you need, use Power and sample size (the Plan a study button).