Lab 3 - Experimental Design and ANOVA Flashcards
classes of variation withing statistics
treatment
experimental unit
experimental design
treatment
the factor imposed by the researcher to generate variation
experimental unit
the unit to which the treatment is applied
experimental error
the variation between similar treatments and the experimental units that cannot be accounted for
experimental design
set of statistically-based rules for laying out an experiment that allows data collection for analysis
also reduces random error
principles of experimental design
replication
randomization
local control/blocking
purpose of replication
reduces error magnitude
purpose of randomization
eliminates biasness and each unit has equal chance of receiving each treatment
purpose of local control/blocking
effective in case of environmental gradient
types of common experimental designs
- completely randomized design
- randomized complete block design
- latin square
- split plot design
completely randomized design
- treatments are assigned randomly so that each treatment has the same chance of receiving any one treatment
- simplest design, low error
when is completely randomized design best used
best for experiments where experimental materials are homogenous and not heterogenous
what experimental design is this
completely randomized design
randomized complete block design
- uses blocking/grouping of similar experimental units where each block is the same size and equal to the number of treatments
- each treatment is randomly assigned t 1 experimental unit in every block
- helps with heterogeneity
what type of experimental design is this
randomized complete block design
latin square
- 2 blocking/groupings of treatment and experiments, where for every treatment, there is a replicate
- treatment occurs once in every row and column
when is the latin square design most suitable
when there are a small number of entries/treatments
what type of experimental design is this
Latin square
split plot design
main plot treatments are randomly assigned to main plots and then the subplot treatments are randomly assigned in each main plot
what type of experimental design is this
split plot design
what is ANOVA
analysis of variance stats methods the separates observed variance into its different components
2 types of ANOVAs
one way
two-way
one way ANOVA
evaluates the impact of one independent variable on one dependent variable
two-way ANOVA
evaluates the impact of 2 or more independent variables on one dependent variable
what is the assumption that needs to be met before we can conduct an ANOVA
they need to be normally distributed
