By default, SAS will include a constant term, or intercept , in the model.
Which model should be used? This depends on ...
● the design of the experiment (completely randomized design, randomized block design, split-plot
design, etc.)
● the manner in which strawberry plants and treatments were chosen and subsequently measured (for
example, repeated measures vs. destructive sampling)
● whether each factor is considered to be fixed or random (if levels of the factor were randomly
chosen from some population to represent variability in that population, the factor is random; if
levels were specifically chosen to be compared to each other, the factor is fixed)
● the principle of parsimony; the simplest appropriate model that adequately describes the results
should be used
● results from prior experience and from other statistical models
● the statistical reference book you decide to use (for example, some authors advocate using a
block*subplot interaction term in a split-plot experiment, while others assume that this interaction
will be small and do not include it in the model).
SAS can provide the numbers you will need to conduct your own data analyses for any approach which
involves linear models, but you should consult an appropriate reference when deciding which model to use.
To examine the output of PROC GLM, consider the example in which yield is modeled as a function of
strawberry variety, type of fertilizer, and their interaction.
By default, SAS will include a constant term, or intercept , in the model.
Which model should be used? This depends on ...
● the design of the experiment (completely randomized design, randomized block design, split-plot
design, etc.)
● the manner in which strawberry plants and treatments were chosen and subsequently measured (for
example, repeated measures vs. destructive sampling)
● whether each factor is considered to be fixed or random (if levels of the factor were randomly
chosen from some population to represent variability in that population, the factor is random; if
levels were specifically chosen to be compared to each other, the factor is fixed)
● the principle of parsimony; the simplest appropriate model that adequately describes the results
should be used
● results from prior experience and from other statistical models
● the statistical reference book you decide to use (for example, some authors advocate using a
block*subplot interaction term in a split-plot experiment, while others assume that this interaction
will be small and do not include it in the model).
SAS can provide the numbers you will need to conduct your own data analyses for any approach which
involves linear models, but you should consult an appropriate reference when deciding which model to use.
To examine the output of PROC GLM, consider the example in which yield is modeled as a function of
strawberry variety, type of fertilizer, and their interaction.
การแปล กรุณารอสักครู่..
