-4

the "mixedDesign" function was written by the professor for this homework, but I think the argument "sd" is standard as I've found it on other functions but with no description of what the value should actually describe. What am I actually saying when I designate a value for sd? Professor says 0.1 is "too low". Is there a standard value for this argument? 

simdata  <- mixedDesign(B=c(2, 2), W=2, n=10, M=Means, 
            SD=0.1, R=.42, empirical = TRUE, long = TRUE)

I'm aware that standard deviation is 34.1% either side of the mean!

Many thanks in advance!

aybry
  • 316
  • 1
  • 7
  • 5
    How on earth do you expect us to answer that? Ask your professor or show us the entire function definition. – Rich Scriven Dec 07 '15 at 23:33
  • Because, as I said, 'I think the argument "sd" is standard' - but it is apparently not?! I'll have a look... – aybry Dec 07 '15 at 23:36
  • 1
    The argument `SD=` it not "standard" – MrFlick Dec 07 '15 at 23:43
  • 1
    There aren't "standard" arguments. That's not a thing. – Gregor Thomas Dec 07 '15 at 23:56
  • No. Guessing from the rest of the context, this is some sort of function for simulating data for an experimental design to be fed into a mixed-model analysis. `SD` **might** be specifying a residual standard deviation. But as other commenters have said, there's just not enough to go on here. – Ben Bolker Dec 08 '15 at 00:12
  • Many thanks, Ben Bolker and others. I think I've figured it out. And I'll try to be a little more specific with future questions... – aybry Dec 08 '15 at 00:14

1 Answers1

1

A bit of googling leads here:

SD: Matrix specifying the cell standard deviations of crossing between- and within-subject factors (dimensions: prod(B)-by-prod(W) matrix) (for pure within-subjects designs, it is possible to input a vector of size prod(W) as an alternative to a 1-by-prod(W) matrix) OR a scalar (single number) that will be the standard deviation of all cells

Ben Bolker
  • 211,554
  • 25
  • 370
  • 453