Power Analysis Nested ANOVA

[DP_Sydney]

Hi All,

 I need to calculate the sample size needed for a nested ANOVA aquacultural experiment examining diet (factor A) for both replicates of the nested factor (factor B; fish tanks) and replicate subsamples (fish). I am unsure how to use data from a pilot study in power analysis calculations.

 I have read a paper by Ling & Cotter (2003) 'Statistical power in comparative aquaculture studies' Aquaculture 224: 159-168, which provides complex (for me) formulae for calculating statistical power. This requires an estimate of the "variance of level B within level A" and the "error variances estimated by MSerror".

 How do I use the pilot data to obtain estimates of these two sources of variance? I could obtain the estimate of MSerror from an ANOVA table if I run a nested ANOVA using the pilot data, and I presume the MSB(A) value in the ANOVA table could be used for the "variance of level B within level A". Does this seem like the right approach, or is there an alternative means to calculating variances directly from the data?

Any help would be great since I'm out of my depth!

Cheers,
Dean

[Looman, Wendy]

I am out of the office March 12, 2012 through March 15, 2012. If you need assistance prior to March 16th, please contact: David M. Bass, Vice President for Research, Benjamin Rose Institute on Aging, Margaret Blenkner Research Institute, 11900 Fairhill Road, #300, Cleveland, OH 44120-1053. Phone: [phone]. Email: [hidden email]

 

If your message concerns the HHCAHPS surveys, please contact: Sue Ambro, Benjamin Rose Institute on Aging, Margaret Blenkner Research Institute, 11900 Fairhill Road, Suite 300, Cleveland, OH 44120-1053. Phone: [phone]. Email: [hidden email]

-----------------------------------------------------------------------------------------------------------------
CONFIDENTIALITY NOTICE: The documents accompanying this email may contain confidential information, which is legally privileged. This information is intended only for the use of the recipient named above. If you are not the addressee or the employee or agent of the intended recipient, you are hereby notified that you are strictly prohibited from printing, storing, disseminating, distributing, or copying this communication. If you have received this communication in error, please notify us immediately by replying to the message and deleting it from your computer.
-----------------------------------------------------------------------------------------------------------------

[MaxJasper]

Hi Dean,

 

Have you tried to check formulas and examples in:

 

Efficient analysis of growth trial data

Edward Nigel Ling

Aquaculture Research, 2007, 38, 728^732

 

Max.

 

 

 

Hi All,

 I need to calculate the sample size needed for a nested ANOVA aquacultural experiment examining diet (factor A) for both replicates of the nested factor (factor B; fish tanks) and replicate subsamples (fish). I am unsure how to use data from a pilot study in power analysis calculations.

 I have read a paper by Ling & Cotter (2003) 'Statistical power in comparative aquaculture studies' Aquaculture 224: 159-168, which provides complex (for me) formulae for calculating statistical power. This requires an estimate of the "variance of level B within level A" and the "error variances estimated by MSerror".

 How do I use the pilot data to obtain estimates of these two sources of variance? I could obtain the estimate of MSerror from an ANOVA table if I run a nested ANOVA using the pilot data, and I presume the MSB(A) value in the ANOVA table could be used for the "variance of level B within level A". Does this seem like the right approach, or is there an alternative means to calculating variances directly from the data?

Any help would be great since I'm out of my depth!

Cheers,
Dean

---

 

[Marta Garcia-Granero]

Hi Dean

The estimates you want can be computed used MIXED or VARCOMP:

VARCOMP
   Depvar BY Diet Tank
   /RANDOM = Tank
   /METHOD = SSTYPE(1)
   /DESIGN = Diet Tank WITHIN Diet.


HTH,
Marta GG

El 10/03/2012 7:18, DP_Sydney escribió:

> Hi All,
>
>   I need to calculate the sample size needed for a nested ANOVA aquacultural
> experiment examining diet (factor A) for both replicates of the nested
> factor (factor B; fish tanks) and replicate subsamples (fish). I am unsure
> how to use data from a pilot study in power analysis calculations.
>
>   I have read a paper by Ling&  Cotter (2003) 'Statistical power in
> comparative aquaculture studies' Aquaculture 224: 159-168, which provides
> complex (for me) formulae for calculating statistical power. This requires
> an estimate of the "variance of level B within level A" and the "error
> variances estimated by MSerror".
>
>   How do I use the pilot data to obtain estimates of these two sources of
> variance? I could obtain the estimate of MSerror from an ANOVA table if I
> run a nested ANOVA using the pilot data, and I presume the MSB(A) value in
> the ANOVA table could be used for the "variance of level B within level A".
> Does this seem like the right approach, or is there an alternative means to
> calculating variances directly from the data?
>
> Any help would be great since I'm out of my depth!
>
> Cheers,
> Dean
>
> --
> View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Power-Analysis-Nested-ANOVA-tp5552347p5552347.html
> Sent from the SPSSX Discussion mailing list archive at Nabble.com.
>
> =====================
> To manage your subscription to SPSSX-L, send a message to
> [hidden email] (not to SPSSX-L), with no body text except the
> command. To leave the list, send the command
> SIGNOFF SPSSX-L
> For a list of commands to manage subscriptions, send the command
> INFO REFCARD
>

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD