[R-sig-ME] Fwd: Re: [R] ANOVA and Pseudoreplication in R

Bert Gunter gunter.berton at gene.com
Sun Feb 27 02:31:43 CET 2011


Hi Ben:

1) Confession: I did not and have not read your post in detail.

2) IMHO, the following advice:

> "Pseudo replication is really about a lack of independence between
> measurements, So you need to work backwards and see where you are building
> in a known lack of independence.  And where that is the case you need to use
> means of all the values."

is baloney. The "lack of independence" part is correct. The baloney
part is "take the mean." That is exactly where mixed models -- or
hierarchical modeling of some sort -- is required. That's why I
referred you to R-sig-mixed-models.

Again, please note: IMHO. Maybe I'm the one full of baloney. It's free
advice, after all, so beware of what you get.

Cheers,
Bert


On Sat, Feb 26, 2011 at 7:44 AM, Ben Ward <benjamin.ward at bathspa.org> wrote:
> On 25/02/2011 21:22, Ben Ward wrote:
>>
>> -------- Original Message --------
>> Subject:        Re: [R] ANOVA and Pseudoreplication in R
>> Date:   Fri, 25 Feb 2011 12:10:14 -0800
>> From:   Bert Gunter<gunter.berton at gene.com>
>> To:     Ben Ward<benjamin.ward at bathspa.org>
>> CC:     r-help<r-help at r-project.org>
>>
>>
>>
>> I can hopefully save bandwidth here by suggesting that this belongs on
>> the R-sig-mixed-models list.
>>
>> -- Bert
>>
>> As an aside, shouldn't you be figuring this out yourself or seeking local
>> consulting expertise?
>
> I did consult with the lecturer at university that knows most about stats,
> and he advised me:
>
> "Pseudo replication is really about a lack of independence between
> measurements, So you need to work backwards and see where you are building
> in a known lack of independence.  And where that is the case you need to use
> means of all the values."
>
>
> And I have done this and came to the conclusion I mentioned as to where I
> thought Pseudoreplicaton was comming from, however, I do not know about the
> one other 'potential' source as it really is for me at least, a grey area.
> I've consulted a few forums that deal with the theory more and await any
> response. Until then I'll have to try and get as many opinions on it as
> possible.
>
> -Ben W.
>
>> On Fri, Feb 25, 2011 at 9:08 AM, Ben Ward<benjamin.ward at bathspa.org>
>> wrote:
>>>
>>>  Hi, As part of my dissertation, I'm going to be doing an Anova,
>>> comparing
>>>  the "dead zone" diameters on plates of microbial growth with little
>>> paper
>>>  disks "loaded" with antimicrobial, a clear zone appears where death
>>> occurs,
>>>  the size depending on the strength and succeptibility. So it's basically
>>> 4
>>>  different treatments, and I'm comparing the diameters (in mm) of
>>> circles.
>>>  I'm concerned however, about Pseudoreplication and how to deal with it
>>> in R,
>>>  (I thought of using the Error() term.
>>>
>>>  I have four levels of one factor(called "Treatment"): NE.Dettol,
>>> EV.Dettol,
>>>  NE.Garlic, EV.Garlic.   ("NE.Dettol" is E.coli not evolved to dettol,
>>>  exposed to dettol to get "dead zones". And the same for NE.Garlic, but
>>> with
>>>  garlic, not dettol. "EV.Dettol" is E.coli that has been evolved against
>>>  dettol, and then tested afterwards against dettol to get the "dead
>>> zones".
>>>  Same applies for "EV.Garlic" but with garlic).  You see from the four
>>> levels
>>>  (or treatments) there are two chemicals involved. So my first concern is
>>>  whether they should be analysed using two seperate ANOVA's.
>>>
>>>  NE.Dettol and NE.Garlic are both the same organism - a lab stock E.coli,
>>>  just exposed to two different chemicals.
>>>  EV.Dettol and EV.Garlic, are in principle, likely to be two different
>>> forms
>>>  of the organism after the many experimental doses of their respective
>>>  chemical.
>>>
>>>  For NE.Garlic and NE.Dettol I have 5, what I've called "Lineages",
>>> basically
>>>  seperate bottles of them (10 in total).
>>>  Then I have 5 Bottles (Lineages) of EV.Dettol, and 5 of EV.Garlic. -
>>> This
>>>  was done because there was the possiblity that, whilst I'm expecting
>>> them
>>>  all to respond in a similar manner, there are many evolutionary paths to
>>> the
>>>  same result, and previous research and reading shows that occasionally
>>> one
>>>  or two react differently to the rest through random chance.
>>>  The point I observed above ("NE.Dettol and NE.Garlic are both the same
>>>  organism...") is also applicable to the 5 bottles: The 5 bottles each of
>>>  NE.Garlic and NE.Dettol are supposed to be all the same organism - from
>>> a
>>>  stock one kept in store in the lab.
>>>  There is potential though for the 5 of EV.Garlic, to be different from
>>> one
>>>  another, and potential for the 5 EV.Dettol to be different from one
>>> another.
>>>
>>>  The Lineage (bottle) is also a factor then, with 5 levels (1,2,3,4,5).
>>>  Because they may be different.
>>>
>>>  To get the measurements of the diamter of the zones. I take out a small
>>>  amount from a tube and spread it on a plate, then take three paper
>>> disks,
>>>  soaked in their respective chemical, either Dettol or Garlic. and press
>>> them
>>>  and and incubate them.
>>>  Then when the zones have appeared after a day or 2. I take 4 diameter
>>>  measurements from each zone, across the zone at different angles, to
>>> take
>>>  account for the fact, that there may be a weird shape, or not quite
>>>  circular.
>>>
>>>  I'm concerned about pseudoreplication, such as the multiple readings
>>> from
>>>  one disk, and the 5 lineages - which might be different from one another
>>> in
>>>  each of the Two "EV." treatments, but not with "NE." treatments.
>>>
>>>  I read that I can remove pseudoreplication from  the multiple readings
>>> from
>>>  each disk, by using the 4 readings on each disk, to produce a mean for
>>> the
>>>  disks, and analyse those means - Exerciseing caution where there are
>>> extreme
>>>  values. I think the 3 disks for each lineage themselves are not
>>>  pseudoreplication, because they are genuinley 3 disks on a plate: the
>>> "Disk
>>>  Diffusion Test" replicated 3 times - but the multiple readings from one
>>> disk
>>>  if eel, is pseudoreplication. I've also read about including Error()
>>> terms
>>>  in a formula.
>>>
>>>  I'm unsure of the two NE. Treatments comming from the same culture does
>>> not
>>>  introduce pseudoreplications at Treatment Factor Level, because of the
>>> two
>>>  different antimicrobials used have two different effects.
>>>
>>>  I was hoping for a more expert opinion on whether I have identified
>>>  pseudoreplication correctly or if there is indeed pseudoreplication in
>>> the 5
>>>  Lineages or anywhere else I haven't seen. And how best this is dealt
>>> with in
>>>  R. At the minute my solution to the multiple readings from one disk is
>>> to
>>>  simply make a new factor, with the means on and do Anova from that, or
>>> even
>>>  take the means before I even load the dataset into R. I'm wondering if
>>> an
>>>  Error() term would be correct.
>>>
>>>  Thanks,
>>>  Ben W.
>>>
>>>  ______________________________________________
>>>  R-help at r-project.org mailing list
>>>  https://stat.ethz.ch/mailman/listinfo/r-help
>>>  PLEASE do read the posting guide
>>> http://www.R-project.org/posting-guide.html
>>>  and provide commented, minimal, self-contained, reproducible code.
>>>
>>
>>
>
>



-- 
Bert Gunter
Genentech Nonclinical Biostatistics




More information about the R-sig-mixed-models mailing list