R help - Oct 2015 - leveneTest between 2-levels of 3-level factor in R?

Still feeling my way around using R...

What I have done so far:

I have a data.frame 'results' with response 'Fail', and three
factors
'PREP', 'CLEAN' & 'ADHES'. ADHES has 3 levels: Crest
Cryst Poly

I calculated the variances:

sigma..k=tapply(Fail,ADHES,var)
print(sqrt(sigma..k)):

Crest    Cryst     Poly 17.56668 41.64679 39.42669

then used leveneTest to test for constance of variance:

print(leveneTest(Fail~ADHES))

Levene's Test for Homogeneity of Variance (center = median)
      Df F value  Pr(>F)
group  2   3.929 0.02588 *
      51

The Question:
Now I want to use Levene's to test for constance of variance between only
the Cryst & Poly levels of the factor ADHES, but I cant work out the syntax
to do this in R.

Could anyone help, please?
?

	[[alternative HTML version deleted]]

On 20/10/15 01:02, Dr. Leigh S. Sutherland wrote:
> Still feeling my way around using R...
>
> What I have done so far:
>
> I have a data.frame 'results' with response 'Fail', and
three factors
> 'PREP', 'CLEAN' & 'ADHES'. ADHES has 3 levels:
Crest Cryst Poly
>
> I calculated the variances:
>
> sigma..k=tapply(Fail,ADHES,var)
> print(sqrt(sigma..k)):
>
> Crest    Cryst     Poly 17.56668 41.64679 39.42669
>
> then used leveneTest to test for constance of variance:
>
> print(leveneTest(Fail~ADHES))
How does leveneTest() find the variables "Fail" and "ADHES"
given that
you do not provide it with a "data" argument?
> Levene's Test for Homogeneity of Variance (center = median)
>        Df F value  Pr(>F)
> group  2   3.929 0.02588 *
>        51
>
> The Question:
> Now I want to use Levene's to test for constance of variance between
only
> the Cryst & Poly levels of the factor ADHES, but I cant work out the
syntax
> to do this in R.
>
> Could anyone help, please?
I am not at all sure that I understand your somewhat incoherently 
expressed question, but I *think* that what you want to do might be done 
by setting

    ok <- with(results,ADHES%in%c("Cryst","Poly"))
    leveneTest(Fail ~ ADHES, data=results[ok,])

It's hard to say, since you do not provide a reproducible example.

cheers,

Rolf Turner

-- 
Technical Editor ANZJS
Department of Statistics
University of Auckland
Phone: +64-9-373-7599 ext. 88276

Thanks Rolf,

Sorry, I was trying to be coherent, but obviously failed abysmally!
I am new to this list, I will give a reproducible example next time, and
more of the relevant script.

I managed to do what I wanted to with subset:

leveneTest(subset(results,ADHES == 'Poly'|ADHES == 'Cryst')[,5],
                 subset(results,ADHES == 'Poly'|ADHES ==
'Cryst')[,3])

which gives the same results as your suggestion, which I can now look into
how it works

many thanks.

Leigh

On 19 October 2015 at 22:13, Rolf Turner <r.turner at auckland.ac.nz>
wrote:
> On 20/10/15 01:02, Dr. Leigh S. Sutherland wrote:
>
>> Still feeling my way around using R...
>>
>> What I have done so far:
>>
>> I have a data.frame 'results' with response 'Fail', and
three factors
>> 'PREP', 'CLEAN' & 'ADHES'. ADHES has 3
levels: Crest Cryst Poly
>>
>> I calculated the variances:
>>
>> sigma..k=tapply(Fail,ADHES,var)
>> print(sqrt(sigma..k)):
>>
>> Crest    Cryst     Poly 17.56668 41.64679 39.42669
>>
>> then used leveneTest to test for constance of variance:
>>
>> print(leveneTest(Fail~ADHES))
>>
>
> How does leveneTest() find the variables "Fail" and
"ADHES" given that you
> do not provide it with a "data" argument?
>
> Levene's Test for Homogeneity of Variance (center = median)
>>        Df F value  Pr(>F)
>> group  2   3.929 0.02588 *
>>        51
>>
>> The Question:
>> Now I want to use Levene's to test for constance of variance
between only
>> the Cryst & Poly levels of the factor ADHES, but I cant work out
the
>> syntax
>> to do this in R.
>>
>> Could anyone help, please?
>>
>
> I am not at all sure that I understand your somewhat incoherently
> expressed question, but I *think* that what you want to do might be done by
> setting
>
>    ok <- with(results,ADHES%in%c("Cryst","Poly"))
>    leveneTest(Fail ~ ADHES, data=results[ok,])
>
> It's hard to say, since you do not provide a reproducible example.
>
> cheers,
>
> Rolf Turner
>
> --
> Technical Editor ANZJS
> Department of Statistics
> University of Auckland
> Phone: +64-9-373-7599 ext. 88276
>


-- 
Leigh Sutherland

Centre for Marine Technology and Ocean Engineering (CENTEC)
Instituto Superior T?cnico
Av. Rovisco Pais
1049-001 LISBOA
PORTUGAL

Tel: +351 218 417 947

	[[alternative HTML version deleted]]