, levels = 8:1)
>>> str(coloursf2)
>>>
>>> Duncan
>>>
>>> Duncan
>>> Duncan Mackay
>>> Department of Agronomy and Soil Science
>>> University of New England
>>> Armidale NSW 2351
>>> Email: home: mac...@northnet.com
Science
>> University of New England
>> Armidale NSW 2351
>> Email: home: mac...@northnet.com.au
>>
>>
>> ordered used in
>> used in MASS::polr and GEE for polytomous logistic regression
>>
>> -Original Message-
>> From: r-help-
> There are two reasons. First is that in the "day_of_week", the starting day is
> Friday so if you plot a graph, the left most column will start with Friday.
> You
> may like to start the column with Monday.
Plotting levels in a particular order by default does not usually require an
ordered f
Hi,
There are two reasons. First is that in the "day_of_week", the starting day
is Friday so if you plot a graph, the left most column will start with
Friday. You may like to start the column with Monday. The second reason is
that instead of having all these long factor names (Monday,...), the cod
au
>
>
> ordered used in
> used in MASS::polr and GEE for polytomous logistic regression
>
> -Original Message-
> From: r-help-boun...@r-project.org [mailto:r-help-boun...@r-project.org]
> On
> Behalf Of Bill
> Sent: Monday, 2 December 2013 21:24
> To: r-help@r-
r polytomous logistic regression
-Original Message-
From: r-help-boun...@r-project.org [mailto:r-help-boun...@r-project.org] On
Behalf Of Bill
Sent: Monday, 2 December 2013 21:24
To: r-help@r-project.org
Subject: [R] why change days of the week from a factor to an ordered factor?
I am readin
Thanks to all, still a bit of a mystery. It is for the graphing I guess but
not sure why. I don't think there is too much sense to saying that Tuesday
precedes Wednesday but there could, as some of you suggest, be
circumstances where this is useful.
On Mon, Dec 2, 2013 at 12:00 PM, Bert Gunter w
Did you not see Mark Leeds's post?
The OP apparently did not really mean R's "ordered factors" as
produced by the R ordered() constructor; rather, he meant "factors
with levels ordered differently than the default", for which Rich's
answer was apropos. Mine -- and now yours -- in which we wrongly
On 12/2/2013 9:35 AM, Bert Gunter wrote:
> Not true, Rich.
The point about alphabetical ordering explains why the author likely
explicitly set the levels for the factor, though.
As to why ordered factors, we may never know, but one possible
explanation is that at some point he was going to use
Bert,
the issue is the sort order of the levels. Time series graphs in the
alphabetical sort
order will be uninterpretable. I show the three sets of contrasts for
factors, factors
with specified levels, and ordered factors.
week <-
c("Sunday","Monday","Tuesday","Wednesday","Thursday","Friday","
On 02 Dec 2013, at 16:35 , Bert Gunter wrote:
> Not true, Rich.
>
>> z <-factor(letters[1:3],lev=letters[3:1])
>> sort(z)
> [1] c b a
> Levels: c b a
>
> What you say is true only for the **default** sort order.
>
> (Although maybe the code author didn't realize this either)
The coding is ce
Not true, Rich.
> z <-factor(letters[1:3],lev=letters[3:1])
> sort(z)
[1] c b a
Levels: c b a
What you say is true only for the **default** sort order.
(Although maybe the code author didn't realize this either)
-- Bert
On Mon, Dec 2, 2013 at 7:24 AM, Richard M. Heiberger wrote:
> If days of
If days of the week is not an Ordered Factor, then it will be sorted
alphabetically.
Fr Mo Sa Su Th Tu We
Rich
On Mon, Dec 2, 2013 at 6:24 AM, Bill wrote:
> I am reading the code below. It acts on a csv file called dodgers.csv with
> the following variables.
>
>
>> print(str(dodgers)) # check t
"BIll" :
(Sorry -- Doubt that this will be helpful, but I couln't resist)
"I don't understand why the author of the code decided to make the factor
days_of_week into an ordered factor. Anyone know why this should be done?"
A definitive answer would require either psychic abilities or asking
the
I am reading the code below. It acts on a csv file called dodgers.csv with
the following variables.
> print(str(dodgers)) # check the structure of the data frame
'data.frame': 81 obs. of 12 variables:
$ month : Factor w/ 7 levels "APR","AUG","JUL",..: 1 1 1 1 1 1 1 1 1
1 ...
$ day
15 matches
Mail list logo
