[R-sig-eco] Ecological datasets for teaching statistics

Fri Jun 19 13:15:01 CEST 2020

Dear Edward,
You are right with your observation. I analyzed the data based on 999 nboots without pairwise comparison and it worked fine. However,  I am not able to generate any output for the pairwise comparison at this high nboot (somehow the R could not output the results). I, therefore, reduced it to 199 to achieve these results.
Thanks for your help.
Xorla

From: Eduard Szöcs [mailto:szoecs using uni-landau.de]
Sent: Friday, June 19, 2020 1:01 PM
Cc: Kanfra, Xorla; r-sig-ecology using r-project.org
Subject: Re: [R-sig-eco] Ecological datasets for teaching statistics

> but the p-values generated were
all the same for each pair of comparison. Is this output normal?
With 199 bootstrap samples the smallest p you can get is 1/199=0,00502512...
Hope this helps,
Eduard
Am 19. Juni 2020, um 11:33, Thierry Onkelinx <thierry.onkelinx using inbo.be<mailto:thierry.onkelinx using inbo.be>> schrieb:

Try to send plain text mail without attachments.

ir. Thierry Onkelinx
Statisticus / Statistician

Vlaamse Overheid / Government of Flanders
INSTITUUT VOOR NATUUR- EN BOSONDERZOEK / RESEARCH INSTITUTE FOR NATURE AND
FOREST
Team Biometrie & Kwaliteitszorg / Team Biometrics & Quality Assurance
thierry.onkelinx using inbo.be
Havenlaan 88 bus 73, 1000 Brussel
www.inbo.be<http://www.inbo.be>

///////////////////////////////////////////////////////////////////////////////////////////
To call in the statistician after the experiment is done may be no more
than asking him to perform a post-mortem examination: he may be able to say
what the experiment died of. ~ Sir Ronald Aylmer Fisher
The plural of anecdote is not data. ~ Roger Brinner
The combination of some data and an aching desire for an answer does not
ensure that a reasonable answer can be extracted from a given body of data.
~ John Tukey
///////////////////////////////////////////////////////////////////////////////////////////

<https://www.inbo.be>

Op vr 19 jun. 2020 om 11:23 schreef Kanfra, Xorla <
xorla.kanfra using julius-kuehn.de>:

 Dear Roman,

 Thank you for the suggestion. I tried a few minutes ago but go this
 response  ‘’The message's content type was not explicitly allowed’’.

 Please can you advise?

 Kind regards

 Xorla

 *From:* Roman Luštrik [mailto:roman.lustrik using gmail.com]
 *Sent:* Friday, June 19, 2020 8:40 AM
 *To:* Kanfra, Xorla
 *Cc:* Thierry Onkelinx; Manuel Spínola; r-sig-ecology using r-project.org
 *Subject:* Re: [R-sig-eco] Ecological datasets for teaching statistics

 Your question is likely to get lost in the debris. Try posting it as a new
 topic.

 Cheers,

 Roman

 On Fri, Jun 19, 2020 at 7:40 AM Kanfra, Xorla <
 xorla.kanfra using julius-kuehn.de> wrote:

 Dear All,
 Sorry to use this thread for communication. I subscribed to this forum
 just a few days ago but couldn’t manage to communicate across the board.
 Please pardon me. If any of you can help me get around a statistical
 challenge
 I have just started using manyglm to analyze nematode with associated
 microbial community data and I am really happy with how robust the
 algorithm works. I have a question about the adjusted p- values. I run a
 pairwise comparison for one of my factors but the p-values generated were
 all the same for each pair of comparison. Is this output normal? besides,
 are the p-values adjusted when running the pairwise comparison? Which
 method does it use? ex Tukey, FDR, Holm, BH, etc. I am a bit confused as I
 do not understand really what is going on behind the scripts.
 Below is the script I used.
 Thanks and hoping to hear from you soon
 Xorla Kanfra

 mod_pairwise <-anova.manyglm(otutable.glm, nBoot=199, cor.type = "I", test
 = "LR", p.uni="adjusted",pairwise.comp = ~envdata$soil)

 Analysis of Deviance Table

 Model: manyglm(formula = otutable.fv, family = "negative.binomial")

 Multivariate test:
                               Res.Df Df.diff   Dev Pr(>Dev)
 (Intercept)                       88
 envdata$soil                      81       7 42968    0.005 **
 envdata$response                  80       1  4824    0.005 **
 envdata$soil:envdata$response     73       7 16066    0.005 **
 ---
 Signif. codes:  0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1 ‘ ’ 1

 Pairwise comparison results:
                                    Observed statistic Free Stepdown
 Adjusted P-Value
 envdata$soil:H vs envdata$soil:HH                6945
     0.005 **
 envdata$soil:E vs envdata$soil:HH                6881
     0.005 **
 envdata$soil:HH vs envdata$soil:R                6471
     0.005 **
 envdata$soil:HH vs envdata$soil:M                6458
     0.005 **
 envdata$soil:E vs envdata$soil:EE                6301
     0.005 **
 envdata$soil:HH vs envdata$soil:K                6032
     0.005 **
 envdata$soil:R vs envdata$soil:RR                5993
     0.005 **
 envdata$soil:EE vs envdata$soil:RR               5855
     0.005 **
 envdata$soil:EE vs envdata$soil:M                5835
     0.005 **

 -----Original Message-----
 From: R-sig-ecology [mailto:r-sig-ecology-bounces using r-project.org] On
 Behalf Of Thierry Onkelinx
 Sent: Thursday, June 18, 2020 9:07 PM
 To: Manuel Spínola
 Cc: r-sig-ecology using r-project.org
 Subject: Re: [R-sig-eco] Ecological datasets for teaching statistics

 Dear Manuel,

 Our institute has published 54 datasets under an open data licence at GBIF
 (

 https://www.gbif.org/dataset/search?publishing_org=1cd669d0-80ea-11de-a9d0-f1765f95f18b
 )

 You can look for local data on GBIF too.

 Best regards,

 ir. Thierry Onkelinx
 Statisticus / Statistician

 Vlaamse Overheid / Government of Flanders
 INSTITUUT VOOR NATUUR- EN BOSONDERZOEK / RESEARCH INSTITUTE FOR NATURE AND
 FOREST
 Team Biometrie & Kwaliteitszorg / Team Biometrics & Quality Assurance
 thierry.onkelinx using inbo.be
 Havenlaan 88 bus 73, 1000 Brussel
 www.inbo.be<http://www.inbo.be>

 ///////////////////////////////////////////////////////////////////////////////////////////
 To call in the statistician after the experiment is done may be no more
 than asking him to perform a post-mortem examination: he may be able to say
 what the experiment died of. ~ Sir Ronald Aylmer Fisher
 The plural of anecdote is not data. ~ Roger Brinner
 The combination of some data and an aching desire for an answer does not
 ensure that a reasonable answer can be extracted from a given body of data.
 ~ John Tukey

 ///////////////////////////////////////////////////////////////////////////////////////////

 <https://www.inbo.be>

 Op do 18 jun. 2020 om 19:58 schreef Manuel Spínola <mspinola10 using gmail.com>:

 Thank you very much Rich.

 Yes, you are right, is a very broad spectrum.

 I teach mainly to wildlife ecology students.

 I was using several datasets from different sources for homeworks and
 final projects, but some of the students were "sharing" the results of

 the

 assignments so I decided to assign different datasets to each student for
 their homeworks.  This means that I need several datasets for each
 assignment, and the data need to be similar in the structure, for

 example,

 logistic regression, the response variable needs to be binary, and so on.

 Manuel

 El jue., 18 jun. 2020 a las 11:41, Rich Shepard (<

 rshepard using appl-ecosys.com

)

 escribió:

 On Thu, 18 Jun 2020, Manuel Spínola wrote:

 I teach statistics to students in ecology and environmental sciences
 fields and I would like to know if you could point me in the right
 direction of sources of ecological/environmental datasets within and
 outside packages, especially for general/generalized linear models

 and

 multivariate statistics.

 Manuel,

 Ecology, and it's applied focus Environmental science, are very broad.

 I've

 been working with these data for several decades so I need to ask what
 types
 of data you want.

 I don't know what's available from Costa Rican agencies but I do know

 that

 in the US you can get geochemical, biological, hydrologidal, and other

 data

 from the US Geological Survay, Environmental Protection Agency (if

 they've

 not removed them), Department of Agriculture's Forest Service and

 Natural

 Resources Conservation Service.

 You can also look at StreamNet run by the Pacific States Marine

 Fisheries

 Council, The Army Corps of Engineers for hydraulic, flow, and sediment
 transport data.

 That's a start.

 Rich

________________________________

 R-sig-ecology mailing list
 R-sig-ecology using r-project.org
 https://stat.ethz.ch/mailman/listinfo/r-sig-ecology

 --
 *Manuel Spínola, Ph.D.*
 Instituto Internacional en Conservación y Manejo de Vida Silvestre
 Universidad Nacional
 Apartado 1350-3000
 Heredia
 COSTA RICA
 mspinola using una.cr <mspinola using una.ac.cr>
 mspinola10 using gmail.com
 Teléfono: (506) 8706 - 4662
 Personal website: Lobito de río <
 https://sites.google.com/site/lobitoderio/>
 Institutional website: ICOMVIS <http://www.icomvis.una.ac.cr/>

         [[alternative HTML version deleted]]

________________________________

 R-sig-ecology mailing list
 R-sig-ecology using r-project.org
 https://stat.ethz.ch/mailman/listinfo/r-sig-ecology

         [[alternative HTML version deleted]]

________________________________

 R-sig-ecology mailing list
 R-sig-ecology using r-project.org
 https://stat.ethz.ch/mailman/listinfo/r-sig-ecology

________________________________

 R-sig-ecology mailing list
 R-sig-ecology using r-project.org
 https://stat.ethz.ch/mailman/listinfo/r-sig-ecology

 --

 In God we trust, all others bring data.

 [[alternative HTML version deleted]]

________________________________

R-sig-ecology mailing list
R-sig-ecology using r-project.org
https://stat.ethz.ch/mailman/listinfo/r-sig-ecology

	[[alternative HTML version deleted]]