Analysis and modelling of the COLOSS international data set 2010-2011

Research output: Chapter in Book/Report/Conference proceedingOther chapter contribution

Abstract

Colony Losses have been calculated for Netherlands, Germany, Poland, Hungary, Austria, Denmark, Ireland and the UK. The following approach was used.
Initial analysis was conducted using examination of the data and descriptive statistics.
Some inconsistencies in the data were found, for further consideration. Using appropriate selection of cases, colony losses were calculated for the countries above individually and also overall, using overall proportion of loss, CDS type loss and loss due to queen problems, and also the mean/median of the individual loss rates per beekeeper. The proportions of beekeepers suffering any loss of any kind were also calculated. We both included and excluded Scotland as part of the UK. Differences between countries were found based on 95% confidence intervals for proportion of losses.
Some suggestions are made in relation to data quality control and data modelling. Logistic regression/binomial-logit generalised linear modelling may be used to model the chance of any loss, any CDS loss, or loss due to queen problems.
LanguageEnglish
Title of host publicationProceedings of the COLOSS WG1 workshop on modelling and standardization
Place of PublicationTersoal Netherlands
Number of pages1
Publication statusPublished - 2012

Fingerprint

modeling
loss
analysis
data quality
quality control
confidence interval
logistics

Keywords

  • analysis and modelling
  • COLOSS international data set
  • 2010-2011
  • colony losses
  • netherlands centre for bee research

Cite this

Gray, A. (2012). Analysis and modelling of the COLOSS international data set 2010-2011. In Proceedings of the COLOSS WG1 workshop on modelling and standardization Tersoal Netherlands.
Gray, Alison. / Analysis and modelling of the COLOSS international data set 2010-2011. Proceedings of the COLOSS WG1 workshop on modelling and standardization. Tersoal Netherlands, 2012.
@inbook{c2626f776a2c4c7c94404e4f270e29c3,
title = "Analysis and modelling of the COLOSS international data set 2010-2011",
abstract = "Colony Losses have been calculated for Netherlands, Germany, Poland, Hungary, Austria, Denmark, Ireland and the UK. The following approach was used.Initial analysis was conducted using examination of the data and descriptive statistics.Some inconsistencies in the data were found, for further consideration. Using appropriate selection of cases, colony losses were calculated for the countries above individually and also overall, using overall proportion of loss, CDS type loss and loss due to queen problems, and also the mean/median of the individual loss rates per beekeeper. The proportions of beekeepers suffering any loss of any kind were also calculated. We both included and excluded Scotland as part of the UK. Differences between countries were found based on 95{\%} confidence intervals for proportion of losses.Some suggestions are made in relation to data quality control and data modelling. Logistic regression/binomial-logit generalised linear modelling may be used to model the chance of any loss, any CDS loss, or loss due to queen problems.",
keywords = "analysis and modelling, COLOSS international data set, 2010-2011 , colony losses, netherlands centre for bee research",
author = "Alison Gray",
year = "2012",
language = "English",
booktitle = "Proceedings of the COLOSS WG1 workshop on modelling and standardization",

}

Gray, A 2012, Analysis and modelling of the COLOSS international data set 2010-2011. in Proceedings of the COLOSS WG1 workshop on modelling and standardization. Tersoal Netherlands.

Analysis and modelling of the COLOSS international data set 2010-2011. / Gray, Alison.

Proceedings of the COLOSS WG1 workshop on modelling and standardization. Tersoal Netherlands, 2012.

Research output: Chapter in Book/Report/Conference proceedingOther chapter contribution

TY - CHAP

T1 - Analysis and modelling of the COLOSS international data set 2010-2011

AU - Gray, Alison

PY - 2012

Y1 - 2012

N2 - Colony Losses have been calculated for Netherlands, Germany, Poland, Hungary, Austria, Denmark, Ireland and the UK. The following approach was used.Initial analysis was conducted using examination of the data and descriptive statistics.Some inconsistencies in the data were found, for further consideration. Using appropriate selection of cases, colony losses were calculated for the countries above individually and also overall, using overall proportion of loss, CDS type loss and loss due to queen problems, and also the mean/median of the individual loss rates per beekeeper. The proportions of beekeepers suffering any loss of any kind were also calculated. We both included and excluded Scotland as part of the UK. Differences between countries were found based on 95% confidence intervals for proportion of losses.Some suggestions are made in relation to data quality control and data modelling. Logistic regression/binomial-logit generalised linear modelling may be used to model the chance of any loss, any CDS loss, or loss due to queen problems.

AB - Colony Losses have been calculated for Netherlands, Germany, Poland, Hungary, Austria, Denmark, Ireland and the UK. The following approach was used.Initial analysis was conducted using examination of the data and descriptive statistics.Some inconsistencies in the data were found, for further consideration. Using appropriate selection of cases, colony losses were calculated for the countries above individually and also overall, using overall proportion of loss, CDS type loss and loss due to queen problems, and also the mean/median of the individual loss rates per beekeeper. The proportions of beekeepers suffering any loss of any kind were also calculated. We both included and excluded Scotland as part of the UK. Differences between countries were found based on 95% confidence intervals for proportion of losses.Some suggestions are made in relation to data quality control and data modelling. Logistic regression/binomial-logit generalised linear modelling may be used to model the chance of any loss, any CDS loss, or loss due to queen problems.

KW - analysis and modelling

KW - COLOSS international data set

KW - 2010-2011

KW - colony losses

KW - netherlands centre for bee research

UR - http://www.coloss.org/publications/jan-2012-workshop-proceedings-wg1-tersoal

M3 - Other chapter contribution

BT - Proceedings of the COLOSS WG1 workshop on modelling and standardization

CY - Tersoal Netherlands

ER -

Gray A. Analysis and modelling of the COLOSS international data set 2010-2011. In Proceedings of the COLOSS WG1 workshop on modelling and standardization. Tersoal Netherlands. 2012