Premium
Current statistical issues in Weed Research
Author(s) -
ONOFRI A,
CARBONELL E A,
PIEPHO HP,
MORTIMER A M,
COUSENS R D
Publication year - 2010
Publication title -
weed research
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.693
H-Index - 74
eISSN - 1365-3180
pISSN - 0043-1737
DOI - 10.1111/j.1365-3180.2009.00758.x
Subject(s) - presentation (obstetrics) , scope (computer science) , computer science , selection (genetic algorithm) , set (abstract data type) , reading (process) , data science , management science , work (physics) , weed , statistical analysis , operations research , artificial intelligence , statistics , political science , mathematics , engineering , medicine , mechanical engineering , agronomy , biology , law , radiology , programming language
O nofri A, C arbonell EA, P iepho H‐P, M ortimer AM & C ousens RD (2010). Current statistical issues in Weed Research . Weed Research 50 , 5–24. Summary The correct design of experimental studies, the selection of the appropriate statistical analysis of data and the efficient presentation of results are key to the good conduct and communication of science. The last Guidance for the use and presentation of statistics in Weed Research was published in 1988. Since then, there have been developments in both the scope of research covered by the journal and in the statistical techniques available. This paper addresses the changes in statistics and provides a reference work that will aid researchers in the design and analysis of their work. It will also provide guidance for editors and reviewers. The paper is organised into sections, which will aid the selection of relevant paragraphs, as we recognise that particular approaches require particular statistical analysis. It also uses examples, questions and checklists, so that non‐specialists can work towards the correct approach. Statistics can be complex, so knowing when to seek specialist advice is important. The structure and layout of this contribution should help weed scientists, but it cannot provide a comprehensive guide to every technique. Therefore, we provide references to further reading. We would like to reinforce the idea that statistical methods are not a set of recipes whose mindless application is required by convention; each experiment or study may involve subtleties that these guidelines cannot cover. Nevertheless, we anticipate that this paper will help weed scientists in their initial designs for research, in the analysis of data and in the presentation of results for publication.