ISO/TS TECHNICAL SPECIFICATION 20281 First edition 2006-04-01 Water quality Guidance on statistical interpretation of ecotoxicity data Qualite de I'eau - Lignes directrices relatives a I'interpretation statistique de données écotoxicologiques Reference number ISO/TS 20281:2006(E) @ISO2006 by IHS under lic ted without license from IHS Not for Resale ISO/TS 20281:2006(E) PDF disclaimer This PDF file may contain embedded typefaces. In accordance with Adobe's licensing policy, this file may be printed or viewed but shall not be edited unless the typefaces which are embedded are licensed to and installed on the computer performing the editing. In accepts no liability in this area. Adobe is a trademark of Adobe Systems Incorporated. Details of the software products used to create this PDF fle can be found in the General Info relative to the file; the PDF-creation parameters were optimized for printing. Every care has been taken to ensure that the file is suitable for use by IsO member bodies. In the unlikely event that a problem relating to it is found, please inform the Central Secretariat at the address given below. ?ISO2006 All rights reserved. Unless otherwise specified, no part of this publication may be reproduced or utilized in any form or by any means, IsO's member body in the country of the requester. ISO copyright office Case postale 56. CH-1211 Geneva 20 Tel. + 41 22 749 01 11 Fax + 41 22 749 09 47 E-mail [email protected] Web www.iso.org Published in Switzerland @ Iso 2006 - Allrights reserved ed without license from IHS Not for Resale ISO/TS 20281:2006(E) Contents Page Foreword. .xii Introduction .xili 1 Scope 2 Normative references . 3 Terms and definitions 4 General statistical principles.... 4.1 Different statistical approaches . 4.1.1 .8 4.1.2 Hypothesis.testing methods ......... 4.1.3 Concentration-response modelling methods ... ..10 4.1.4 Biology-based methods... 4.2 Experimental design issues ..... 11 4.2.1 4.2.2 NOEC or ECx: Implications for design... 12 4.2.3 .Randomization....... 12 4.2.4 .13 4.2.5 Multiple controls included in the experimental design..... 4.3 Process of data analysis... 14 4.3.1 General... 4.3.2 Data inspection and outliers..... 4.3.3 Datainspectionandassumptions 15 4.3.3.1 Scatter... .15 4.3.3.2 Heterogeneous variances and distribution .... .15 4.3.3.3 Heterogeneous variances and true variation in response.... 4.3.3.4 Consequences for the analysis .... .16 4.3.4 Transformation of data.... 4.3.5 Parametric and non-parametric methods 4.3.5.1 General.. .17 4.3.5.2 Parametric methods.. 4.3.5.3 Generalized linear models (GLMs).. .18 4.3.5.4 Non-parametric methods.... 4.3.5.5 Howto choose?.. 18 4.3.6 Pre-treatment of data.. 4.3.7 Model fitting..... .19 4.3.8 Model checking ... 4.3.8.1 Analysis of residuals .. 20 4.3.8.2 Validation of fitted dose-response model 21 4.3.9 Reporting the results.... 21 5 Hypothesis testing..... 5.1 Introduction.. 21 5.1.1 General. 21 5.1.2 NOEC: What it is, and what it is not.... 25 5.1.3 Hypothesis used to determine NOEC 5.1.3.1 Understanding the question to be answered .... 25 5.1.3.2 One-sided hypothesis.... 26 5.1.3.3 5.1.3.4 Trend or pair-wise test.... 5.1.4 Comparisons of single-step (pair-wise comparisons) or step-down trend tests to 28 ili Copyrght International Organizaionsadardizalonghts reserved ted without license from IHS Not for Resale ISO/TS 20281:2006(E) 5.1.4.1 5.1.4.2 Single-step procedures. .28 5.1.4.3 Step-down procedures.... 29 5.1.4.4 Deciding between the two approaches, .30 5.1.5 Dose metric in trend tests ... .31 5.1.6 Role of power in toxicity experiments . 31 5.1.7 Experimental design . 32 5.1.8 Treatment of covariates and other adjustments to analysis ................... 5.2 Quantal data (e.g. mortality, survival).... 5.2.1 Hypothesis testing with quantal data to determine NOEC values .. 34 5.2.2 Parametric versus non-parametric tests .... 35 5.2.2.1 5.2.2.2 Single-step procedures... ..36 5.2.2.3 Step-down procedures... 36 5.2.2.3.1 Choice