ALBERT

Note: Contents: Preface. - 1. Introduction and data manipulation. - 1.1. Why ordination?. - 1.2. Terminology. - 1.3. Types of analyses. - 1.4. Response variables. - 1.5. Explanatory variables. - 1.6. Handling missing values in data. - 1.7. Importing data from spreadsheets - WCanoImp program. - 1.8. Transformation of species data. - 1.9. Transformation of explanatory variables. - 2. Experimental design. - 2.1. Completely randomized design. - 2.2. Randomized complete blocks. - 2.3. Latin square design. - 2.4. Most frequent errors - pseudoreplications. - 2.5. Combining more than one factor. - 2.6. Following the development of objects in time - repeated observations. - 2.7. Experimental and observational data. - 3. Basics of gradient analysis. - 3.1. Techniques of gradient analysis. - 3.2. Models of species response to environmental gradients. - 3.3. Estimating species optima by the weighted averaging method. - 3.4. Calibration. - 3.5. Ordination. - 3.6. Constrained ordination. - 3.7. Basic ordination techniques. - 3.8. Ordination diagrams. - 3.9. Two approaches. - 3.10. Testing significance of the relation with environmental variables. - 3.11. Monte Carlo permutation tests for the significance of regression. - 4. Using the Canoco for Windows 4.5 package. - 4.1. Overview of the package. - 4.2. Typical flow-chart of data analysis with Canoco for Windows. - 4.3. Deciding on the ordination method: unimodal or linear?. - 4.4. PCA or RDA ordination: centring and standardizing. - 4.5. DCA ordination: detrending. - 4.6. Scaling of ordination scores. - 4.7. Running CanoDraw for Windows 4.0. - 4.8. New analyses providing new views of our data sets. - 5. Constrained ordination and permutation tests. - 5.1. Linear multiple regression model. - 5.2. Constrained ordination model. - 5.3. RDA: constrained PCA. - 5.4. Monte Carlo permutation test: an introduction. - 5.5. Null hypothesis model. - 5.6. Test statistics. - 5.7. Spatial and temporal constraints. - 5.8. Split-plot constraints. - 5.9. Stepwise selection of the model. - 5.10. Variance partitioning procedure. - 6. Similarity measures. - 6.1. Similarity measures for presence-absence data. - 6.2. Similarity measures for quantitative data. - 6.3. Similarity of samples versus similarity of communities. - 6.4. Principal coordinates analysis. - 6.5. Non-metric multidimensional scaling. - 6.6. Constrained principal coordinates analysis (db-RDA). - 6.7. Mantel test. - 7. Classification methods. - 7.1. Sample data set. - 7.2. Non-hierarchical classification (K-means clustering). - 7.3. Hierarchical classifications. - 7.4. TWINSPAN. - 8. Regression methods . - 8.1. Regression models in general. - 8.2. General linear model: terms. - 8.3. Generalized linear models (GLM). - 8.4. Loess smoother. - 8.5. Generalized additive models (GAM). - 8.6. Classification and regression trees. - 8.7. Modelling species response curves with CanoDraw. - 9. Advanced use of ordination. - 9.1. Testing the significance of individual constrained ordination axes. - 9.2. Hierarchical analysis of community variation. - 9.3. Principal response curves (PRC) method. - 9.4. Linear discriminant analysis. - 10. Visualizing multivariate data. - 10.1. What we can infer from ordination diagrams: linear methods. - 10.2. What we can infer from ordination diagrams: unimodal methods. - 10.3. Visualizing ordination results with statistical models. - 10.4. Ordination diagnostics. - 10.5. t-value biplot interpretation. - 11. Case study 1: Variation in forest bird assemblages. - 11.1. Data manipulation. - 11.2. Deciding between linear and unimodal ordination. - 11.3. Indirect analysis: portraying variation in bird community. - 11.4. Direct gradient analysis: effect of altitude. - 11.5.Direct gradient analysis: additional effect of other habitat characteristics. - 12. Case study 2: Search for community composition patterns and their environmental correlates: vegetation of spring meadows. - 12.1. The unconstrained ordination. - 12.2. Constrained ordinations. - 12.3. Classification. - 12.4. Suggestions for additional analyses. - 13. Case study 3: Separating the effects of explanatory variables. - 13.1. Introduction. - 13.2. Data. - 13.3. Data analysis. - 14. Case study 4: Evaluation of experiments in randomized complete blocks. - 14.1. Introduction. - 14.2. Data. - 14.3. Data analysis. - 15. Case study 5: Analysis of repeated observations of species composition from a factorial experiment. - 15.1. Introduction. - 15.2. Experimental design. - 15.3. Sampling. - 15.4. Data analysis. - 15.5. Univariate analyses. - 15.6. Constrained ordinations. - 15.7. Further use of ordination results. - 15.8. Principal response curves. - 16. Case study 6: Hierarchical analysis of crayfish community variation. - 16.1. Data and design. - 16.2. Differences among sampling locations. - 16.3. Hierarchical decomposition'of community variation. - 17. Case study 7: Differentiating two species and their hybrids with discriminant analysis. - 17.1. Data. - 17.2. Stepwise selection of discriminating variables. - 17.3. Adjusting the discriminating variables. - 17.4. Displaying results. - Appendix A: Sample datasets and projects. - Appendix B: Vocabulary. - Appendix C: Overview of available software. - References. - Index.