1. Create and evaluate a candidate set of at least five general linear models predicting species diversity using combinations of the other continuous variables (evenness, abundance, richness, and dominance)-in isolation (simple linear regression) or as additive or interactive effects (multiple linear regression). Be creative when making your candidate model set but do not include correlated variables in the same model. Include a global (fully saturated) and null (intercept-only) model in your candidate set. 2. In the below table: Number each model, include the predictor(s) used in the model, the number of parameters, how each model ranks against one another using AIC corrected for small sample size, the relative likelihood a given model is the best, and a measure of model fit. Which of your models is the "best" (most supported)? What led you to this conclusion? Evaluate whether this model's predictor(s) had any meaningful explanatory power and interpret the parameter estimate(s). Did you need to use model averaging? Why or why not? (3 points) Model # Parameters K AAICO W Log-likelihood Create a scatterplot with a regression line based on your top-ranked model. Include appropriately labeled axes and provide a caption below the figure describing the displayed relationship. (1 point) In the below table: From each of your models, provide the parameter estimate(s), measures of uncertainty around the estimate(s) (standard error [SE], lower [LCL] and upper [UCL] confidence limits), and test statistics (t values and p-values). You will need to add additional rows to this table if you're evaluating more than one parameter per model. (2 points) Model # Parameter Estimate SE LCL UCL It value p-value How did you evaluate the assumptions of your models? Were they met? If not, what could you do to address such concerns? Include a diagnostic plot from your global model to support your answer. (2 points) In a few sentences, compare the methods you used to make model-based inferences. Do your conclusions about species diversity in these tidal pools differ based on the statistical paradigms used? (2 points) Data set 1 Diversity Evenness Dominanci Abundance Richness 0.47 0.98 3.24 16 0.8 3.75 16 0.4 0.84 2.43 17 3 0.84 0.93 8.55 19 8 0.8 0.94 7.77 19 2 0.54 0.77 3.05 19 S 0.72 0.92 5.77 25 6 0.77 0.85 27 8 0.91 0.96 9.95 28 0.76 0.9 5.64 29 0.81 0.84 0.62 0.79 3.73 36 6 0.62 0.69 3.32 38 0.58 0.92 78 39 0.72 0.93 5 40 6 0.53 2.38 0.36 0.75 2.12 42 1.07 0.91 11.13 15 1.06 0.97 12.97 48 12 0.71 0.84 4.58 49 7 1.01 0.94 11.14 52 12 0.74 0.88 4.83 59 7 0.77 0.77 5.02 62 10 0.86 0.86 6.08 68 10 1.27 0.96 21.35 75 21 12 0.92 15.48 77 20 0.95 0.95 8.54 78 10 0.64 0.67 3.33 100 9 0.73 0.87 4.6 131 7 Diversity Evenness Dominance Abundance Richness 0.47 0.98 3.24 16 3 0.63 0.8 3.75 16 6 0.4 0.84 2.43 17 3 0.84 0.93 8.55 19 8 0.8 0.94 7.77 1999 7 0.54 0.77 3.05 19 5 0.72 0.92 5.77 25 25 6 0.77 0.85 5.4 27 27 8 0.91 0.95 9.95 28 9 0.76 0.9 5.64 29 7 0.81 0.84 5.7 32 9 0.62 0.79 3.73 36 6 0.8 0.89 6.06 37 8 0.62 0.69 3.32 38 8 0.88 0.92 7.8 39 9 0.72 0.93 5 40 40 6 0.53 0.68 2.38 41 0.36 0.75 2.12 42 1.07 0.91 11.13 46 1.05 0.97 12.97 48 6 14 42 485 3 15 12 0.71 0.84 4.58 1.01 0.94 11.14 425 49 7 52 12 0.74 0.88 4.83 59 7 0.77 0.77 5.02 62 62 10 10 0.86 0.86 6.03 68 1.27 0.96 21.35 75 85 10 21 1.2 0.92 15.48 77 0.95 0.95 8.94 78 78 20 20 10 10 0.64 0.67 3.33 100 9 0.73 0.87 4.6 131 7