Testing correlation for different size

Testing Correlation for Different Size Populations Are two correlation coefficients (rl and r2 different based on a difference in the number of observations for each (N) ... 

The neurobehavioral test battery used in the 66-month Seychelles study was designed to assess multiple developmental domains (Davidson et al. 1998). The tests were considered to be sufficiently sensitive and accurate to detect neurotoxicity in the presence of a number of statistical covariates. On-site test administration reliability was assessed by an independent scorer, and mean interclass correlations for interscorer reliability were 0.96-0.97 (Davidson et al. 1998). The sample size was determined to be sufficient to detect a 5.7-point difference on any test with a mean (SD) of 100 (16) between low (0-3 ppm) and high (>12 ppm) hair mercury concentration groups for a 2-sided test (A = 0.05 at 80% power). 

The following sections detail existing in vitro/in vivo correlations for the major aerosol modalities and, where appropriate, comparisons are made with lung model predictions. Measures of particle diameter, particle size statistics, and aerosol test methods are also discussed. Aerosol test methodologies are included in the discussion because, as described above, sizing results are highly dependent upon the method and apparatus used. The correlations that have been developed and any predictions that can be made from them are therefore specific to the use of particular experimental methods, and it is important that the applicability of the different instruments/methods be understood. 

For the Ishigami test function, Figure 1 shows the results of eight different correlation ratio methods for sample sizes 100, 300, 1000, 3000, and 10000 analysing the influence of the first input parameter on the output. Some of these methods study different algorithmic approaches (Variance of the Conditional Expectation VCE, Expectation of the Conditional Variance ECV), others the influence of different sampling schemes (Simple Random Sampling SRS, Latin... 

Interpretation of the effects evoked by the explanatory variables on injury or mortality involves hypotheses about injury causation as well as the expected trend. Since impact speed is a dominant determinant for injury and fatality, any explanatory variable that is associated with impact speed could act as a surrogate for impact speed and thus as a confounder, i.e., be significant without having a causal relationship or mask the original effect size due to the association with impact speed. Potential associations were tested using Pearson and Spearman correlations for continuous variables and t-tests and Mann-Whitney-Tests for non-continuous (binary) variables. P-values refer to the hypotheses of a correlation (for continuous variables) or to differences between the two groups (for binary variables). 

---