Estimated Marginal Means: Year of Construction

Figure 1. Estimated marginal means for Year of Construction

This table displays the model-estimated marginal means, standard errors, and confidence intervals for the linear predictor of Number of damage incidents at the factor levels of Year of Construction. This table is useful for exploring the differences between the levels of this factor. In this example, the linear predictor ranges from a low of -6.410 for ships constructed between 1960-64 to a high of -5.591 for ships constructed between 1970-74. To see whether the values in this table represent real differences or are likely due to chance variation, look at the test results.

Figure 2. Pairwise comparisons for Year of Construction

The pairwise comparisons table displays the differences between each pair of ship construction periods and tests whether each difference is due to chance variation.

The parameter estimates table shows that ships constructed between 1965-69 and 1970-74 have higher damage rates than those built between 1960-64. These differences are still significant after the p values are adjusted by the sequential Sidak method.

The pairwise comparisons reveal no other significant differences.

Figure 3. Overall test results for Year of Construction

The overall test table reports the results of a test of all of the contrasts in the pairwise comparisons table. Since there are no interaction effects involving Year of Construction, this test is equivalent to the model effects test for Year of Construction.