Figure 8
The vertical axis is labeled with four categories and arranged from top to bottom as “Native female,” “C I female,” “Native male,” and “C I male.” The horizontal axis represents a range of values from 0 to 1 with an interval of 0.2. Each group is represented by colored dots, with error bars indicating confidence intervals. The dots are color-coded to represent different combinations of G P A and education type: blue represents “Low G P A, G E N,” green represents “High G P A, G E N,” red represents “Low G P A, V E T,” and orange represents “High G P A, V E T.” The dots are horizontally arranged for each group, with a red dot on the left and a green dot on the right. For the “Native female” group, the red dot lies around 0.1, the orange dot lies around 0.4, the blue dot lies around 0.5, and the green dot lies around 0.9. For the “C I female” group, the red dot lies just after 0.1, the orange dot lies around 0.3, the blue dot lies just before 0.7, and the green dot lies around 0.9. For the “Native male” group, the red dot lies just before 0.1, the orange dot lies just after 0.2, the blue dot lies around 0.55, and the green dot lies around 0.9. For the “C I male” group, the red dot lies just after 0.1, the orange dot lies around 0.5, the blue dot lies just after 0.6, and the green dot lies just before 1. The error bars for C I female and C I male are larger than for native female and native male error bars. Note: All numerical data values are approximated.Predicted probabilities of membership in higher-education clusters (clusters 1 and 5). Source: Figure created by authors