Reports for particular tests give conservatively estimated minimum g factor loadings. These are obtained by taking the square root of the weighted average of that test's correlations with other tests, and estimate the correlation of that test with g, the common factor between tests for mental abilities.
This estimation is called conservative because the correlations found with high-range tests tend to be attenuated (depressed) by causes like restriction of range and incomplete reporting by candidates. Restriction of range, as in considering only the high range as opposed the full range of intelligence, results in lower correlations as it eliminates much of the possible variance.