All Tables

Table 2: Analysis of correlations between p₀ and the indices described in Sect. 2.3 as well as correlations between the indices themselves. Three samples of BAL QSOs are considered. n is the number of objects considered in the tests. The last column indicates whether the correlation is positive or negative (identical for the three BAL QSO samples). Boldfaced entries are the least likely to be due to random chance.
Table 3: Results of the Principal Component Analysis. The first three principal components out of a total of 8 are given in order of their contribution to the total variance. The first row gives the variances of the data along the direction of the corresponding principal component. The sum of all the variances is equal to 8, the number of variables considered in the analysis. The variance of each principal component is given as "proportion'' in the second row and as "the cumulative proportion'' in the third one. The rest of the table gives the correlation coefficients between the original variables and the principal components. Significant correlations are boldfaced, i.e. those ones with coefficients whose absolute values are > $1/ \sqrt 8$ = 0.3536.
Table 4: The spectropolarimetric indices. The measurements were done using the spectropolarimetric data illustrated in Figs. 3 and 4 and the polarization spectra of Ogle et al. (1999). The continuum polarization, p₀, and its associated error, $\sigma _{\rm p}$ , are taken from Table 1. p₀, $\sigma _{\rm p}$ and $p_{{\rm max}}$ are in % while the other quantities are unitless.
Table 5: Results of correlation tests between spectropolarimetric indices, p₀ and three indices derived from optical spectra. $P_{\tau }$ and $P_{r_{\rm s}}$ give the probabilities that the correlation between the two indices are due to chance. n is the number of points considered for the correlation. A correlation is emphasized if $P \leq 0.05$ . The last column gives the sign of the correlation.