deletions | additions       diff --git a/figures/Refiloe_s_fracRadiovsSep2/caption.tex b/figures/Refiloe_s_fracRadiovsSep2/caption.tex  index 8dc6cc2..da9d029 100644  --- a/figures/Refiloe_s_fracRadiovsSep2/caption.tex  +++ b/figures/Refiloe_s_fracRadiovsSep2/caption.tex 
...
Figure 11. the observed fractions of optically-selected quasar pairs that have radio counterparts are almost equal amongst our bins. The error bars largely overlap. The the x and y uncertainties were calculated as described in the statistical analysis section. The straight line at 0.10 shows that the typical fraction of radio detection is $\approx 10\%$ of SDSS-selected quasars \cite{Jiang_2007},whereas the vertical line indicates the typical size of the stellar component of our quasars. The red data point (with an uncertainty of 4.7 = $sqrt{0.4 \times 0.6 \times 91}$ ) is taken from a separate study done at the same redshift at 1.4 GHz in the X-ray band. band on post-merger data.  The trend in our data points (as well as the additional red point) suggest that there is a correlation between dual AGN and projected separation i.e. we are more likely to detect dual AGN as the line passes through the points / their uncertainties.