\begin{table}%t2
%\centering
\par
\caption{\label{tab:med_diff}Differences in the physical parameters derived with other models or constraints than the fiducial ones.}
\begin{tabular}{lrr}
\hline\hline \noalign{\smallskip}
Parameter                    & PKS1138         & GOODS\\
\hline
age [Gyr]                    & $+$0.13 (75\%)  & 0    (81\%)\\
SF time scale [Gyr]          & 0       (100\%) & 0    (70\%)\\
SFR [$M_\odot~{\rm yr}^{-1}$] & $-$0.08 (88\%)  & $-$0.6 (81\%)\\
$A_V$ [mag]              & 0       (100\%) & 0     (88\%)\\
\hline
%\hline
age [Gyr]                    & 0       (92\%)  & 0    (96\%)\\
SF time scale [Gyr]          & 0       (90\%)  & 0    (100\%)\\
SFR [$M_\odot~{\rm yr}^{-1}$] & 0       (76\%)  & 0    (92\%)\\
$A_V$ [mag]              & 0       (80\%)  & 0    (97\%)\\\hline
%\hline
age [Gyr]                    & 0      (70\%)   & 0    (79\%)\\
SF time scale [Gyr]          & 0      (72\%)   & 0    (80\%)\\
SFR [$M_\odot~{\rm yr}^{-1}$] & $-$0.05 (56\%)  & 0    (73\%)\\
$A_V$ [mag]              & $-$0.50 (58\%)  & $-$0.75 (37\%)\\
\hline
\end{tabular}
\tablefoot{The top panel shows the median differences between the parameters derived at $z_{\rm phot}$ and those at $z_{\rm spec}$.  The numbers in the parenthesis shows the fraction at which the two parameters agree within 1$\sigma$. The middle panel shows the differences we obtain if we do not use the logical age constraint. The numbers in the bottom panel mean the median differences between parameters from the secondary burst models and those from the fiducial~$\tau$ models.}
\end{table}