<s>
Sensitivity	B-General_Concept
auditing	I-General_Concept
is	O
an	O
extension	O
of	O
sensitivity	O
analysis	O
for	O
use	O
in	O
policy-relevant	O
modelling	O
studies	O
.	O
</s>
<s>
In	O
order	O
to	O
take	O
these	O
concerns	O
into	O
due	O
consideration	O
,	O
sensitivity	B-General_Concept
auditing	I-General_Concept
extends	O
the	O
instruments	O
of	O
sensitivity	O
analysis	O
to	O
provide	O
an	O
assessment	O
of	O
the	O
entire	O
knowledge	O
-	O
and	O
model-generating	O
process	O
.	O
</s>
<s>
Likewise	O
,	O
sensitivity	B-General_Concept
auditing	I-General_Concept
has	O
been	O
developed	O
to	O
provide	O
pedigrees	O
of	O
models	O
and	O
model-based	O
inferences	O
.	O
</s>
<s>
Sensitivity	B-General_Concept
auditing	I-General_Concept
is	O
especially	O
suitable	O
in	O
an	O
adversarial	O
context	O
,	O
where	O
not	O
only	O
the	O
nature	O
of	O
the	O
evidence	O
,	O
but	O
also	O
the	O
degree	O
of	O
certainty	O
and	O
uncertainty	O
associated	O
to	O
the	O
evidence	O
,	O
is	O
the	O
subject	O
of	O
partisan	O
interests	O
.	O
</s>
<s>
Sensitivity	B-General_Concept
auditing	I-General_Concept
-	O
together	O
with	O
post-normal	O
science	O
is	O
one	O
of	O
the	O
lenses	O
recommended	O
to	O
study	O
sustainability	O
in	O
.	O
</s>
<s>
Sensitivity	B-General_Concept
auditing	I-General_Concept
is	O
recommended	O
by	O
the	O
European	O
Commission	O
for	O
use	O
in	O
impact	O
assessments	O
in	O
order	O
to	O
improve	O
the	O
quality	O
of	O
model-based	O
evidence	O
used	O
to	O
support	O
policy	O
decisions	O
.	O
</s>
<s>
Sensitivity	B-General_Concept
auditing	I-General_Concept
is	O
summarised	O
by	O
seven	O
rules	O
or	O
guiding	O
principles	O
:	O
</s>
<s>
Both	O
rules	O
originate	O
from	O
the	O
practice	O
of	O
impact	O
assessment	O
,	O
where	O
a	O
modeling	O
study	O
presented	O
without	O
a	O
proper	O
SA	O
,	O
or	O
as	O
originating	O
from	O
a	O
model	O
which	O
is	O
in	O
fact	O
a	O
black	B-Device
box	I-Device
,	O
may	O
end	O
up	O
being	O
rejected	O
by	O
stakeholders	O
.	O
</s>
<s>
Your	O
model	O
is	O
a	O
black	B-Device
box	I-Device
-	O
why	O
should	O
I	O
trust	O
your	O
results	O
?	O
</s>
<s>
Sensitivity	B-General_Concept
auditing	I-General_Concept
is	O
described	O
in	O
the	O
European	O
Commission	O
Guidelines	O
for	O
impact	O
assessment	O
.	O
</s>
<s>
"	O
[	O
…	O
]	O
where	O
there	O
is	O
a	O
major	O
disagreement	O
among	O
stakeholders	O
about	O
the	O
nature	O
of	O
the	O
problem	O
,	O
…	O
then	O
sensitivity	B-General_Concept
auditing	I-General_Concept
is	O
more	O
suitable	O
but	O
sensitivity	O
analysis	O
is	O
still	O
advisable	O
as	O
one	O
of	O
the	O
steps	O
of	O
sensitivity	O
auditing.	O
"	O
</s>
<s>
"	O
Sensitivity	B-General_Concept
auditing	I-General_Concept
,	O
 [ … ] 	O
is	O
a	O
wider	O
consideration	O
of	O
the	O
effect	O
of	O
all	O
types	O
of	O
uncertainty	O
,	O
including	O
structural	O
assumptions	O
embedded	O
in	O
the	O
model	O
,	O
and	O
subjective	O
decisions	O
taken	O
in	O
the	O
framing	O
of	O
the	O
problem.	O
"	O
</s>
<s>
"	O
In	O
general	O
sensitivity	B-General_Concept
auditing	I-General_Concept
stresses	O
the	O
idea	O
of	O
honestly	O
communicating	O
the	O
extent	O
to	O
which	O
model	O
results	O
can	O
be	O
trusted	O
,	O
taking	O
into	O
account	O
as	O
much	O
as	O
possible	O
all	O
forms	O
of	O
potential	O
uncertainty	O
,	O
and	O
to	O
anticipate	O
criticism	O
by	O
third	O
parties.	O
"	O
</s>
<s>
The	O
European	O
Academies’	O
association	O
of	O
science	O
for	O
policy	O
SAPEA	O
describes	O
in	O
detail	O
sensitivity	B-General_Concept
auditing	I-General_Concept
in	O
its	O
2019	O
report	O
entitled	O
“”	O
.	O
</s>
<s>
Application	O
of	O
concepts	O
from	O
sensitivity	B-General_Concept
auditing	I-General_Concept
are	O
to	O
the	O
OECD-PISA	O
study	O
,	O
food	O
security	O
,	O
nutrition	O
,	O
the	O
ecological	O
footprint	O
.	O
</s>
<s>
Sensitivity	B-General_Concept
auditing	I-General_Concept
is	O
among	O
the	O
tools	O
recommended	O
in	O
the	O
context	O
of	O
a	O
possible	O
ethics	O
of	O
quantification	O
,	O
which	O
aims	O
to	O
identify	O
common	O
ethical	O
elements	O
in	O
different	O
problems	O
seen	O
in	O
quantification	O
,	O
such	O
as	O
metrics	O
fixation	O
,	O
misuse	O
of	O
statistics	O
,	O
poor	O
modelling	O
and	O
unethical	O
algorithms	O
.	O
</s>
