<s>
Grouped	O
data	O
are	O
data	O
formed	O
by	O
aggregating	O
individual	O
observations	O
of	O
a	O
variable	O
into	O
groups	O
,	O
so	O
that	O
a	O
frequency	O
distribution	O
of	O
these	O
groups	O
serves	O
as	O
a	O
convenient	O
means	O
of	O
summarizing	O
or	O
analyzing	B-General_Concept
the	O
data	O
.	O
</s>
<s>
There	O
are	O
two	O
major	O
types	O
of	O
grouping	O
:	O
data	B-General_Concept
binning	I-General_Concept
of	O
a	O
single-dimensional	O
variable	O
,	O
replacing	O
individual	O
numbers	O
by	O
counts	O
in	O
bins	O
;	O
and	O
grouping	O
multi-dimensional	O
variables	O
by	O
some	O
of	O
the	O
dimensions	O
(	O
especially	O
by	O
independent	O
variables	O
)	O
,	O
obtaining	O
the	O
distribution	O
of	O
ungrouped	O
dimensions	O
(	O
especially	O
the	O
dependent	O
variables	O
)	O
.	O
</s>
<s>
Note	O
that	O
the	O
result	O
of	O
this	O
will	O
be	O
different	O
from	O
the	O
sample	O
mean	O
of	O
the	O
ungrouped	B-General_Concept
data	I-General_Concept
.	O
</s>
