<s>
In	O
statistics	O
,	O
homogeneity	B-General_Concept
and	O
its	O
opposite	O
,	O
heterogeneity	B-General_Concept
,	O
arise	O
in	O
describing	O
the	O
properties	O
of	O
a	O
dataset	B-General_Concept
,	O
or	O
several	O
datasets	B-General_Concept
.	O
</s>
<s>
They	O
relate	O
to	O
the	O
validity	O
of	O
the	O
often	O
convenient	O
assumption	O
that	O
the	O
statistical	O
properties	O
of	O
any	O
one	O
part	O
of	O
an	O
overall	O
dataset	B-General_Concept
are	O
the	O
same	O
as	O
any	O
other	O
part	O
.	O
</s>
<s>
In	O
meta-analysis	O
,	O
which	O
combines	O
the	O
data	O
from	O
several	O
studies	O
,	O
homogeneity	B-General_Concept
measures	O
the	O
differences	O
or	O
similarities	O
between	O
the	O
several	O
studies	O
(	O
see	O
also	O
Study	B-General_Concept
heterogeneity	I-General_Concept
)	O
.	O
</s>
<s>
Homogeneity	B-General_Concept
can	O
be	O
studied	O
to	O
several	O
degrees	O
of	O
complexity	O
.	O
</s>
<s>
For	O
example	O
,	O
considerations	O
of	O
homoscedasticity	B-General_Concept
examine	O
how	O
much	O
the	O
variability	O
of	O
data-values	O
changes	O
throughout	O
a	O
dataset	B-General_Concept
.	O
</s>
<s>
However	O
,	O
questions	O
of	O
homogeneity	B-General_Concept
apply	O
to	O
all	O
aspects	O
of	O
the	O
statistical	B-General_Concept
distributions	I-General_Concept
,	O
including	O
the	O
location	O
parameter	O
.	O
</s>
<s>
An	O
intermediate-level	O
study	O
might	O
move	O
from	O
looking	O
at	O
the	O
variability	O
to	O
studying	O
changes	O
in	O
the	O
skewness	B-General_Concept
.	O
</s>
<s>
In	O
addition	O
to	O
these	O
,	O
questions	O
of	O
homogeneity	B-General_Concept
apply	O
also	O
to	O
the	O
joint	O
distributions	O
.	O
</s>
<s>
The	O
concept	O
of	O
homogeneity	B-General_Concept
can	O
be	O
applied	O
in	O
many	O
different	O
ways	O
and	O
,	O
for	O
certain	O
types	O
of	O
statistical	O
analysis	O
,	O
it	O
is	O
used	O
to	O
look	O
for	O
further	O
properties	O
that	O
might	O
need	O
to	O
be	O
treated	O
as	O
varying	O
within	O
a	O
dataset	B-General_Concept
once	O
some	O
initial	O
types	O
of	O
non-homogeneity	O
have	O
been	O
dealt	O
with	O
.	O
</s>
<s>
Differences	O
in	O
the	O
typical	O
values	O
across	O
the	O
dataset	B-General_Concept
might	O
initially	O
be	O
dealt	O
with	O
by	O
constructing	O
a	O
regression	O
model	O
using	O
certain	O
explanatory	O
variables	O
to	O
relate	O
variations	O
in	O
the	O
typical	O
value	O
to	O
known	O
quantities	O
.	O
</s>
<s>
There	O
should	O
then	O
be	O
a	O
later	O
stage	O
of	O
analysis	O
to	O
examine	O
whether	O
the	O
errors	O
in	O
the	O
predictions	O
from	O
the	O
regression	O
behave	O
in	O
the	O
same	O
way	O
across	O
the	O
dataset	B-General_Concept
.	O
</s>
<s>
Thus	O
the	O
question	O
becomes	O
one	O
of	O
the	O
homogeneity	B-General_Concept
of	O
the	O
distribution	O
of	O
the	O
residuals	O
,	O
as	O
the	O
explanatory	O
variables	O
change	O
.	O
</s>
<s>
The	O
initial	O
stages	O
in	O
the	O
analysis	O
of	O
a	O
time	O
series	O
may	O
involve	O
plotting	O
values	O
against	O
time	O
to	O
examine	O
homogeneity	B-General_Concept
of	O
the	O
series	O
in	O
various	O
ways	O
:	O
stability	O
across	O
time	O
as	O
opposed	O
to	O
a	O
trend	O
;	O
stability	O
of	O
local	O
fluctuations	O
over	O
time	O
.	O
</s>
<s>
There	O
can	O
then	O
be	O
questions	O
of	O
examining	O
the	O
homogeneity	B-General_Concept
across	O
sites	O
of	O
the	O
distribution	O
of	O
the	O
scaled	O
values	O
.	O
</s>
<s>
In	O
meteorology	O
,	O
weather	O
datasets	B-General_Concept
are	O
acquired	O
over	O
many	O
years	O
of	O
record	O
and	O
,	O
as	O
part	O
of	O
this	O
,	O
measurements	O
at	O
certain	O
stations	O
may	O
cease	O
occasionally	O
while	O
,	O
at	O
around	O
the	O
same	O
time	O
,	O
measurements	O
may	O
start	O
at	O
nearby	O
locations	O
.	O
</s>
<s>
There	O
are	O
then	O
questions	O
as	O
to	O
whether	O
,	O
if	O
the	O
records	O
are	O
combined	O
to	O
form	O
a	O
single	O
longer	O
set	O
of	O
records	O
,	O
those	O
records	O
can	O
be	O
considered	O
homogeneous	B-General_Concept
over	O
time	O
.	O
</s>
<s>
An	O
example	O
of	O
homogeneity	B-General_Concept
testing	O
of	O
wind	O
speed	O
and	O
direction	O
data	O
can	O
be	O
found	O
in	O
Romanić	O
et	O
al.	O
,	O
2015	O
.	O
</s>
<s>
Simple	O
populations	O
surveys	O
may	O
start	O
from	O
the	O
idea	O
that	O
responses	O
will	O
be	O
homogeneous	B-General_Concept
across	O
the	O
whole	O
of	O
a	O
population	O
.	O
</s>
<s>
Assessing	O
the	O
homogeneity	B-General_Concept
of	O
the	O
population	O
would	O
involve	O
looking	O
to	O
see	O
whether	O
the	O
responses	O
of	O
certain	O
identifiable	O
subpopulations	O
differ	O
from	O
those	O
of	O
others	O
.	O
</s>
<s>
A	O
test	O
for	O
homogeneity	B-General_Concept
,	O
in	O
the	O
sense	O
of	O
exact	O
equivalence	O
of	O
statistical	B-General_Concept
distributions	I-General_Concept
,	O
can	O
be	O
based	O
on	O
an	O
E-statistic	O
.	O
</s>
<s>
A	O
location	B-General_Concept
test	I-General_Concept
tests	O
the	O
simpler	O
hypothesis	O
that	O
distributions	O
have	O
the	O
same	O
location	O
parameter	O
.	O
</s>
