PART
B
OF
THE
SUPPORTING
STATEMENT
INFORMATION
COLLECTION
REQUEST
(
EPA
ICR
Number
1996.01)

National
Survey
on
Environmental
Management
of
Asthma
U.
S.
Environmental
Protection
Agency
Office
of
Radiation
and
Indoor
Air
Indoor
Environments
Division
B­
1
INFORMATION
COLLECTION
REQUEST
(
EPA
ICR
No.
1996.01)

U.
S.
Environmental
Protection
Agency
Office
of
Radiation
and
Indoor
Air
Indoor
Environments
Division
1
B
SURVEY
OBJECTIVES,
KEY
VARIABLES,
AND
OTHER
PRELIMINARIES
1(
a)
Survey
Objectives
The
Indoor
Environments
Division
(
IED)
is
working
to
integrate
the
management
of
environmental
factors
with
the
medical
treatment
of
asthma,
particularly
among
children
and
lowincome
populations.
In
an
effort
to
evaluate
the
effectiveness
of
its
current
outreach
efforts,
EPA
is
proposing
to
conduct
a
survey
of
U.
S.
households
to
gain
information
regarding
the
number
of
individuals
with
asthma
who
have
taken
steps
to
improve
the
quality
of
their
indoor
environment
as
part
of
their
approach
to
managing
the
disease,
as
well
as
any
barriers
they
may
have
encountered
while
attempting
to
do
so.
EPA
will
compare
the
data
gained
from
this
survey
to
the
Agency's
established
Government
Performance
and
Results
Act
of
1993
(
GPRA)
goal
of
reducing
the
number
of
people
with
asthma
who
are
exposed
to
environmental
asthma
triggers.

Specifically,
EPA's
goal
is
that
2.5
million
people
with
asthma,
including
one
million
children
and
200,000
low­
income
adults,
will
have
taken
steps
to
reduce
their
exposure
to
indoor
environmental
asthma
triggers
by
2005.

EPA
expects
the
following
issues
to
be
addressed
by
the
National
Survey
on
Environmental
Management
of
Asthma:

Issue
1:
The
extent
to
which
measures
to
reduce
exposure
to
indoor
environmental
asthma
triggers
are
being
used
by
individuals
with
asthma.

Issue
2:
The
barriers
to
implementation,
if
any,
that
have
been
encountered
by
adults
with
asthma
or
parents
of
children
with
asthma
in
attempting
to
improve
their
indoor
environment.
B­
2
Issue
3:
Ways
in
which
the
Agency
might
better
target
its
outreach
efforts
to
encourage
individuals
with
asthma,
or
parents
of
children
with
asthma,
to
improve
the
quality
of
their
indoor
environment.

1(
b)
Key
Variables
The
key
variables
associated
with
this
survey
effort
include:

$
The
level
of
knowledge
possessed
by
adults
with
asthma
(
and
the
parents
of
children
with
asthma)
about
indoor
environmental
asthma
triggers.

$
Indoor
environmental
management
measures
being
taken
to
avoid,
or
minimize
the
impact
of,
these
triggers.

$
Barriers
to
implementing
indoor
environmental
management
measures.

These
key
variables
will
be
evaluated
in
terms
of
the
demographic
characteristics
of
the
respondents,
including
age,
race/
ethnicity,
household
income,
and
level
of
education.
The
most
significant
variable,
in
terms
of
its
potential
influence
on
survey
results,
is
the
variety
of
indoor
environmental
management
measures
currently
being
used
by
people
with
asthma.
This
variable
is
directly
related
to
the
survey's
primary
objective,
which
is
to
determine
the
extent
to
which
such
measures
are
used
by
these
individuals.
The
other
key
variables
identified
above
may
influence
the
likelihood
that
an
asthmatic
will
implement
measures
to
improve
the
indoor
environment
and
therefore
may
also
influence
survey
results.

1(
c)
Statistical
Approach
In
response
to
the
GPRA,
EPA
has
set
a
specific,
quantitative
goal
related
to
reducing
the
exposure
of
people
with
asthma,
particularly
children
and
lower­
income
populations,
to
indoor
environmental
asthma
triggers.
The
primary
objective
in
conducting
the
National
Survey
on
Environmental
Management
of
Asthma
is
to
measure
the
extent
to
which
individuals
with
asthma
are
taking
action
to
manage
their
exposure
to
indoor
environmental
asthma
triggers.
EPA
estimates
that
one
in
10
individuals
in
the
United
States
suffers
from
asthma
(
Mannino).
Given
the
size
of
the
potential
sample
population,
EPA
does
not
believe
that
it
is
practical
to
survey
every
person
with
asthma
in
the
United
States.
Consequently,
EPA
will
survey
a
subset
of
this
group
in
such
a
way
as
to
allow
the
Agency
to
draw
conclusions
about
the
group
as
a
whole
from
B­
3
the
responses
received.
Anecdotal
information
is
not
sufficient
for
this
purpose;
hence,
EPA
has
chosen
a
statistical
approach
for
the
National
Survey
on
Environmental
Management
of
Asthma.

The
survey
asks
a
series
of
questions
about
the
use
of
various
measures
to
manage
indoor
environmental
asthma
triggers.
An
analysis
of
the
results
will
allow
EPA
to
assess
its
progress
toward
the
Agency's
GPRA
goal
and
evaluate
the
effectiveness
of
current
outreach
efforts.
In
addition,
the
survey
responses
will
also
provide
EPA
with
an
indication
of
any
barriers
asthmatics
may
be
experiencing
in
their
attempts
to
implement
environmental
management
practices.
These
data
will
allow
the
Agency
to
design
outreach
materials
and
activities
to
assist
this
group
in
overcoming
obstacles
and
adopting
sound
indoor
air
quality
management
practices.

EPA
will
be
assisted
in
this
survey
effort
by
an
independent
contractor
who
will
be
responsible
for
developing
the
survey
response
database;
identifying
the
survey
sample
pool;
overseeing
the
conduct
of
the
telephone
survey
and
capture
of
survey
results;
tabulating
and
analyzing
data;
and
reporting
results.
The
actual
survey
will
be
conducted
by
a
firm
that
specializes
in
administering
telephone
surveys
to
the
public
about
health­
related
topics.
This
firm
will
be
chosen
through
a
competitive
process
and
will
be
required
to
carry
out
the
survey
in
accordance
with
the
methodology
proposed
in
this
ICR.

1(
d)
Feasibility
EPA
has
reviewed
the
administrative
procedures
necessary
to
conduct
the
National
Survey
on
Environmental
Management
of
Asthma
and
has
concluded
that
it
is
feasible
to
undertake
the
survey.
EPA
has
sufficient
funding
to
conduct
the
survey
and
will
put
a
contract
in
place
to
provide
the
necessary
logistical
support.
The
survey
was
peer­
reviewed
by
staff
of
EPA's
IED
and
EPA
Regional
offices
to
ensure
that
the
questions
asked
will
reveal
sufficient
information
to
adequately
evaluate
the
implementation
of
measures
to
reduce
exposure
to
indoor
environmental
asthma
triggers
and
to
ensure
that
the
questions
do
not
collect
redundant
information.

EPA
estimates
that
it
will
take
approximately
seven
to
eight
months
to
administer
the
National
Survey
on
Environmental
Management
of
Asthma,
capture
and
analyze
survey
responses,
and
report
its
findings.
EPA
plans
to
initiate
its
survey
within
weeks
of
receiving
OMB
approval
and,

therefore,
expects
to
complete
the
survey
well
within
the
period
for
which
this
ICR
is
in
effect.
B­
4
2
B
SURVEY
DESIGN
2(
a)
Target
Population
and
Coverage
Because
there
is
no
known
source
of
information
available
to
EPA
which
identifies
individuals
in
the
United
States
with
asthma,
the
Agency
is
targeting
U.
S.
households
with
telephones
to
identify
eligible
candidates
for
its
survey.
Random
digit
dialing
will
be
used
to
select
residential
telephone
numbers
across
the
country.
The
Agency
will
over
sample
in
those
zip
codes
designated
by
U.
S.
Census
data
to
have
a
higher
proportion
of
low­
income
populations.
The
National
Survey
on
Environmental
Management
of
Asthma
will
be
administered
to
only
those
households
in
which
the
respondent
identifies
that
an
individual
with
asthma
resides
in
the
home.

2(
b)
Sample
Design
2(
b)
i
Sampling
Frame
87,652
households
will
comprise
the
sample
frame
for
the
National
Survey
on
Environmental
Management
of
Asthma.
These
households
will
be
chosen
randomly
from
all
U.
S.
households
with
telephones.
EPA
will
ensure
that
the
contractor
selected
to
generate
the
sample
frame
and
administer
the
survey
has
the
capability
for
random
digit
dialing
residential
telephones
and
access
to
regional
data
regarding
low­
income
populations.

2(
b)
ii
Sample
Size
Of
the
87,652
households
in
the
sample
frame,
EPA
expects
to
reach
52,591
individuals
during
the
period
for
which
this
ICR
is
in
effect.
Of
these
individuals,
EPA
expects
that
15
percent
of
households
will
contain
an
asthmatic,
resulting
in
7,889
eligible
individuals.
After
responding
to
several
screening
questions,
adults
with
asthma
and
parents
of
children
with
asthma
will
be
invited
to
participate
in
a
longer,
more
in­
depth
telephone
survey.
EPA
estimates
that
approximately
74
percent,
or
5,837
individuals,
will
agree
to
participate
in
the
Agency's
survey.

Only
one
individual
per
eligible
household
will
be
selected
for
a
more
in­
depth
survey.
If
the
household
contains
more
than
one
asthmatic,
the
following
formula
will
apply:

S
If
both
an
adult
and
a
child
have
asthma,
questions
will
be
asked
about
the
child.

S
If
more
than
one
child
in
the
household
has
asthma,
a
random
determination
will
be
made
as
to
which
child
the
questions
will
be
asked
about.
B­
5
S
If
more
than
one
adult
in
the
household
has
asthma,
but
there
are
no
asthmatic
children
in
the
household,
a
random
determination
will
be
made
as
to
which
adult
the
questions
will
be
asked
of.

Because
EPA
may
conduct
some
or
all
of
this
phase
of
its
survey
during
a
second
telephone
interview,
it
is
likely
that
a
small
percentage
of
eligible
participants
will
be
unreachable.
In
addition,
it
is
likely
that
passive
refusals
and
terminations
will
be
encountered
at
this
phase.
(
See
Appendix
2
for
a
complete
discussion
of
passive
refusals
and
terminations.)
Ultimately,
EPA
expects
to
obtain
completed
survey
responses
from
4,961
individuals.

In
selecting
the
sample
size
for
the
survey,
the
Agency
evaluated
a
number
of
parameters
including
confidence
levels,
precision,
and
cost.
EPA
also
considered
the
likely
response
rates
for
a
survey
of
this
kind.
EPA
selected
a
confidence
level
of
90
percent
and
a
precision
rate
of
three
percent
for
each
sample
subset,
i.
e.,
children
with
asthma
and
low­
income
adults
with
asthma.

This
will
result
in
better
than
a
two
percent
precision
rate
for
the
survey
as
a
whole
at
the
95
percent
confidence
level.
The
selection
of
these
parameters,
coupled
with
the
expected
response
rates,
led
to
the
selection
of
the
sample
size.

EPA
did
consider
higher
and
lower
precision
rates
for
the
sample
subsets.
However,
the
Agency
believed
that
a
precision
rate
lower
than
the
three
percent
selected
would
not
yield
sufficient
information
to
draw
valid
conclusions
about
the
management
of
indoor
environmental
asthma
triggers
among
children
and
low­
income
adults
with
asthma.
On
the
other
hand,
EPA
determined
that
the
added
cost
of
substantially
increasing
the
sample
size,
in
order
to
achieve
a
greater
than
three
percent
precision
rate
among
the
sample
subsets,
was
not
warranted
given
the
survey's
ultimate
objective.
Moreover,
the
sample
size
selected
will
ensure
a
95
percent
confidence
level
in
the
survey's
overall
results
and
a
90
percent
confidence
level
for
each
of
the
two
sample
subsets.
This
was
considered
sufficient
to
demonstrate
the
achievement
of
EPA's
GPRA
goal
with
a
high
degree
of
confidence.

2(
b)
iii
Stratification
Variables
EPA
believes
that
the
number
of
individuals
in
the
United
States
suffering
from
asthma
is
reaching
epidemic
proportions.
Within
the
affected
population,
the
rate
of
increase
is
largest
among
children
and
low­
income
adults.
Consequently,
EPA
is
creating
two
substrata
for
the
National
Survey
on
Environmental
Management
of
Asthma:
children
with
asthma
and
low­
income
adults
with
asthma.
Because
of
the
confidentiality
associated
with
medical
conditions
and
treatments,

EPA
is
unable
to
access
information
which
would
specifically
identify
children
and
low­
income
adults
with
asthma.
Thus,
the
Agency
cannot
create
separate
sampling
frames
for
these
two
substrata.
To
compensate,
EPA
has
designed
the
survey
to
ensure
that
a
sufficient
number
of
1See
Appendix
2
for
a
full
explanation
of
how
response
rates
were
derived.

B­
6
households
will
be
sampled
to
provide
enough
information
about
children
and
low­
income
adults
with
asthma
to
draw
valid
conclusions
about
these
populations.
Of
the
7,889
eligible
respondents,

EPA
expects
that
25
percent
will
live
with
a
child
with
asthma
(
Mannino).
Based
on
its
best
professional
judgment,
EPA
anticipates
that
the
majority
of
these
respondents
will
participate
in
the
full
survey,
thus
providing
sufficient
data
about
children
with
asthma
(
Hemmelgarn).
In
addition,
EPA
intends
to
over
sample
in
communities
known
to
have
a
high
percentage
of
lowincome
households
in
order
to
ensure
that
survey
results
will
yield
sufficient
information
to
draw
valid
conclusions
about
this
population.
As
a
result
of
this
stratification,
the
precision
rate
of
the
overall
survey
will
improve.

2(
b)
iv
Sampling
Method
To
select
the
survey
sample,
EPA
will
employ
random
digit
dialing
to
obtain
a
sample
of
87,652
publicly
listed,
residential
telephone
numbers,
over
sampling
in
low­
income
regions
throughout
the
country.
EPA
will
use
U.
S.
Census
data
to
identify
telephone
exchanges
where
a
higher
number
of
low­
income
individuals
reside.
If
information
by
telephone
exchange
is
not
available,

information
will
be
broken
down
by
zip
codes
to
identify
areas
with
a
higher
proportion
of
lowincome
residents.
However,
more
calls
may
be
required
to
obtain
the
required
number
of
lowincome
residents.

2(
b)
v
Multi­
Stage
Sampling
EPA
is
conducting
a
two­
phase
sampling
in
the
National
Survey
on
Environmental
Management
of
Asthma.
The
survey
will
use
random­
digit
dialing
with
a
minimum
of
fourteen
call
attempts
for
the
initial
contact.
In
the
first
phase,
individual
respondents
will
be
screened
to
identify
residents
with
asthma.
EPA
estimates
screening
approximately
52,591
individuals
during
this
phase
of
the
survey1.
Successful
completion
of
the
screening
phase
will
identify
the
actual
number
of
eligible
respondents.
At
this
time,
EPA
estimates
that
approximately
15
percent
of
U.
S.
households
include
at
least
one
person
with
asthma
(
NCHS).
Respondents
determined
to
be
eligible
in
the
first
phase
will
be
asked
to
participate
in
the
second
phase
of
the
survey
consisting
of
a
longer,

more
in­
depth
telephone
survey.
EPA
estimates
that
74
percent
of
those
eligible
to
participate
in
the
second
phase
of
the
survey
will
agree
to
do
so
(
Hornik,
et.
al.).

Since
respondent
time
or
contractor
staffing
may
not
allow
completion
of
the
2nd
phase
immediately
after
completion
of
the
screening
survey,
eligible
respondents
who
have
completed
B­
7
the
screening
and
have
agreed
to
participate
in
the
full
survey
may
need
to
be
re­
contacted.
The
primary
obstacle
to
the
response
rate
for
this
phase
of
the
survey
is
the
ability
to
reach
eligible
respondents
at
a
convenient
time
when
they
are
at
home.
To
overcome
this,
each
individual
will
be
asked
during
the
first
phase
about
the
best
time
for
future
contacts.
This
will
reduce
the
chance
that
a
respondent
who
has
agreed
to
participate
cannot
be
contacted.
EPA
expects
to
reach
85
percent
of
those
respondents
who
agreed
in
the
first
phase
of
the
survey
to
participate
in
the
full
asthma
survey
(
Hornik,
et.
al.
and
Mishra,
et.
al.).
B­
8
2(
c)
Precision
Requirements
2(
c)
i
Precision
Targets
EPA's
survey
has
been
designed
to
ensure
that,
at
the
95
percent
confidence
level,
its
estimate
of
the
number
of
individuals
taking
steps
to
reduce
their
exposure
to
indoor
environmental
asthma
triggers
is
within
two
percentage
points
of
the
true
value.
For
each
sample
subset,
the
Agency
expects
to
achieve
results
within
three
percentage
points
of
the
true
value
at
the
90
percent
confidence
level.
EPA
feels
that
these
precision
rates
will
be
more
than
adequate
to
characterize
the
extent
to
which
the
results
measured
by
its
survey
are
true
characteristics
of
our
nation's
asthmatic
population
and
not
a
function
of
the
margin
of
error.

2(
c)
ii
Nonsampling
Error
EPA
expects
that
the
largest
nonsampling
errors
which
could
potentially
bias
its
survey
are
associated
with
eligibility
and
nonresponse.
The
eligibility
rate
will
have
the
greatest
impact
on
the
survey
sample
due
to
the
relatively
low
rate
of
incidence
(
approximately
10
percent)
of
asthma
in
the
general
population.
However,
because
the
survey
does
not
seek
to
measure
data
on
the
indoor
environmental
management
practices
of
the
population
as
a
whole,
this
factor
(
i.
e.,

ineligibility
to
participate)
has
no
effect
on
the
accuracy
and
precision
of
survey
results.

Another
potential
source
of
bias
is
that
eligible
respondents
may
choose
not
to
participate
in
the
survey.
This
can
occur
for
two
reasons.
It
may
be
impossible
to
reach
potential
respondents.
To
minimize
this
possibility,
EPA
will
utilize
random­
digit
dialing
with
a
minimum
of
fourteen
call
attempts
for
each
phase
of
the
survey.
Another
reason
for
nonresponse
may
be
the
refusal
of
potential
respondents
to
participate
in
the
survey.
However,
health­
related
surveys
report
a
better
than
average
response
rate
(
Hemmelgarn).
In
addition,
the
salesman's
"
foot­
in­
the­
door"
effect
(
Hornik,
et
al.)
suggests
that
compliance
with
a
small
request
significantly
enhances
the
likelihood
of
compliance
with
a
subsequent
"
target"
request.
In
other
words,
people
who
are
willing
to
participate
in
the
screening
survey
can
be
expected
to
participate
in
the
full
asthma
survey.

With
this
in
mind,
EPA
estimates
that
of
the
87,652
households
which
make
up
the
sampling
frame,
60
percent,
or
52,591,
will
be
contacted
successfully
and
will
agree
to
participate
in
the
screening
survey.
Of
these
52,591
individuals,
EPA
expects
that
15
percent,
or
7,889
individuals,
will
either
have
asthma
or
live
with
someone
who
does.
Only
those
individuals
who
have
asthma
or
live
with
someone
who
does
are
considered
to
be
eligible
respondents.
B­
9
EPA
estimates
that
approximately
74
percent
of
the
7,889
eligible
respondents,
or
5,837
individuals
will
agree
to
participate
in
the
Agency's
survey.
The
Agency
expects
to
achieve
completed
surveys
for
4,961
of
these
individuals
The
overall
response
rate
or
the
number
of
survey
responses
completed
vs.
the
number
of
eligible
households
(
4,961
complete
survey
responses
÷
7,889
eligible
households)
equals
63
percent.

Based
on
this
expected
response
rate,
EPA
has
selected
a
sample
size
large
enough
to
yield
the
number
of
responses
necessary
to
achieve
a
two
percent
precision
rate
at
the
95
percent
confidence
level
for
the
overall
survey
results
and
a
three
percent
precision
rate
at
the
90
percent
confidence
level
for
each
substrata.

Although
EPA
realizes
that
an
advance
letter
would
increase
the
response
rate,
resource
constraints
and
time
constraints
do
not
allow
us
to
send
out
an
advance
letter
for
this
survey.

EPA
does,
however,
expect
a
strong
response
to
the
National
Survey
on
Environmental
Management
of
Asthma
based
on
two
factors.
First,
the
survey
will
be
conducted
by
telephone
and
multiple
attempts
will
be
made
to
reach
potential
respondents.
Second,
the
literature
suggests
that
surveys
that
collect
information
about
children's
health
issues
have
higher
response
rates
than
other
types
of
surveys.
However,
the
Agency
will
take
additional
steps
to
increase
response
rates,

including:

°
Calling
at
hours
when
people
are
likely
to
be
at
home,
e.
g.,
after
6
PM.

°
Scheduling
call
back
times
when
it
is
convenient
for
people
to
take
the
survey.

2(
d)
Questionnaire
Design
A
discussion
of
the
data
elements
contained
in
the
National
Survey
on
Environmental
Management
of
Asthma
is
included
in
Section
4(
b)(
ii)
of
Part
A
of
the
Supporting
Statement
for
this
ICR.
The
survey
seeks
information
about
the
measures
taken
by
people
with
asthma
(
and
parents
of
children
with
asthma)
to
minimize
exposure
to
indoor
environmental
asthma
triggers.

All
of
the
questions
asked
are
linked
to
the
survey's
objective
of
determining
the
extent
to
which
indoor
environmental
management
measures
are
used
by
these
individuals.

EPA
has
designed
a
questionnaire
in
which
the
respondents
are
asked
to
provide
primarily
yes/
no
responses.
In
some
cases,
respondents
are
given
a
range
of
responses
in
the
form
of
multiple
B­
10
choice
questions
and
are
asked
to
indicate
the
one
which
best
defines
their
response.
The
survey
seeks
information
on
those
environmental
management
measures
that
the
Agency
considers
important
in
reducing
an
individual's
exposure
to
known
indoor
environmental
asthma
triggers.

The
survey
is
approximately
ten
pages
long.
By
using
yes/
no
and
multiple
choice
questions,
the
Agency
has
substantially
reduced
the
amount
of
time
necessary
for
the
respondent
to
complete
the
survey
and
has
ensured
consistency
in
data
response
and
interpretation.

The
survey
instrument
was
developed
in
consultation
with
staff
from
EPA's
IED,
EPA
Regional
offices,
and
the
National
Center
for
Health
Statistics
(
NCHS)
to
ensure
that
respondents
will
understand
the
questions
asked
and
will
provide
the
type
of
data
necessary
to
measure
the
Agency's
objectives.
The
survey
was
also
designed
with
the
help
of
a
statistician
to
ensure
the
reliability
of
the
data.
B­
11
3
 
PRETESTS
AND
PILOT
TESTS
To
pretest
the
National
Survey
on
Environmental
Management
of
Asthma,
nine
adult
asthmatics
or
parents
of
children
with
asthma
were
selected
from
a
list
of
potential
respondents
identified
by
EPA
staff
and
its
cooperative
partner,
the
Asthma
and
Allergy
Network/
Mothers
of
Asthmatics.

All
nine
responded
to
the
pretest
by
completing
both
the
screening
survey
and
the
survey
itself
through
a
telephone
interview.
All
respondents
found
the
survey
to
be
easy
to
understand
and
complete.
In
addition,
the
time
it
took
for
each
respondent
to
participate
in
the
survey
was
not
considered
to
be
overly
burdensome.
Several
of
the
pretest
respondents
made
suggestions
regarding
the
survey's
format
and
content,
and
EPA
has
made
minor
adjustments
to
the
survey
instrument
based
on
that
feedback.
EPA
also
submitted
the
draft
questionnaire
to
the
NCHS
cognitive
testing
laboratory
and
made
adjustments
to
the
questionnaire
based
on
the
feedback
received.

4
 
COLLECTION
METHODS
AND
FOLLOW­
UP
4(
a)
Collection
Methods
EPA
has
chosen
to
conduct
its
survey
via
telephone
interview.
This
was
done
for
three
reasons.

The
first
is
related
to
the
fact
that
the
Agency
has
no
way
of
targeting
the
intended
respondents,

i.
e.,
people
with
asthma.
Consequently,
EPA
considered
it
more
efficient
to
call
individual
households,
rather
than
distribute
the
survey
by
mail,
to
determine
their
eligibility.
Secondly,
the
survey
questions
are
tailored
to
the
responses
of
the
participant
and
therefore
require
an
interactive
interview
which
cannot
be
accomplished
using
a
mail
survey.
Lastly,
based
on
the
statistical
literature,
EPA
expects
a
higher
response
rate
with
a
telephone
survey
than
the
Agency
would
otherwise
realize
with
a
survey
sent
in
the
mail.
A
higher
response
rate
decreases
the
size
of
the
sample
frame
needed,
and
the
level
of
follow­
up
required,
to
achieve
the
Agency's
target
precision
rates
and
confidence
levels.
Hence,
EPA
chose
to
administer
the
National
Survey
on
Environmental
Management
of
Asthma
using
telephone
interviews.

EPA
will
seek
to
find
a
firm
knowledgeable
in
environmental
health
issues,
particularly
indoor
air
quality
and
asthma,
to
administer
the
survey.
Interviewers
will
receive
training
on
the
intent
of
the
survey,
the
range
of
potential
responses,
and
definitions
of
key
terms
used
or
addressed
by
the
survey.
B­
12
4(
b)
Survey
Response
and
Follow­
Up
The
target
response
rate
for
the
National
Survey
on
Environmental
Management
of
Asthma
is
63
percent.
Actual
response
rates,
for
both
the
screening
survey
and
the
survey
itself,
will
be
measured
using
the
following
formula:

Survey
Screening:
Response
Rate
=
Number
of
Responses
÷
Number
of
Different
Telephone
Numbers
Called
Survey:
Response
Rate
=
Number
of
Responses
÷
Number
of
Eligible
Respondents
Identified
in
the
Survey
Screening
The
Agency's
contractor
will
be
responsible
for
tracking
survey
responses,
conducting
survey
interviews,
entering
survey
results
into
the
program
database,
maintaining
the
data
in
a
secured
environment,
and
providing
quality
assurance/
quality
control
of
all
survey
activities.
The
contractor
will
also
be
responsible
for
ensuring
the
accuracy
and
completeness
of
collected
information.
A
computer
database
will
be
developed
to
guide
interviewers
through
the
survey
and
will
prompt
the
interviewer
to
ask
appropriate
questions
based
on
the
responses
received.

The
interviewers
will
enter
data
as
the
survey
is
being
conducted,
and
the
database
will
have
error­
checking
features
to
ensure
that
data
are
entered
correctly
and
provide
sufficient
information
to
interpret
results.
Following
data
entry,
a
final
review
of
survey
responses
will
be
performed
to
identify
numerical
outliers
in
individual
responses.
Outliers
will
be
evaluated,

modified
within
the
database
where
appropriate,
or
marked
as
invalid.
Those
records
marked
as
invalid
will
be
dropped
from
the
response
pool.

5
 
ANALYZING
AND
REPORTING
SURVEY
RESULTS
5(
a)
Data
Preparation
Data
will
be
entered
into
the
National
Survey
on
Environmental
Management
of
Asthma
database
by
EPA
contractor
staff
at
the
time
of
the
interview.
As
mentioned
in
Section
4(
b),
the
database
will
be
developed
to
lead
the
interviewer
and
respondent
through
a
series
of
questions
tailored
to
the
responses
given.
This
design
feature
will
eliminate
the
possibility
for
interviewer
error
in
the
type
and
progression
of
questions
asked
and
will
minimize
the
potential
for
error
due
to
incomplete
responses.
The
database
will
incorporate
error­
checking
features
and
will
prompt
the
interviewer
during
the
course
of
the
survey
when
errors
in
data
entry
or
completeness
are
detected.
This
design
feature
will
reduce
the
potential
for
outliers
or
invalid
responses
to
individual
questions.
B­
13
5(
b)
Analysis
The
data
obtained
through
this
survey
will
be
aggregated
and
analyzed
for
the
purpose
of
evaluating
the
progress
made
toward
meeting
EPA's
GPRA
goal.
Data
concerning
specific
asthma
triggers
will
be
correlated
to
the
appropriate
indoor
environmental
management
measures.

For
example,
if
a
respondent
considers
molds
to
be
an
asthma
trigger,
that
information
will
be
correlated
with
the
questions
concerning
ways
in
which
to
control
mold
in
the
home.
In
this
way,

EPA
can
determine
if
respondents
are
taking
appropriate
actions
to
minimize
or
avoid
the
impact
of
indoor
environmental
asthma
triggers.
Based
on
the
number
of
respondents
who
indicate
that
they
are
using
indoor
environmental
management
measures,
EPA
will
extrapolate
a
percentage
value
as
it
relates
to
the
asthmatic
population
as
a
whole
in
the
United
States.
By
doing
so,
the
Agency
can
determine
its
progress
towards
meeting
its
GPRA
goal.

Data
will
also
be
tabulated
for
each
section
of
the
questionnaire
based
on
the
number
of
like
responses
to
each
question.
In
this
way,
EPA
can
evaluate
trends
in
terms
of
common
barriers,

control
measures
most
likely
to
be
taken,
etc.
This
information
can
then
be
used
by
EPA
to
guide
its
asthma
program
and
the
implementation
of
future
outreach
initiatives.
Data
will
be
compiled
in
a
set
of
tables
similar
to
the
following
example:

Table
5.1
 
Control
Method
Used
for
Roaches
Asthma
Trigger
Do
Nothing
Use
Roach
Pesticide
Spray
Use
Roach
Traps
Use
Boric
Acid
Other
Cockroaches
Each
type
of
analysis
will
include
an
estimate
of
the
relative
error
associated
with
the
data
reported.

Nonresponse
is
unavoidable
in
any
survey.
Nonresponse
may
be
a
source
of
bias
in
the
survey
estimates
if
the
nonrespondents
tend
to
fall
into
certain
categories.
For
example,
if
most
nonrespondents
live
in
the
Midwest,
this
may
indicate
a
source
of
bias.
If
the
response
rates
are
not
equal
for
all
sampled
units,
this
may
suggest
nonresponse
bias
and
the
estimates
calculated
using
base
weights
applied
only
to
the
sample
respondents
may
be
biased.
To
evaluate
this
possibility,

survey
subclasses
(
cells)
will
be
developed
based
on
geographic
region,
metropolitan
area
vs.

nonmetropolitan
area,
and
income
class.
Unit
nonresponse
rates
will
be
estimated
for
each
cell.
If
the
unit
nonresponse
rates
differ
significantly
across
the
defined
cells,
then
it
will
be
necessary
to
made
adjustments
for
the
nonrespondents
when
calculating
the
survey
estimates.
B­
14
To
reduce
nonresponse
bias
as
much
as
possible,
adjustment
factors
will
be
computed
independently
for
each
cell.
The
base
weights
of
respondents
in
each
cell
will
be
adjusted
upward,

so
that
the
respondents
in
each
cell
will
represent
both
the
nonrespondents
and
the
unsampled
households
in
that
cell.
Although
it
is
not
possible
to
eliminate
nonresponse
bias
entirely,
the
weight
adjustment
procedure
is
expected
to
reduce
nonresponse
bias
in
the
survey
statistics
to
an
acceptable
level.

This
approach
for
analyzing
the
potential
for
nonresponse
bias
is
based
on
the
following
two
references:

°
Survey
of
Research
and
Development
Funding
and
Performance
by
Nonprofit
Organizations:
1996
and
1997,
Methodology
Report,
National
Science
Foundation,
2000.

(
http://
www.
nsf.
gov/
sbe/
srs/
srdfpnpo/
meth9697/
start.
htm)

°
Commercial
Buildings
Energy
Consumption
Survey
(
CBECS),
Technical
and
Methodological
Information:
Nonsampling
and
Sampling
Errors,
U.
S.
Department
of
Energy,
1999.

(
http://
www.
eia.
doe.
gov/
emeu/
cbecs/
tech_
errors_
nonresponse.
html)

5(
c)
Reporting
Results
The
results
of
the
survey
will
be
compiled
into
a
summary
document
and
will
be
made
available
to
the
U.
S.
Congress
during
EPA's
reporting
of
its
GPRA
goals.
Additionally,
IED
will
report
to
the
Information
Strategies
Branch
on
the
results
of
the
survey
process
and,
upon
request,
make
copies
of
the
survey
instrument
and
this
ICR
available
as
a
reference
for
other
EPA
Program
Offices
undertaking
large­
scale
survey
and
program
evaluation
efforts.
EPA
will
not
have
direct
access
to
personal
identifiers.
Raw
survey
data
will
be
stripped
of
all
personal
identifiers
by
the
contractor
before
being
released
to
the
Agency.
