Agency
Identifier
December
1997
Development
of
NERL/
CHAD
The
National
Exposure
Research
Laboratory
Consolidated
Human
Activity
Database
National
Exposure
Research
Laboratory
Office
of
Research
and
Development
U.
S.
Environmental
Protection
Agency
Research
Triangle
Park,
NC
27711
ii
Agency
Identifier
December
1997
Development
of
NERL/
CHAD:
The
National
Exposure
Research
Laboratory
Consolidated
Human
Activity
Database
by
Graham
Glen,
Yeshpal
Lakkadi,
Jo
Anne
Tippett,
and
Maria
del
Valle­
Torres
ManTech
Environmental
Technology,
Inc.
Research
Triangle
Park,
NC
Contract
68­
D5­
0049
Work
Assignment
Manager
Thomas
McCurdy
Human
Exposure
and
Atmospheric
Sciences
Division
National
Exposure
Research
Laboratory
Research
Triangle
Park,
NC
27711
National
Exposure
Research
Laboratory
Office
of
Research
and
Development
U.
S.
Environmental
Protection
Agency
Research
Triangle
Park,
NC
27711
iii
Notice
The
United
States
Environmental
Protection
Agency
through
its
Office
of
Research
and
Development
managed
and
funded
the
research
described
here
under
Contract
68­
D5­
0049
to
ManTech
Environmental
Technology,
Inc.
It
has
not
been
subjected
to
the
Agency's
peer
and
administrative
review
and
therefore
does
not
necessarily
reflect
the
views
of
the
Agency,
and
no
official
endorsement
should
be
inferred.
Mention
of
trade
names
or
commercial
products
does
not
constitute
endorsement
or
recommendation
for
use.
iv
Abstract
This
document
describes
the
creation
and
development
of
the
Consolidated
Human
Activity
Database
(
CHAD)
for
the
Environmental
Protection
Agency's
National
Exposure
Research
Laboratory.
This
database
was
created
to
support
exposure/
intake
dose/
risk
assessments.
The
overall
design
incorporates
comments
received
from
potential
users
knowledgeable
about
human
activity
pattern
research
and
risk
assessment.

CHAD
is
a
relational
database
with
a
graphical
user
interface
that
facilitates
queries
and
report
generation.
It
contains
databases
from
previously
existing
human
activity
pattern
studies,
which
were
incorporated
in
two
forms:
(
1)
as
the
original
raw
data
and
(
2)
as
data
modified
according
to
predefined
format
requirements.
The
latter
involved
development
of
a
common
activity/
location
code
system,
compilation
of
background
questionnaire
information,
and
the
application
of
data
quality
flags.
To
enhance
the
applicability
to
risk
assessment,
the
ability
to
estimate
metabolic
activity
ratios
for
respondents
has
been
added.

This
report
covers
a
period
from
May
to
December
1997,
and
work
was
completed
as
of
December
31,
1997.
v
Contents
Abstract
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
iii
Figures
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
vi
Tables
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
vi
Acronyms
and
Abbreviations
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
vii
Acknowledgment
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
viii
1
Introduction
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
1
2
Recommendations
for
Future
Work
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
2
3
Scoping
Activities
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
3
3.1
Contacting
Potential
Users
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
3
3.2
Data
Acquisition
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
4
4
Code
Consolidation
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
7
4.1
Activity
and
Location
Codes
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
7
4.2
Standardization
of
Time
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
8
4.3
Questionnaire
Data
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
9
5
Database
Design
and
Development
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
10
5.1
Data
Format
and
Content
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
10
5.2
Platform
Selection
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
11
5.3
Development
Process
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
12
6
MET
Data
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
17
7
Quality
Assurance
Procedures
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
19
7.1
Determination
of
Quality
Flags
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
19
7.2
QA
Issues
in
Coding
CHAD
Variables
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
24
8
Discussion
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
27
8.1
California
Activity
Codes
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
27
8.2
Resolution
of
Quality
Flags
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
27
8.3
Nonrandom
Sampling
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
27
References
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
29
Appendices
A
Exposure
Analysts
and
Modelers
Contacted
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
31
B
CHAD
Activity
Codes
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
33
C
CHAD
Location
Codes
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
43
D
Occupational
Codes
Used
in
CHAD
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
51
E
CHAD
Questionnaire
Information
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
55
F
Program
for
Recoding
NHAPS
Variables
into
CHAD
Format
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
56
G
Modeling
Multi­
Route
Intake
Dose
in
an
Internally
Consistent
Manner
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
73
H
METS
Distributions
for
CHAD
Activities
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
77
vi
Figures
5­
1
Information
Formats
Available
Through
Microsoft
Access
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
11
5­
2
Sample
Query
Data
in
ASCII
Format
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
12
5­
3
Sample
Screen
of
CHAD
Questionnaire
Data
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
13
5­
4
Sample
Screen
of
Diary
Data
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
14
5­
5
CHAD
Main
Screen
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
14
5­
6
Sample
Screen
of
Filter
Options
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
15
5­
7
Example
of
Query
Results
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
16
6­
8
Sample
Screen
of
METS
Data
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
18
Tables
7­
1
Quality
Indicator
Variables
in
CHAD
Version
of
the
Data
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
20
7­
2
ive
CHAD
Variables
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
21
A­
1
Modelers
Contacted
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
32
B­
1
HAD
Activity
Codes
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
34
C­
1
AD
Location
Codes
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
44
H­
1
for
CHAD
Activities
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
.
77
vii
Acronyms
and
Abbreviations
CARB 
California
Air
Resources
Board
CHAD 
Consolidated
Human
Activity
Database
CHADID 
A
unique
identification
number
for
each
individual
in
the
CHAD
database
GUI 
Graphical
User
Interface
METS 
Ratio
of
an
activity­
specific
metabolic
rate
to
a
person's
resting
metabolic
rate
NHAPS 
National
Human
Activity
Pattern
Survey
NHEXAS 
National
Human
Exposure
Assessment
Survey
PAI 
Physical
Activity
Index
PEM 
Personal
exposure
monitor
QC
Variable 
Quality
count
variable
viii
Acknowledgment
ManTech
Environmental
Technology,
Inc.,
would
like
to
thank
William
C.
Steen
for
supporting
this
project;
James
Capel
and
Ted
Johnson
for
providing
valuable
data;
and
Ted
Palma
for
his
encouragement.
In
addition,
thanks
go
to
the
numerous
people
contacted
as
sources
of
information
and
data
that
have
enhanced
this
database.
1
Chapter
1
Introduction
This
report
documents
the
design
and
development
of
EPA's
National
Exposure
Research
Laboratory/
Consolidated
Human
Activity
Database
(
CHAD),
a
comprehensive
human
activity
database.
CHAD
contains
data
obtained
from
pre­
existing
human
activity
studies
that
were
collected
at
city,
state,
and
national
levels.
The
purpose
of
CHAD
is
to
"
provide
a
basis
for
conducting
multi­
route,
multi­
media
exposure/
intake
dose/
risk
assessments"
(
McCurdy,
1997a).

The
database
was
developed
in
two
segments.
Segment
1
was
begun
in
May
1997
and
focused
on
initial
scoping
efforts,
comparisons
of
acquired
target
databases,
development
of
a
code
system,
and
creation
of
data
quality
flags.
Segment
2
tasks
were
started
in
June
1997
and
concentrated
on
platform
selection,
creation
of
the
database
structure,
conflation
of
the
target
databases
into
CHAD
format,
application
of
data
quality
flags,
incorporation
of
the
CHAD­
formatted
data
and
the
original
source
data
(
along
with
pre­
existing
flags)
into
the
database,
resolution
of
data
quality
flags,
identification
of
future
variables
to
be
added,
and
further
scoping
efforts.

This
report
describes
the
development,
design,
and
present
state
of
CHAD.
Chapter
2
presents
anticipated
future
work
on
this
database.
Chapter
3
presents
the
scoping
activities
that
lay
the
groundwork
for
the
design
and
development
of
CHAD.
Chapter
4
presents
the
rationale
and
method
for
consolidating
the
codes
from
the
various
contributing
databases.
Chapter
5
describes
the
overall
design
of
the
consolidated
database,
a
view
of
its
infrastructure,
and
a
discussion
of
the
first
three
developmental
phases
of
CHAD.
Chapter
6
presents
Phase
4
of
the
development,
which
centers
on
the
METS
data.
Chapter
7
is
a
description
of
the
quality
assurance
procedures
that
were
implemented
in
the
development
of
CHAD.
Chapter
8
is
a
discussion
of
outstanding
issues 
the
California
activity
codes,
resolution
of
quality
flags,
and
nonrandom
sampling.
References
and
appendices
are
included
at
the
end
of
the
text.
2
Chapter
2
Recommendations
for
Future
Work
As
with
the
creation
of
any
database
system,
it
is
expected
that
system
test
errors
or
other
such
problems
will
be
found.
Also,
tasks
identified
previously
in
this
document
need
to
be
completed 
such
as
the
addition
of
variables.
Refinement
of
CHAD
will
therefore
continue.
The
following
is
a
brief
description
of
some
of
the
tasks
that
have
been
identified
for
future
completion.

1.
Once
CHAD
is
reformatted
in
Oracle,
it
will
be
placed
on
the
World
Wide
Web
in
a
manner
consistent
with
the
EPA's
Office
of
Information
Resource
Management
standards.
This
will
enable
increased
access
to
environmental
and
human
activity
data.
Designer
2000,
a
web
development
software
application,
will
be
used
to
accomplish
this.
Users
will
be
able
to
download
CHAD
query
results
and
raw
data
batch
files
from
the
target
databases.

2.
Estimates
of
body
mass
(
weight
in
kg)
will
be
added
to
CHAD.

3.
Climatic
data
(
e.
g.,
temperature
and
precipitation)
from
the
closest
available
monitoring
sites
will
be
obtained
for
the
days
of
human
activity
currently
in
the
database.
If
the
day
of
response
or
the
location
of
an
individual
can
not
be
determined,
then
seasonally
and/
or
spatially
averaged
climatic
data
contained
in
the
EPA
standard
climate
atlas
will
be
used.

4.
The
METS
data
will
be
modified
so
that
they
account
for
post­
exercise
"
excess"
oxygen
consumption.
This
will
affect
only
the
METS
expenditures
equal
to
or
greater
than
3.0.
In
addition,
CHAD
will
be
modified
so
that
daily
physical
activity
for
a
given
individual
can
be
estimated.
Ultimately,
this
will
result
in
a
Physical
Activity
Index
(
PAI)
value.

5.
The
search
for
more
data
applicable
to
CHAD
will
continue.
3
Chapter
3
Scoping
Efforts
3.1
Contacting
Potential
Users
Prior
to
initiating
the
design
and
development
of
CHAD,
scoping
efforts
began
in
June
1997
to
gather
baseline
information
about
human
activity
studies,
availability
of
data,
and
requirements
of
potential
users.
These
efforts
began
with
contacting
individuals
known
to
have
experience
with
human
activity
studies
and/
or
human
exposure
assessment
research.
In
total,
33
exposure
analysts
and
modelers
were
contacted
via
telephone
or
attended
meetings
to
discuss
CHAD
(
Table
A­
1,
Appendix
A).
Information
from
these
contacts
was
recorded
in
a
telephone
log
or
as
meeting
notes.
In
addition,
demonstrations
of
CHAD
to
potential
users 
both
inside
and
outside
of
EPA 
were
given
in
October
and
November
1997.
A
variety
of
design
and
development
suggestions
were
made
by
these
potential
users,
most
of
which
were
incorporated
into
CHAD.
The
following
are
examples
of
the
suggestions
received;
the
chapter
where
each
issue
is
discussed
in
this
report
is
given
in
parentheses.

!
A
multi­
digit,
collapsible
code
system
should
be
used
as
the
code
structure
for
CHAD
(
Chapter
4).

!
The
earliest
"
clean"
version
of
each
database
should
be
used
when
possible
(
because
of
potential
data
loss
in
later
versions),
and
the
original
coding
systems
used
in
each
original
database
should
be
retained
in
addition
to
the
new
CHAD
codes
(
Chapter
5).

!
Data
contained
in
CHAD
should
include
basic
background
information
about
the
participant:
a
first
name
identifier
of
each
individual,
date
of
birth,
gender,
health
classes,
ethnicity,
smoking
status,
occupational
code;
a
logical
breakdown
of
hours
in
a
day;
and
demographic
status
(
Chapter
5).

!
CHAD
should
be
easy
to
use
(
Chapter
5).

!
The
data
should
be
easily
interpretable,
well
documented
(
Chapter
5).

!
A
relational
database
structure
should
be
used
(
Chapter
5).

!
Quality
assurance
checks
should
be
incorporated
that
serve
as
indicators
of
data
completeness
and
should
contain
the
original
data
with
flags
(
Chapter
7).

!
Consideration
should
be
given
to
physical
activity
(
amount
of
exertion)
data
(
Chapter
6
METS
data).
4
!
Specific
information
about
environmental
conditions
should
be
provided
(
Chapter
7).

!
The
availability
of
data
on
sensitive
groups
such
as
asthma
sufferers
should
be
determined
(
Chapter
7).

!
Data
should
be
easily
accessible;
consideration
should
be
given
to
storing
CHAD
on
a
Unix
central
processor
and
using
NT
or
Unix
parameters
in
this
database
(
Chapter
7).

!
Compatibility
with
current
exposure
models
and
other
databases,
along
with
potential
linkages,
should
be
considered
(
Chapter
5).

!
Include
ability
to
filter
queries
by
state
and/
or
region.

!
Add
the
capability
to
link
to
spreadsheet
applications.

!
Provide
option
to
retrieve
full
citations
for
publications
of
target
studies.

3.2
Data
Acquisition
This
process
was
simultaneous
with
contacting
the
potential
users.
It
began
with
identification
of
data
that
would
serve
as
the
foundation
of
CHAD.
Existing
databases
from
several
human
activity
studies
were
targeted
because
of
their
comprehensiveness
and
availability
of
documentation.
Because
of
the
passage
of
time,
it
was
difficult
to
locate
the
original
databases.
However,
the
best
known
sources
were
contacted
for
each.
If
the
primary
source
did
not
have
access
to
it,
then
often
a
second
or
third
potential
source
was
contacted.
In
some
cases,
it
was
not
possible
to
locate
the
original
database;
second
generation
databases
were
then
used
instead.
At
the
end
of
this
process,
databases
were
obtained
from
the
following
studies:

!
A
Study
of
Personal
Exposure
to
Carbon
Monoxide
in
Denver,
Colorado.
This
study
was
conducted
in
1983
with
a
total
of
454
study
participants
(
Johnson,
1984).
Each
participant
carried
a
personal
exposure
monitor
(
PEM)
and
an
activity
diary
for
two
consecutive
24­
hour
sampling
periods.
Approximately
900
person
days
of
PEM
and
activity
data
were
collected.
Detailed
background
questionnaire
data
were
also
collected.
The
electronic
data
were
obtained
from
James
Capel
(
Table
A­
1,
Appendix
A).
The
questionnaire
information
was
incomplete,
and
to
date,
another
source
for
the
remaining
questionnaire
information
has
not
been
found.

!
National
Human
Activity
Pattern
Study
(
NHAPS).
This
study
was
conducted
between
October
1992
and
September
1994
(
Klepeis
et
al.,
1995),
during
which
period
24­
hour
activity
diaries
were
collected
from
9386
participants.
Questionnaire
data
were
also
collected.
This
survey
was
the
largest
and
most
geographically
diverse
(
data
from
48
states)
of
the
studies
going
into
CHAD.
It
was
conducted
as
a
next­
day
telephone
interview.
This
method
has
the
advantage
that
an
interviewer
can
insist
on
filling
in
gaps
in
a
diary,
but
suffers
from
the
limitations
of
a
participant's
memory
regarding
the
exact
time
and
duration
of
activities.
The
original
data
for
this
study
were
obtained
from
William
Nelson
of
EPA/
NERL.
5
!
Activity
Patterns
of
California
Residents.
This
study
was
conducted
from
October
1987
to
September
1988
by
the
California
Air
Resources
Board
(
CARB;
Wiley
et
al.,
1991).
It
involved
1762
randomly
selected
participants
aged
12
and
above.
Retrospective
time
diaries
were
collected
via
telephone
for
all
activities
occurring
on
the
previous
day.
A
total
of
36,
918
records
were
collected.
Questionnaire
data
were
collected
on
activity
locations,
presence
of
smokers,
and
housing
and
socioeconomic
characteristics.
Like
NHAPS,
this
was
a
next­
day
telephone
survey.
The
data
for
this
study
were
also
obtained
from
James
Capel.

!
Activity
Pattern
Survey
for
California
Children.
This
survey
was
conducted
by
CARB
from
1989
to
1990
(
Wiley,
1991).
A
total
of
27,048
activity
records
were
collected
from
1200
children.
The
same
methodology
was
used
as
described
above,
although
the
questionnaires
and
diary
formats
differed
in
certain
details.
James
Capel
also
provided
these
data.

!
Study
of
Carbon
Monoxide
Exposure
of
Residents
of
Washington,
DC.
This
EPA­
sponsored
investigation
was
carried
out
concurrently
with
the
Denver
study
and
covered
705
respondents
for
one
day
each
(
Settergren
et
al.,
1984).
The
version
of
the
data
acquired
by
for
this
project
has
incomplete
questionnaire
data.
Also,
the
list
of
potential
activity
and
location
codes
is
fairly
short,
limiting
the
resolution
in
activity
description
that
is
possible
in
this
study.
James
Capel
was
the
source
for
these
data.

!
Human
Activity
Patterns
in
Cincinnati,
Ohio.
This
study
was
funded
by
the
Electric
Power
Research
Institute
(
EPRI)
and
was
conducted
in
1987
(
Johnson,
1989).
Activity
diaries
were
carried
by
973
randomly
selected
participants
for
three
consecutive
24­
hour
periods
in
March
or
August
1985,
resulting
in
2800
subject
days
of
diary
data.
Detailed
background
questionnaires
were
also
completed
by
the
participants.
The
electronic
data
were
provided
by
James
Capel.

!
Valdez
Air
Health
Study.
This
study
was
conducted
during
1990 
91
(
Goldstein
et
al.,
1992).
A
total
of
289
individuals
from
different
households
were
interviewed
as
part
of
a
continuous
monitoring
program
that
also
measured
criteria
pollutants,
volatile
organic
compounds,
and
weather
conditions.
This
sample
covered
23%
of
the
residences
in
the
Valdez
area.
The
data
for
this
study
were
provided
by
Ted
Johnson
(
Table
A­
1,
Appendix
A).

!
Los
Angeles
Area
Studies:
Development
of
Improved
Methods
to
Measure
Effective
Doses
of
Ozone.
These
studies
of
outdoor
workers
exposed
to
oxidant
pollution
(
Shamoo
et
al.,
1991),
elementary
and
high
school
students
exposed
to
oxidant
pollution
(
Spier
et
al.,
1992),
and
ozone­
exposed
construction
workers
were
conducted
between
1989
and
1992.
The
activity
records
from
these
studies
cover
only
a
work­
day
or
school­
day
period
(
approximately
eight
hours),
as
the
purpose
of
these
studies
was
to
focus
on
exposures
in
specific
microenvironments.
The
original
data
were
obtained
from
Kenneth
Clark
(
Table
A­
1,
Appendix
A).
Because
of
the
limited
study
sizes
and
the
difficulty
of
extending
the
diaries
to
a
standard
24­
hour
period,
these
studies
were
not
incorporated
into
the
CHAD
database.

Additional
data
source
contact
information
is
provided
in
a
footnote
in
Table
A­
1,
Appendix
A.
Literature
references
pertaining
to
these
studies
are
included
at
the
end
of
the
text.
6
Chapter
4
Code
Consolidation
To
conflate
the
selected
databases
and
incorporate
them
into
CHAD,
a
single
code
system
was
created
to
account
for
all
activity
and
location
codes
across
studies.
This
was
accomplished
after
examining
the
code
systems
of
each
database
and
then
comparing
them
for
common
categories.

4.1
Activity
and
Location
Codes
The
NHAPS
and
California
have
approximately
100
different
activity
codes
and
another
100
location
code,
in
contrast
with
earlier
studies
such
as
the
Denver
and
Washington
studies,
which
have
only
10
to
20
codes
each
for
activities
and
locations.
In
the
NHAPS
and
California
systems,
a
nesting
format
was
used
in
which
main
activity
categories
were
divided
into
subcategories.
In
some
cases,
these
subcategories
were
further
divided
into
more
specific
categories.
This
structure
accommodated
a
specific
and
comprehensive
documentation
of
participants'
activities
over
time.
Because
of
the
strong
similarity
between
the
coding
systems
of
the
NHAPS
and
the
California
studies,
use
of
one
of
these
as
the
CHAD
coding
system
would
mean
that
the
studies
with
the
greatest
number
of
diary
entries
would
need
relatively
little
modification.
The
NHAPS
system
is
slightly
more
succinct
than
the
California
code
system.
For
these
reasons,
the
NHAPS
coding
system
was
selected
as
the
foundation
for
the
CHAD
system.

The
NHAPS
system
has
been
refined
and
regrouped
as
necessary
to
incorporate
coding
systems
from
other
studies.
In
some
cases,
these
other
systems
had
finer
categories
than
did
NHAPS,
although
more
often
they
used
broader
categories.
Generally,
all
the
NHAPS
coding
categories
are
retained,
along
with
several
extra
ones
reflecting
the
categories
used
in
other
studies.
Thus
there
are
more
than
100
each
of
CHAD
activity
and
location
codes.
Another
regrouping
put
all
the
travel
codes
together
in
CHAD
(
all
starting
with
18);
they
are
not
grouped
together
in
NHAPS.

In
the
CHAD
coding
system,
activities
and
locations
are
each
designated
by
five
digits.
For
activities,
the
first
digit
is
always
1,
and
for
locations,
it
is
3;
this
prevents
activity
and
location
codes
from
being
confused.
Subsequent
digits
define
finer
distinctions.
With
a
five­
digit
system,
there
is
room
to
add
new
codes
as
they
are
needed.
The
resulting
CHAD
activity
and
location
code
systems
are
presented
in
Appendix
B
and
Appendix
C,
respectively.

In
retrospect,
certain
difficulties
are
evident
with
the
NHAPS
system.
For
instance,
both
indoor
and
outdoor
(
e.
g.,
back
yard)
locations
are
parts
of
the
primary
location
category
of
"
at
home".
For
exposure
purposes,
the
primary
7
distinction
is
the
indoor/
outdoor
one,
while
the
type
of
building
is
secondary.
The
recoding
was
already
underway
when
this
issue
became
apparent,
so
no
further
changes
to
the
CHAD
codes
were
made
at
this
time.

4.2
Standardization
of
Time
The
original
studies
used
a
variety
of
protocols
regarding
diary
start
time,
stop
time,
and
resolution.
All
of
the
studies
incorporated
into
CHAD
so
far
have
nominal
24­
hour
(
or
multiples
thereof)
durations
and
theoretical
resolutions
of
one
minute.
In
practice,
the
studies
using
written
diaries
did
not
start
at
a
standard
time,
and
they
ended
only
approximately
24
hours
later.
The
telephone
recall
surveys
generally
enforced
a
standard
midnight­
tomidnight
time
frame,
but
seldom
were
able
to
pinpoint
activity
start
and
stop
times
to
the
nearest
minute,
so
that
five­
minute
or
in
some
cases
only
15­
minute
resolution
was
achieved.

All
diaries
are
standardized
to
an
exact
24­
hour
length
for
CHAD.
If
the
original
diaries
were
longer,
then
the
first
24
hours
were
retained.
If
the
diaries
were
shorter,
then
extra
records
with
missing
activities
and
locations
were
added
to
fill
the
missing
time.
Activities
that
cross
hourly
boundaries
(
i.
e.,
that
start
before
the
top
of
the
hour
but
finish
after)
are
divided
at
the
hour.
This
was
motivated
by
the
standard
practice
of
hourly
measurement
of
concentrations
of
air
pollutants.
A
single
value
can
be
associated
with
each
activity
record
in
CHAD,
assuming
that
concentration
remains
fixed
within
the
hour.
This
assumption
eases
exposure
modeling
calculations.

Although
each
diary
is
standardized
to
24
hours
in
length,
there
is
no
common
start
time
that
could
be
applied
to
all
diaries.
This
has
been
resolved
by
"
wrapping"
the
diary
around
so
that
in
CHAD,
all
diaries
begin
and
end
at
midnight.
The
true
starting
time
of
the
original
diary
is
indicated
by
the
variable
WRAPTIME,
which
is
discussed
in
the
data
flagging
(
Chapter
7).

The
CHAD
variables
have
been
recoded
with
SAS
programs.
Quality
flags
were
applied
to
the
data
during
this
process.
A
detailed
discussion
of
the
quality
flags
is
presented
in
Chapter
7.
Listings
of
the
SAS
programs
may
be
found
in
Appendix
F.
8
4.3
Questionnaire
Data
Each
component
study
in
CHAD
was
accompanied
by
a
survey
or
questionnaire
that
recorded
personal
data
from
the
respondent,
such
as
age,
gender,
occupation,
and
house
type.
Appendix
D
provides
a
list
of
the
occupation
types
used
in
CHAD.
The
list
of
such
questions
varied
significantly
in
both
length
and
focus,
depending
on
the
purposes
of
the
original
study.
This
made
it
difficult
to
consolidate
the
questions
into
a
single
set
applicable
across
all
studies.

There
are
two
main
philosophies
on
which
a
consolidated
list
can
be
based:
Either
includes
those
questions
common
to
most
studies
(
effectively
the
intersection
set
of
questions),
or
include
all
questions
asked
in
any
study
(
the
union
of
all
questions).
In
the
case
of
the
studies
in
CHAD,
the
union
would
contain
nearly
1000
questions
per
personday
with
the
vast
majority
of
these
responses
missing
for
any
individual
person.
This
would
make
CHAD
unwieldy
and
would
not
effectively
consolidate
the
data,
as
most
questions
would
apply
to
only
one
of
the
component
studies.
Therefore,
a
small
set
(
around
30)
of
fairly
common
questions
has
been
selected
for
CHAD,
for
which
several
(
but
not
necessarily
all)
of
the
original
studies
provided
data
(
Appendix
E).
In
addition,
a
total
of
about
15
new
variables
have
been
composed
and
added
to
the
CHAD
data
to
assist
the
user
and
to
help
indicate
data
quality.
These
variables
are
described
in
detail
in
Chapter
7
of
this
report.
9
Chapter
5
Database
Design
and
Development
This
process
began
in
September
1997.
Much
of
the
initial
database
structure
was
based
on
comments
received
during
contacts
with
potential
users.
The
CHAD
design
was
further
modified
after
further
exchange
of
ideas
during
demonstrations
and
presentations
given
in
October
and
November
1997.

5.1
Data
Format
and
Content
One
of
the
most
common
concerns
from
the
viewpoint
of
potential
users
was
that
a
consolidated
database
might
not
retain
information
of
interest
that
was
collected
in
only
one
or
two
of
the
component
studies.
Also,
the
recoding
of
variables
into
a
standard
format
inevitably
results
in
some
information
loss,
as
the
exact
phrasing
of
the
questions
(
and
the
allowed
responses)
differ
somewhat
from
study
to
study.
To
answer
these
concerns,
the
CHAD
database
system
contains
(
1)
the
CHAD
consolidated
database
itself
and
(
2)
a
section
consisting
of
the
original
(
or
raw)
data
obtained
from
the
various
component
studies.
The
earliest
"
clean"
versions
of
the
original
questionnaire
and
diary
data
and
the
modified
data
were
entered
into
CHAD.
Queries
can
be
made
of
either
the
original
data
or
the
modified
(
according
to
CHAD
quality
standards)
data.
The
original
data
contain
many
questions
or
details
that
pertain
to
only
one
or
two
of
the
studies
and
therefore
were
not
common
enough
to
be
part
of
a
single,
consolidated
database.
In
the
consolidated
portion,
a
user
can
select
a
single
search
variable
(
e.
g.,
respondent's
age)
and
apply
it
to
all
of
the
studies
in
CHAD.
Also,
interpretation
of
the
variables
(
e.
g.,
activity
codes)
is
identical
for
all
studies,
which
is
not
the
case
for
the
original
data.

Graphical
user
interface
(
GUI)
screens
were
developed
for
the
CHAD
data
and
the
raw
data.
For
the
raw
data,
part
of
this
process
involved
examining
the
questionnaire
record
sheets
and
grouping
the
variables
that
covered
similar
topics.
For
instance,
all
variables
relating
to
bathing
or
showering
(
i.
e.,
BATH,
BATH#)
were
grouped.
This
ensured
that
all
of
the
necessary
questionnaire
data
were
entered
into
the
database
correctly
and
that
access
to
the
data
would
be
convenient
for
future
users.
The
modified
data
are
augmented
by
a
set
of
quality
flags
(
see
Chapter
7)
and
METS
variables
(
see
Chapter
6).
Certain
new
variables
such
as
the
body
weight
of
the
individuals
and
daily
temperature
will
be
estimated
and
added
to
the
database
in
the
near
future.
These
data
were
not
part
of
the
original
surveys
but
can
be
estimated
or
obtained
from
other
sources
and
statistically
modified
for
use.
Place
markers
for
these
variables
are
used
in
the
first
version
of
this
database
until
the
data
estimates
are
obtained.

5.2
Platform
Selection
10
Figure
5­
1.
Information
formats
available
through
Access.
Access
was
chosen
as
the
front­
end
interface
because
it
is
a
relational
database.
A
relational
database
is
a
linked
collection
of
tables,
each
containing
data
about
one
subject.
Because
the
tables
are
related,
information
from
more
than
one
table
can
be
used
at
a
time.
A
data
value
is
stored
just
once,
and
this
reduces
required
disk
storage
space
and
reduces
time
required
for
updating
and
retrieving
data.
For
example,
personal
data
on
a
respondent
does
not
need
to
be
repeated
on
every
line
of
the
activity
diary,
because
it
is
constant
for
a
diary
day.
In
this
relational
database,
the
personal
data
file
is
linked
to
the
diary
file
via
CHADID,
which
uniquely
identifies
a
single
diaryday
(
Further
details
on
CHADIDs
are
given
in
Section
5.3.)
Files
and
reports
can
then
be
generated
from
both
the
diary
file
and
the
personal
(
questionnaire)
data
at
the
same
time.

The
tools
of
Access
can
manage
information
from
other
desktop
databases,
SQL
databases,
or
applications
such
as
spreadsheets
or
word
processors.
Access
also
(
1)
provides
front­
end
development
and
database
development;
(
2)
is
user­
friendly;
(
3)
allows
records
to
be
read,
added,
edited,
and
updated;
(
4)
prints
data
in
professional
print
formats
such
as
tables,
forms,
and
reports
(
Figure
5­
1);
(
5)
produces
charts
from
a
given
set
of
data;
(
6)
performs
complex
mathematical
calculations;
and
(
7)
is
cost­
effective.

5.3
Development
Process
The
development
process
has
occurred
in
four
phases.
The
first
three
phases
are
described
below.
(
The
fourth
phase,
focused
on
METS
data,
is
presented
in
Chapter
6.)
Again,
emphasis
was
placed
on
structuring
CHAD
so
that
both
questionnaire
and
activity
data
could
be
easily
queried.
The
phases
are
described
individually
below.

5.3.1
Phase
I:
Entering
Data
into
CHAD
Tables
The
earliest
versions
of
the
original
questionnaire
and
diary
data,
available
in
ASCII
format,
have
been
read
into
tables
contained
in
the
database.
Additional
tables
have
been
created
to
link
each
field
to
an
associated
text.
Questionnaire
and
diary
information
is
stored
in
two
different
tables
and
is
linked
through
a
unique
CHADID.
(
See
Section
5.3.3
for
further
details
on
CHADID.)
Each
field
in
a
main
table
(
e.
g.,
the
Denver
data
table,
shown
11
Figure
5­
2.
Sample
query
data
in
ASCII
code
format.
in
Figure
5­
2)
is
linked
to
associated
data
tables;
this
link
enables
access
to
textual
information
for
the
specified
ASCII
code.

5.3.2
Phase
II:
Graphical
User
Interface
Development
The
GUI
of
CHAD
allows
a
user
to
browse
through
all
records
of
a
respondent's
information.
The
user­
friendly,
state­
of­
the
art
GUI
design
includes
buttons
and
pull­
down
menus
for
customizing
queries.
A
user
may
access
diary
information
for
each
respondent
through
the
designated
GUI
screen.
Also,
a
user
may
view
explanations
of
the
data
codes
through
pull­
down
menus.

At
present,
GUIs
are
in
place
for
four
studies
(
NHAPS
A
and
B,
California
adults
and
children,
and
Denver),
and
in
the
near
future,
access
to
data
for
each
study
will
be
provided
through
individual
screens.
The
main
screen
for
each
study
depicts
information
for
one
individual
(
Figure
5­
3).
Personal
information
such
as
age,
gender,
ethnicity,
and
education
may
be
viewed
at
the
top
portion
of
the
screen.
A
user
may
view
other
questionnaire
information
by
scrolling
the
bottom
portion
of
the
screen,
and
clicking
on
the
"
Diary
Information"
button
brings
up
the
diary
information
for
that
person
(
Figure
5­
4).
Quality
flag
information
(
discussed
in
Chapter
7)
for
the
questionnaire
and
diary
information
of
that
person
may
also
be
viewed
in
the
same
manner.
12
Figure
5­
3.
Sample
screen
of
CHAD
questionnaire
data.
5.3.3
Phase
III:
Querying
CHAD
Data
A
main
menu
screen
was
created
so
that
the
original
and
CHAD­
formatted
data
could
be
accessed
from
one
screen
(
Figure
5­
5).
Data
for
all
questions
common
to
all
studies
have
been
incorporated
into
CHAD.
Available
data
from
the
studies
may
be
read
into
CHAD
tables.

Each
record
in
the
CHAD
tables
is
identified
by
a
unique
CHADID,
which
consists
of
nine
characters
(
shown
in
F
i
g
u
r
e
5
­
4
a
s
D
E
N
1
0
8
0
7
A
)
.
T
h
e
f
i
r
s
t
t
h
r
e
e
c
h
a
r
a
c
t
e
r
s
13
Figure
5­
4.
Sample
screen
of
diary
data.

Figure
5­
5.
CHAD
main
screen.
represent
the
study
source
(
e.
g.,
DEN
for
Denver).
The
next
five
characters
represent
the
person's
ID
used
in
the
14
Figure
5­
6.
Sample
screen
of
filter
options.
study,
and
the
last
character
may
be
either
A
or
B:
A
to
denote
the
first
person­
day
and
B
to
indicate
the
second
person­
day
of
multiday
studies.

Data
may
be
queried
by
clicking
the
Filter
Data
button
(
Figure
5­
3).
This
button
leads
to
a
screen
where
the
user
can
query
data
from
options
available
on
the
screen
(
Figure
5­
6).
A
user
can
save
the
selected
query
in
one
of
the
boxes
available
on
the
right
portion
of
the
screen,
which
can
then
be
accessed
lateror
used
for
batch
queries.
The
queried
data
can
be
displayed
on
the
screen,
outputted
in
the
form
of
an
ASCII
file,
or
printed
as
a
report.
A
summary
of
queried
data
can
be
viewed
by
clicking
on
the
Data
Analysis
button
on
the
Filter
Data
screen
(
Figure
5­
6).
The
Data
Analysis
screen
(
Figure
5­
7)
displays
the
number
of
respondents
belonging
to
the
selected
criteria.
This
set
of
queried
data
can
be
further
analyzed
through
various
charts
available
to
the
user.
15
Figure
5­
7.
Example
of
query
results.
16
Chapter
6
MET
Data
A
MET
is
defined
as
the
ratio
of
an
activity­
specific
metabolic
rate
to
a
person's
resting
metabolic
rate.
The
resting
metabolic
rate
can
be
estimated
from
personal
data
such
as
gender,
age,
weight,
and
occupation.
Metabolic
rates
are
important
in
the
estimation
of
intake
dose,
pharmacokinetic
modeling,
and
risk
assessment.
Appendix
G
indicates
how
MET
values
are
utilized
in
modeling
efforts
that
seek
to
quantify
the
intake
of
substances
and
relate
this
to
health
effects.

The
MET
values
for
specific
activities
are
not
always
single
point
estimates,
but
in
many
cases
are
provided
as
ranges
or
distributions.
CHAD
provides
for
the
description
of
MET
values
in
the
form
of
uniform,
exponential,
normal,
lognormal,
and
triangular
distributions,
as
well
as
point
estimates.
The
distributions
and
their
associated
parameters
are
estimated
from
data
reported
in
the
literature.
Table
H­
1
in
Appendix
H
presents
METS
distributions
and
associated
statistical
information
by
activity
code.

Figure
6­
1
shows
a
MET
Data
screen,
which
can
be
accessed
by
clicking
the
METS
Data
button
on
the
Diary
Information
Activity
Monitor
screen
(
Figure
5­
4).
The
user
can
conduct
stochastic
sampling
of
MET
values
from
the
specified
distributions,
and
the
resultant
activity­
specific
and
daily
total
energy
expenditures
will
be
calculated.
At
the
press
of
a
button,
a
new
set
of
randomly
based
MET
values
can
be
produced.
To
assist
users
who
wish
to
emphasize
large
MET
values
in
their
analyses,
CHAD
contains
a
flag
that
indicates
cases
for
which
the
median
MET
value
is
greater
than
or
equal
to
3.
17
Figure
6­
1.
Sample
screen
of
METS
data.
18
Chapter
7
Quality
Assurance
Procedures
Data
quality
flags
are
used
to
indicate
problems
with
data
content.
There
are
three
classes
of
such
problems
that
can
arise
in
the
activity
database.
The
first
is
missing
or
blank
data.
This
is
due
to
periods
that
are
not
accounted
for
(
gaps)
in
the
diary
data,
or
diaries
that
are
shorter
than
24
hours,
or
missing
questionnaire
data.
The
second
type
of
problem
occurs
if
data
are
present
but
outside
the
allowable
range.
If
"
clean"
data
sets
are
obtained,
this
type
of
error
should
occur
very
seldom.
Even
so,
such
errors
will
occasionally
be
found.
When
these
errors
are
detected,
either
they
are
corrected
(
if
possible),
or
the
values
are
deleted.
The
third
and
most
difficult
type
of
problem
to
resolve
occurs
when
the
data
are
present
and
the
values
are
within
allowable
range,
but
are
deemed
to
be
of
poor
quality
by
a
researcher
or
analyst.

In
CHAD,
missing
data
are
filled
in
with
dummy
activity
records
that
have
activity='
X'
and
location='
X'.
No
activities
have
been
extended
or
inferred
to
fill
in
these
gaps,
although
this
may
be
done
in
a
future
version
(
in
which
case
a
flag
would
be
set
to
indicate
that
the
data
were
changed).
The
second
type
of
problem
occurs
infrequently
and
can
be
eliminated
with
careful
quality
control.
The
third
type
of
problem
is
by
far
the
most
extensive,
as
there
are
literally
thousands
of
cases
of
unusual
or
unlikely
events
in
the
database.
Some
of
these
problems
can
be
identified
by
using
the
data
quality
flags
added
to
the
CHAD
database.

Interviews
with
prospective
users
made
clear
that
most
users
would
like
to
have
access
to
the
unadjusted
(
original)
data,
even
if
of
poor
quality,
with
the
option
of
deleting
or
replacing
data
according
to
their
needs.
Therefore,
quality
flags
are
available
to
identify
suspect
records,
but
it
is
up
to
the
users
to
determine
which
flags
are
important
to
them.
For
example,
some
diaries
do
not
report
any
sleep
time
or
any
meals.
These
can
be
identified
by
using
the
QCSLEEP
and
QCMEALS
flags.
The
user
must
decide
whether
or
not
the
diaries
are
acceptable
to
them.

7.1
Determination
of
Quality
Flags
In
CHAD,
there
are
two
main
types
of
quality
indicators.
Variables
whose
names
start
with
the
prefix
"
QF"
are
binary
quality
flags
of
the
pass/
fail
type
(
Table
7­
1).
These
are
set
to
"
on"
(
appearing
as
a
filled­
in
dot
in
CHAD)
to
indicate
questionable
data.
The
other
type
of
quality
indicators
use
the
prefix
"
QC".
These
variables
are
counts
(
usually
of
time),
which
indicate
how
often
within
a
diary
some
condition
was
met.
For
example,
the
variable
`
QCSLEEP'
records
the
number
of
hours
of
sleep
time
in
each
diary
day.
This
variable
can
then
be
used
to
select
diaries
that
contain
sleep
durations
within
a
specified
range.
19
Table
7­
1.
Quality
Indicator
Variables
in
CHAD
version
of
the
data
Name
File
Range
Description
QFACTLOC
Diary
0­
1
1=
Inconsistent
activity
and
location
code
pair
QFINFER
Diary
0­
1
1=
Record
changed
(
e.
g.
activity
inferred)
by
analyst
QFMETAB
Diary
0­
1
1=
High
(>=
3.0)
METS
value
QFTRAVEL
Summary
0­
1
1=
Inconsistent
morning­
evening
travel
times
QCSLEEP
Summary
0­
24
Hours
of
sleeping
during
diary
day
QCMISS
Summary
0­
1440
Minutes
of
time
missing
from
diary
QCACTLOC
Summary
0­
1440
Minutes
of
time
with
QFACTLOC=
1
QCMEALS
Summary
0+
Number
of
separate
meals
in
diary
day
QCEATIME
Summary
0­
1440
Minutes
spent
eating
in
diary
day
QCINFER
Summary
0­
1440
Minutes
of
time
with
QFINFER=
1
QCMETAB
Summary
0­
1440
Minutes
of
time
with
QFMETAB=
1
QCHEAVY
Summary
0­
1440
Minutes
of
time
with
heavy
breathing
(
HEAVYBR=
1)

QCLONG
Summary
1­
1440
Duration
(
minutes)
of
longest
activity
in
diary
day
There
are
several
additional
descriptive
variables
that
have
been
added
to
the
CHAD
database
that
can
be
used
for
quality
control
purposes,
even
though
their
names
do
not
start
with
the
letter
Q
(
Table
7­
2).
For
example,
diaries
with
a
small
number
of
records
(
i.
e.,
small
RECCOUNT)
might
be
eliminated
from
consideration
on
the
basis
that
they
do
not
provide
a
detailed
enough
description
of
daily
activities.

All
of
the
following
quality
flags
are
added
to
the
data
by
the
SAS
programs
that
convert
the
original
data
into
CHAD
form.
There
is
one
SAS
program
for
each
of
the
studies
incorporated
into
CHAD.
An
example
program
(
NHAPS.
SAS)
is
given
in
Appendix
F.
The
QFACTLOC
flag
is
set
in
the
first
data
step
(
DATA
NHAPS1),
but
most
of
the
other
flagging
actually
takes
place
in
the
DATA
SUM
step
of
the
program.

Table
7­
2.
Other
Descriptive
CHAD
Variables
Name
File
Type
Description
CHADID
Both
9­
Char
Unique
9
character
ID
identifying
each
person­
day
NDAYS
Summary
1+
Total
#
of
consecutive
diary
days
from
this
person
DAYNUM
Summary
1­
Ndays
Day
number
(
1=
first
24
hours,
2=
second
24
hrs...)

DAYTYPE
Summary
WD,
WE
WD=
Weekday,
WE=
Weekend
(
Sat
or
Sun)
20
RECCOUNT
Summary
24+
Number
of
records
in
diary
WRAPTIME
Summary
0000­
2359
Time
at
which
24­
hour
diary
day
actually
started
Sections
7.1.1 
7.1.3
described
the
quality
indicators
that
have
been
incorporated
into
CHAD.

7.1.1
Quality
Flag
(
QF)
Variables
1.
QFACTLOC 
Indicates
inconsistent
activity­
location
pairs
in
the
diary.

Each
record
in
the
diary
portion
of
the
database
contains
both
an
activity
code
and
a
location
code.
Certain
activities
are
not
compatible
with
certain
locations.
For
example,
travel
activities
(
those
with
codes
over
18000)
should
not
take
place
in
one's
home
(
location
codes
less
than
31000).
A
list
of
incompatible
combinations
was
developed
and
implemented.
Frequency
tables
of
flagged
data
were
prepared
for
certain
studies
(
Denver,
NHAPS,
and
California
adults)
and
were
used
to
revise
the
flagging
criteria.
Overall,
about
1.7%
(
11,
292)
of
the
activity
records
are
flagged.
This
still
amounts
to
several
thousand
records,
so
a
detailed
examination
of
flagged
data
will
take
time.
Since
certain
activity 
location
combinations
are
unlikely
but
not
impossible,
the
identification
of
suspect
combinations
will
always
be
partly
subjective,
and
further
changes
to
the
list
may
be
desired
in
future.
The
total
amount
of
time
that
QFACTLOC
is
"
on"
in
each
diary
is
summarized
in
the
variable
QCACTLOC.

2.
QFTRAVEL 
Flag
indicating
possible
travel
time
inconsistencies.

One
of
the
logical
constraints
on
a
valid
activity
diary
is
that
travel
to
a
remote
location
(
such
as
a
workplace)
should
be
roughly
matched
in
duration
by
a
later
return
trip.
There
are
complicating
factors
such
as
traffic
jams,
and
running
errands,
etc.,
which
may
result
in
differing
travel
times.
Another
difficulty
arises
from
simply
trying
to
identify
trips
that
qualify
as
out­
and­
back
pairs.
Without
examining
all
the
diaries
individually,
a
first
attempt
at
flagging
potential
travel
inconsistencies
was
made.
The
QFTRAVEL
variable
flags
those
person­
days
that
meet
four
conditions:
(
1)
The
person
is
employed
outside
the
home,
(
2)
the
sampled
day
is
a
weekday
(
Monday 
Friday),
(
3)
the
total
travel
time
in
the
morning
rush
hours
(
6 
9
am)
differs
by
more
than
a
factor
of
2
from
the
evening
rush
hour
(
4 
7
pm)
travel
time,
and
(
4)
either
the
morning
or
evening
travel
time
was
at
least
30
minutes.
The
travel
time
used
is
the
travel
time
to/
from
work
(
Activity=
18200,
if
recorded)
plus
the
time
for
unspecified
travel
(
Activity=
18000).
Only
the
NHAPS
and
California
studies
recorded
the
purpose
of
the
travel.
QFTRAVEL
is
on
in
1071
diaries.

Even
with
these
restrictive
conditions,
a
significant
percentage
of
diaries
are
flagged
(
roughly
20%
for
studies
with
only
general
travel,
and
5%
of
diaries
with
commuting
identified
as
such).
Most
of
these
diaries
are
likely
to
be
valid,
but
individual
examination
would
be
necessary
in
order
to
make
any
decisions.
A
refinement
of
the
flag
definition
might
be
possible
in
order
to
better
restrict
the
flagging
to
only
work
commute
trips,
but
it
is
not
clear
how
best
to
do
this
since
the
purpose
of
a
trip
must
be
inferred
from
the
diary
as
a
whole.

3.
QFINFER 
Indicates
that
a
particular
diary
record
was
inferred
(
changed
from
the
original).
21
All
data
in
CHAD
is
faithful
to
the
original
raw
data.
If
the
original
data
contained
gaps
in
time,
or
missing
activity
or
location
codes,
then
these
show
up
as
CHAD
records
with
missing
codes
(
ACTIVITY=
X
and/
or
LOCATION=
X).
At
some
point
it
may
be
desirable
to
assign
inferred
activities
in
place
of
the
missing
data.
If
this
is
done,
then
the
QFINFER
flag
will
be
set
to
indicate
this,
and
the
end
user
will
have
the
option
of
"
turning
off"
the
changes.
Similarly,
diaries
with
no
missing
time
might
be
altered
(
to
add
meals
or
sleep
time,
for
instance),
and
the
QFINFER
flag
would
then
used
to
indicate
this
condition.
QFINFER=
0
for
all
data
since
no
changes
have
yet
been
made.
The
cumulative
variable
QCINFER
indicates
the
total
time
per
diary
with
QFINFER=
1.

4.
QFMETAB 
Indicates
high
(>
3.0)
metabolic
rate
factors.

A
distribution
of
possible
METS
values
is
associated
with
each
activity.
(
See
Chapter
6
for
details.)
The
QFMETAB
flag
indicates
those
activities
for
which
the
median
value
in
the
distribution
is
greater
than
or
equal
to
3.
These
activities
will
have
an
effect
on
the
breathing
rates
of
subsequent
activities
through
the
phenomenon
of
"
excess
post­
exercise
oxygen
consumption"
(
EPOC).
In
the
current
(
December
1997)
implementation
of
CHAD,
the
METS
values
for
these
subsequent
activities
have
not
been
modified,
although
this
is
planned
for
early
1998.
The
total
amount
of
time
per
diary
with
QFMETAB
set
to
"
on"
is
recorded
in
the
count
variable
QCMETAB.
QFMETAB
is
on
in
91370
cases,
or
14.2%
of
the
records.
In
each
of
the
eight
studies
this
flag
is
on
between
12­
18%
of
the
time,
with
Valdez
having
the
highest
rate.

7.1.2
Quality
Count
(
QC)
Variables
1.
QCSLEEP 
The
amount
of
time
spent
sleeping.

The
only
two
activities
that
are
almost
essential
each
day
are
sleeping
and
eating.
For
sleeping,
the
daily
total
time
(
rounded
to
the
nearest
whole
number
of
hours)
is
recorded
in
the
variable
QCSLEEP,
which
appears
with
the
questionnaire
(
summary)
data.
This
flag
does
not
necessarily
indicate
poor
quality
data,
but
may
sometimes
just
point
to
unusual
behavior.
People
may
spend
little
or
no
time
sleeping
on
a
given
day.
Poorly
recorded
diaries
do
not
necessarily
underestimate
sleep
time.
However,
diaries
with
little
or
no
reported
sleep
time
should
probably
be
examined
more
closely
to
allow
the
analyst
to
decide
on
their
reliability.
QCSLEEP
is
equal
to
zero
in
304
diaries.
This
means
1.8%
of
the
diaries
do
not
record
sleep
time.

2.
QCMISS 
Missing
diary
time
(
less
than
24
hours).

The
CHAD
database
is
divided
into
person­
days
that
contain
exactly
24
hours
of
consecutive
diary
data.
In
some
cases,
the
original
data
did
not
span
a
full
24
hours,
or
else
it
contained
one
or
more
gaps
(
time
intervals)
with
no
recorded
activities.
The
total
amount
(
in
minutes)
of
such
time
is
recorded
in
the
QCMISS
variable,
so
that
incomplete
diaries
can
be
eliminated
by
filtering.
QCMISS
is
greater
than
zero
in
876
diaries;
5.2%
are
incomplete
(
have
less
than
24
hours).

3.
QCMEALS 
Number
of
meals
in
the
diary
day.

A
meal
is
defined
as
one
or
more
consecutive
records
of
eating
activity.
Two
meals
must
be
separated
by
at
least
one
non­
eating
activity.
Roughly
6%
of
the
diaries
do
not
report
any
eating
on
that
particular
day.
Examination
22
of
the
diaries
shows
that
in
some
cases
eating
is
not
reported
because
it
is
subsumed
under
some
other
activity
(
e.
g.,
visiting
a
friend
or
relative
for
a
very
long
period
of
time).
Eating
may
also
take
place
while
watching
TV
or
traveling,
for
example.
There
are
a
few
diaries
that
report
excessive
eating
(
more
than
six
hours
per
day).
The
diaries
are
flagged
but
are
not
excluded
because
of
unusual
eating
patterns.
The
analyst
must
decide
on
the
appropriateness
of
using
any
of
these
diaries.
QCMEALS
is
equal
to
zero
in
1684
diaries
which
is
9.9%.

4.
QCLONG 
The
duration
of
the
longest
activity
in
the
diary.

Some
diaries
report
a
single
activity
(
meaning
that
both
the
ACTIVITY
and
LOCATION
codes
do
not
change)
lasting
10
or
12
hours
or
more.
The
duration
in
minutes
of
the
longest
such
activity
is
recorded
in
the
variable
QCLONG.
Diaries
with
large
values
(
say,
>
600)
for
QCLONG
could
be
examined
to
determine
whether
or
not
the
diary
appears
complete.

The
remaining
QC
variables
provide
time
summaries
of
other
flags.
This
allows
diaries
to
be
excluded
on
the
basis
of
thresholds
of
time
flagged
in
each
category.

5.
QCACTLOC 
Total
time
(
minutes)
in
diary
with
QFACTLOC
flag
set
to
"
on".

6.
QCEATIME 
Total
time
(
minutes)
in
diary
spent
eating.

7.
QCINFER 
Total
time
(
minutes)
in
diary
with
QFINFER
set
to
"
on".

8.
QCMETAB 
Total
time
(
minutes)
in
diary
with
QFMETAB
set
to
"
on".

9.
QCHEAVY 
Total
time
(
minutes)
in
diary
with
HEAVYBR=
1
(
Heavy
breathing).

The
NHAPS
and
Cincinnati
studies
asked
the
respondents
to
identify
those
activities
accompanied
by
heavy
breathing.
The
definition
of
heavy
breathing
is
therefore
subjective
and
not
consistent
across
individuals
or
across
studies.
Nevertheless,
this
flag
is
provided
so
that
the
user
can
quickly
find
those
diaries
that
are
likely
to
represent
active
individuals.

7.1.3
Other
Variables
1.
RECCOUNT 
The
number
of
individual
activity
records
in
a
diary
day.

A
small
number
of
reported
activities
over
a
24­
hour
period
may
be
a
sign
of
poor
or
unreliable
diaries.
Some
activity
databases
delete
diaries
that
contain
fewer
than
some
specified
number
of
entries
(
records),
but
any
specific
number
of
activities
would
represent
an
arbitrary
standard.
Using
the
RECCOUNT
variable
as
a
filter,
the
CHAD
user
can
select
any
desired
minimum
number.
Note
that
all
diaries
have
at
least
24
records
because
there
is
one
or
more
for
every
hour.
(
See
Chapter
4
for
a
discussion
of
the
rationale
for
subdividing
records
at
each
hour
boundary.)

2.
WRAPTIME 
The
time
at
which
the
24­
hour
diary
actually
started.
23
For
consistency,
all
of
the
CHAD
diaries
are
organized
into
a
midnight­
to­
midnight
form.
However,
some
studies
(
e.
g.,
Denver)
did
not
have
a
standard
starting
time.
In
such
cases,
a
consecutive
24­
hour
period
was
taken,
starting
at
the
actual
start
time,
and
the
part
up
until
midnight
was
"
chopped
off"
and
added
back
on
at
the
other
end.
The
WRAPTIME
variable
indicates
if
this
has
been
done,
so
the
end
user
could
"
unwrap"
the
diary
into
its
original
form
if
desired.
WRAPTIME
indicates
the
actual
start
time
of
the
diary
in
military
time,
so
it
has
the
value
'
0000'
for
diaries
that
were
never
wrapped
(
always
midnight
to
midnight),
and
can
range
up
to
a
value
of
2359.
For
wrapped
diaries,
the
date
and
the
weekday
given
in
CHAD
represent
the
day
that
contained
more
than
12
hours
of
time.

7.2
QA
Issues
in
Coding
CHAD
Variables
QA
issues
in
coding
CHAD
variables
include
completeness
of
time
of
the
activity
diaries,
coding
of
uncertain
activities,
and
coding
of
missing
activities
or
nonsequential
activities.
These
issues
are
described
below.

7.2.1
Completeness
of
Time
of
the
Activity
Diaries
The
activity
data
in
each
study
should
reflect
activities
for
a
24­
hour
period.
However,
we
found
that
records
may
have
missing
activity
data
such
that
a
complete
24­
hour
period
was
not
represented.
Also,
in
three
major
studies
(
Denver,
Washington,
and
Cincinnati)
the
start
and
end
times
were
not
consistent
from
person
to
person.
In
CHAD,
all
diaries
are
exactly
24
hours
long
and
start
at
midnight.
As
explained
above,
the
variable
WRAPTIME
indicates
the
true
diary
starting
time.
Two
sequencing
numbers
are
provided:
RECNUM
indicates
the
new
midnight­
to­
midnight
order,
while
SEQ
indicates
the
original
chronological
sequence.

The
Denver
study
collected
two
diary
days
from
each
participant,
that
is,
48
continuous
hours.
The
two
diary
days
from
each
respondent
were
first
recombined
into
one
48­
hour
diary.
If
data
were
available
for
less
than
45
hours,
then
only
one
diary
day
was
retained.
The
number
of
consecutive
diary
days
from
the
same
respondent
is
indicated
by
the
variable
NDAYS.
In
most
case
of
the
original
data,
there
was
some
time
overlap
between
the
first
and
second
days.
This
was
due
to
the
extending
or
adding
of
records,
often
with
missing
activity
codes,
so
that
the
diary
days
either
started
or
ended
on
an
hour.
These
records
were
removed
from
the
data,
and
a
stop
time
was
set
exactly
24
hours
after
the
first
record
started.
The
second
diary
day
then
starts
immediately
and
continues
another
24
hours.
If
the
second
day
was
too
short,
then
additional
records
were
added
with
the
activity
and
location
codes
set
to
missing
(
i.
e.,
`
X').
These
missing
values
may
perhaps
be
filled
in
at
a
later
date.
The
different
diary
days
from
the
same
individual
are
indicated
in
two
ways:
(
1)
The
leftmost
eight
characters
in
the
CHADID
are
the
same,
and
the
ninth
(
rightmost)
is
A
for
the
first
diary
day,
B
for
the
second,
and
so
on;
and
(
2)
the
DAYNUM
variable
on
the
personal
summary
screen
indicates
the
day
number
numerically.

7.2.2
Coding
of
Uncertain
Activities
In
a
number
of
diary
records
in
the
target
databases,
the
applicable
activity
code
was
uncertain.
The
code
used
for
these
activities
in
the
Denver
database
is
15.
For
the
data
standardization
described
above,
the
new
code
assigned
to
these
activities
in
the
CHAD
portion
of
the
Denver
data
is
"
U".
This
will
be
done
for
all
the
databases
being
added.
Too
many
instances
of
the
15
code
would
result
in
a
record
with
a
large
amount
of
activity
time
unaccounted
for.
It
was
decided
that
records
with
more
than
eight
hours
total
(
over
two
days)
of
activity
time
24
coded
as
15
would
be
excluded
from
entry
in
CHAD.
Activities
coded
as
U
are
distinct
from
activities
not
recorded
at
all,
since
other
data
may
have
been
provided
(
such
as
the
location,
the
presence
of
smokers,
etc.).
Often,
U
signifies
a
combination
of
activities
(
such
as
eating
and
TV
watching)
for
which
the
respondent
was
uncertain
of
the
proper
coding
method.

7.2.3
Coding
of
Missing
Activities
or
Nonsequential
Activities
In
a
number
of
instances,
blank
spaces
were
found
in
some
of
the
diary
data,
indicating
missing
data.
All
such
time
gaps
were
filled
in
with
new
activities
by
using
"
X"
for
both
the
activity
and
location
code.
In
some
cases
there
were
time
overlaps,
usually
because
two
diary
days
extended
over
each
other.
In
most
cases
this
was
easy
to
resolve,
but
in
35
cases
a
specific
decision
was
made
on
the
best
method
of
removing
the
overlap.
These
cases
are
explicitly
written
in
the
SAS
program
named
DENVER.
SAS,
which
is
used
to
transform
the
raw
Denver
data
into
the
CHAD
form.
Note
that
most
studies
(
including
NHAPS
and
California)
did
not
have
any
gaps
or
overlaps
in
their
diaries.
25
Chapter
8
Discussion
8.1
California
Activity
Codes
The
version
of
the
California
adult
survey
(
1988)
that
was
obtained
by
ManTech
contains
certain
activity
codes
that
are
not
included
in
the
study
documentation.
Telephone
conversations
with
Susan
Lum
of
the
California
Air
Resources
Board
have
revealed
that
our
documentation
is
identical
to
the
"
official"
version,
but
that
our
electronic
data
are
not.
Several
draft
versions
of
the
database
were
developed
by
various
researchers,
and
evidently
we
have
one
of
these.
Our
version
nevertheless
contains
a
finer
breakdown
(
more
codes)
than
does
the
final
version.
Use
of
these
extra
codes
would
require
mapping
the
100
undocumented
codes
into
the
nearest
CHAD
codes;
some
cases
would
require
new
CHAD
codes.
These
tasks
were
beyond
the
scope
of
this
project.
An
alternative
would
be
to
request
the
official
version
of
the
California
data;
the
CHAD
user
could
then
view
the
text
description
corresponding
to
each
activity
to
determine
any
finer
breakdown
of
activity
categories.

8.2
Resolution
of
Quality
Flags
The
data
incorporated
into
CHAD
have
been
assigned
quality
flags
as
described
in
Chapter
7.
As
a
result,
several
thousand
diaries
have
been
flagged
for
one
reason
or
another.
So
far,
no
changes
have
been
made
to
the
flagged
diaries.
To
resolve
the
points
being
flagged,
an
analyst
would
have
to
examine
each
flagged
diary
in
detail
to
determine
whether
changes
should
be
made.
This
is
a
very
labor
intensive
process
since
so
many
diaries
are
involved.
For
the
first
(
beta­
test)
version
of
CHAD,
it
will
be
left
to
the
end
user
to
"
correct"
any
flagged
records
in
the
database.
A
feature
for
modifying
records
should
be
added
to
the
CHAD
user
interface
to
facilitate
this
process.

8.3
Nonrandom
Sampling
Users
may
wish
to
obtain
"
random"
activity
diaries
from
CHAD
that
meet
certain
conditions.
However,
the
respondents
sampled
in
CHAD
are
not
randomly
distributed.
Some
cities
or
states
are
oversampled
because
entire
studies
were
carried
out
in
those
places.
Other
studies
disqualified
large
groups
of
people
such
as
smokers,
or
non­
English
speakers,
or
people
without
telephones.
Many
surveys
were
age­
restricted,
or
they
preferentially
sampled
certain
target
groups
(
such
as
asthmatics
in
Cincinnati).
As
a
result,
users
should
be
cautioned
against
using
CHAD
to
draw
random
individuals
who
represent
Americans
as
a
whole.
One
solution
to
this
problem,
which
is
outside
the
scope
of
this
project,
would
be
to
create
a
set
of
statistical
weights
for
CHAD
to
reflect
the
probability
of
sampling.
26
References
Deyling,
K.
1997.
Development
of
a
Consistent
&
Comprehensive
Data
Base
for
Assessing
Multipathway/
Intake
Dose
in
Children
&
Others:
Part
I.
Consolidated
Report,
prepared
under
Work
Assignment
39
of
Contract
68­
D5­
0049
for
the
U.
S.
Environmental
Protection
Agency,
National
Exposure
Research
Laboratory,
Research
Triangle
Park,
NC.

Goldstein,
B.,
R.
Tardiff,
G.
Hoffnagle,
and
R.
Kester.
1992.
Valdez
Air
Health
Study:
Summary
Report.
Prepared
for
Alyeska
Pipeline
Service
Company,
Anchorage,
AK.

Johnson,
T.
1984.
Study
of
Personal
Exposure
to
Carbon
Monoxide
in
Denver,
Colorado.
Prepared
for
U.
S.
Environmental
Protection
Agency,
Environmental
Monitoring
Systems
Laboratory,
Research
Triangle
Park,
NC.

Johnson,
T.,
J.
Capel,
and
L.
Wijnberg.
1986.
Selected
Data
Analyses
Relating
to
Studies
of
Personal
Carbon
Monoxide
Exposure
in
Denver
and
Washington,
DC.
Prepared
for
U.
S.
Environmental
Protection
Agency,
Environmental
Monitoring
Systems
Laboratory,
Research
Triangle
Park,
NC.

Johnson,
T.
1989.
Human
Activity
Patterns
in
Cincinnati,
Ohio.
Final
Report.
Prepared
for
Electric
Power
Research
Institute,
Health
Studies
Program,
Palo
Alto,
CA.

Klepeis,
N.,
A.
Tsang,
and
J.
Behar.
1995.
Analysis
of
the
National
Human
Activity
Pattern
Survey
(
NHAPS)
Respondents
from
a
Standpoint
of
Exposure
Assessment.
Final
Report.
Prepared
for
U.
S.
Environmental
Protection
Agency,
National
Exposure
Research
Laboratory,
Las
Vegas,
NV.

Linn,
W.,
C.
Spier,
and
J.
Hackney.
1993.
Activity
Patterns
in
Ozone­
Exposed
Construction
Workers.
Journal
of
Occupational
Medicine
and
Toxicology,
2:
1.
pp.
1­
14.

McCurdy,
T.
1997a.
Work
Assignment
39
of
Contract
68­
D5­
0049.
U.
S.
Environmental
Protection
Agency,
National
Exposure
Research
Laboratory,
Research
Triangle
Park,
NC.

McCurdy,
T.
1997b.
Modeling
Multi­
route
Intake
Dose
in
an
Internally
Consistent
Manner.
U.
S.
Environmental
Protection
Agency,
National
Exposure
Research
Laboratory,
Research
Triangle
Park,
NC.
27
Paul,
R.,
J.
Yoder,
T.
Johnson,
and
J.
Capel.
1991.
Documentation
for
the
Multi­
City
Activity
Data
Base.
Prepared
for
U.
S.
Environmental
Protection
Agency,
Office
of
Air
Quality
Planning
and
Standards,
Research
Triangle
Park,
NC.

Settergren,
S.
K.,
T.
D.
Hartwell,
and
C.
A.
Clayton.
1984.
Study
of
Carbon
Monoxide
Exposure
of
Residents
of
Washington,
DC.:
Additional
Analyses.
Prepared
for
U.
S.
Environmental
Protection
Agency,
Environmental
Monitoring
Systems
Laboratory,
Research
Triangle
Park,
NC.

Shamoo,
D.,
T.
Johnson,
S.
Trim,
D.
Little,
W.
Linn,
and
J.
Hackney.
1991.
Activity
Patterns
in
a
Panel
of
Outdoor
Workers
Exposed
to
Oxidant
Pollution.
Journal
of
Exposure
Analysis
and
Environmental
Epidemiology,
1:
4.
pp.
423­
438.

Spier,
C.,
D.
Little,
S.
Trim,
T.
Johnson,
W.
Linn,
and
J.
Hackney.
1992.
Activity
Patterns
in
Elementary
and
High
School
Students
Exposed
to
Oxidant
Pollution.
Journal
of
Exposure
Analysis
and
Environmental
Epidemiology,
2:
3.
pp.
277­
293.

Wiley,
J.,
J.
Robinson,
T.
Piazza,
K.
Garrett,
K.
Cirksena,
Y.
Cheng,
and
G.
Martin.
1991.
Activity
Patterns
of
California
Residents.
Final
Report.
Prepared
for
California
Air
Resources
Board,
Research
Division,
Sacramento,
CA.

Wiley,
J.
1991.
The
Study
of
Children's
Activity
Patterns.
Final
Report.
Prepared
for
California
Air
Resources
Board,
Research
Division,
Sacramento,
CA.
28
Appendix
A
Exposure
Analysts
and
Modelers
Contacted
Table
A­
1.
Exposure
Analysts
and
Modelers
Contacted*

Contact*
Location
Area
of
Expertise
Gerry
Ackland
Research
Triangle
Institute
Denver
and
Washington
study
Maurice
Berry
NERL
Dietary
exposure
assessment
James
Capel*
Independent
Contractor
Ten
City
Study
Kenneth
Clark*
LAREI
Los
Angeles
study
Curtis
Dary
NERL
Human
activity
bibliography
George
Duggan
OAQPS
pNEM
model
Michael
Dusetzina
OAQPS
pNEM
model
Debra
Edwards
HED
Human
activity
pattern
research
William
Engelmann
Lockheed
NHAPS
Jeff
Evans
OPP
Health
effects
research
Michael
Firestone
OPPTS
Health
effects
research
Steve
Hern
NERL
THERDbase
Alan
Huber
NERL
NHAPS
Steve
Hui
CARB
Air
pollution
research
Stephanie
Irene
HED
Human
activity
pattern
research
John
Irwin
OAQPS
Multicity
study
Ted
Johnson*
Independent
Contractor
Los
Angeles
and
Valdez
Neil
Klepeis
Lockheed
NHAPS
Michael
Lebowitz
Arizona
State
University
Allergy
studies
and
NHEXAS
Michael
McCoy
Formerly
of
IT
Los
Angeles
and
Valdez
Table
A­
1.
Exposure
Analysts
and
Modelers
Contacted*

Contact*
Location
Area
of
Expertise
29
Maria
Morandi
University
of
Texas
Houston
Asthma
study
William
Nelson*
NERL
NHAPS
Al
Nielson
OPP
Health
effects
research
Joan
Novak
NERL
MODELS3
Mary
O'Rourke
Arizona
State
University
Allergy
studies
and
NHEXAS
Ted
Palma
OAQPS
Target
study
documentation
James
Quackenboss
NERL
THERDbase
Harvey
Richmond
OAQPS
pNEM
Ingrid
Sunzenauer
OPP
Health
effects
research
Amy
Vasu
OAQPS
TRIM
Karen
Whitby
OPP
Health
effects
research
Ed
Zager
OPP
Health
effects
research
Michael
Zelenka
NERL
NHAPS
*
Abbreviations
used:
HED 
Health
Effects
Division
(
OPPTS/
OPP/
HED)
NERL 
National
Exposure
Research
Laboratory
NHAPS 
National
Human
Activity
Pattern
Survey
NHEXAS 
National
Human
Exposure
Assessment
Survey
OAQPS 
Office
of
Air
Quality
Planning
and
Standards
OPP 
Office
of
Pesticides
Programs
(
OPPTS/
OPP)
OPPTS 
Office
of
Prevention,
Pesticides,
and
Toxic
Substances
pNEM 
Probabilistic
National
Ambient
Air
Quality
Standards
Exposure
Model
THERDbase 
Total
Human
Exposure
Relational
Database
and
Advanced
Simulation
Environment
TRIM 
Total
Risk
Integrated
Model
*
These
contact
persons
were
also
sources
of
data
used
for
CHAD,
as
shown
below.
James
Capel
919­
967­
8008
Denver,
Washington,
Cincinnati,
and
California
Kenneth
Clark
562­
401­
7561
Los
Angeles
Ted
Johnson
919­
929­
8266
Valdez
William
Nelson
919­
541­
3184
NHAPS
30
Appendix
B
CHAD
Activity
Codes
31
Table
B­
1.
CHAD
Activity
Codes
Code
Denver
Washington
Cincinnati
California
1988
California
1990
NHAPS
<
10>
Work
and
Other
Income
Producing
Activities
10000:
work
and
other
income
producing
activities,
general
10100:
work,
general
2,
22
10110:
work,
general,
for
organizational
activities
19
60
10111:
work
for
professional/
union
organizations
60
60
10112:
work
for
special
interest
identity
organizations
61
61
10113:
work
for
political
party
and
civic
participation
62
62
10114:
work
for
volunteer/
helping
organizations
63
63
10115:
work
of/
for
religious
groups
64
64
10116:
work
for
fraternal
organizations
66
66
10117:
work
for
child/
youth/
family
organizations
67
67
10118:
work
for
other
organizations
68
68
10120:
work,
income­
related
only
2
2
1
5
01
10130:
work,
secondary
(
income­
related)
5
05
Table
B­
1.
CHAD
Activity
Codes
Code
Denver
Washington
Cincinnati
California
1988
California
1990
NHAPS
32
10200:
unemployment
2
2
02
10300:
breaks
8
08
<
11>
Household
Activities
11000:
general
household
activities
11100:
prepare
food
3
3,
23
10
10
10
11110:
prepare
and
clean
up
food
8
11200:
indoor
chores
5,
25
11210:
clean
up
food
11
11
11
11220:
clean
house
12
12
12
11300:
outdoor
chores
6
6,
26
11
13
11310:
clean
outdoors
13
13
11400:
care
of
clothes
4
4,
24
9
14
14
14
11410:
wash
clothes
149
11500:
build
a
fire
169
169
11600:
repair,
general
11610:
repair
of
boat
166
166
11620:
paint
home/
room
167
167
11630:
repair/
maintain
car
15
15
15
Table
B­
1.
CHAD
Activity
Codes
Code
Denver
Washington
Cincinnati
California
1988
California
1990
NHAPS
33
11640:
home
repairs
16
11650:
other
repairs
16
16
11700:
care
for
plants
17
17
17
11800:
care
for
pets/
animals
18
19
18
11900:
other
household
10
19
18
19
<
12>
Child
Care
12000:
child
care,
general
12
12100:
care
of
baby
20
20
20
12200:
care
of
child
21
21
21
12300:
help/
teach
22
22
22
12400:
talk/
read
23
23
23
12500:
play
indoors
24
24
24
12600:
play
outdoors
25
25
25
12700:
medical
care­
child
26
26
26
12800:
other
child
care
27
27
27
<
13>
Obtain
Goods
and
Services
13000:
obtain
goods
and
services,
general
13100:
dry
clean
28
28
Table
B­
1.
CHAD
Activity
Codes
Code
Denver
Washington
Cincinnati
California
1988
California
1990
NHAPS
34
13200:
shop/
run
errands,
general
7
7,
27
13
301
13210:
shop
for
food
30
30
30
13220:
shop
for
clothes
or
household
goods
31
31
31
13230:
run
errands
38
38
38
13300:
obtain
personal
care
service
32
32
32
13400:
obtain
medical
service
33
33
33
13500:
obtain
government/
financial
services
34
34
34
13600:
obtain
car
service
35
35
35
13700:
other
repairs
36
36
36
13800:
other
services
37
37
37
<
14>
Personal
Needs
and
Care
14000:
personal
needs
and
care,
general
8,
28
14100:
shower,
bathe,
personal
hygiene
40
40
14110:
shower,
bathe
40
14120:
personal
hygiene
44
14200:
medical
care
41
41
41
14300:
help
and
care
42
42
42
Table
B­
1.
CHAD
Activity
Codes
Code
Denver
Washington
Cincinnati
California
1988
California
1990
NHAPS
35
14400:
eat
8
13,
33
15
43,
44,
6
43,
44,
6,
711
43
14500:
sleep
or
nap
9
11,
31
16
45,46
45,46
45
14600:
dress,
groom
47
47
47
14700:
other
personal
needs
10
17
48
<
15>
Education
and
Professional
Training
15000:
general
education
and
professional
training
12,
32
15100:
attend
full­
time
school
50
50
50
15110:
attend
day­
care
3
52
15120:
attend
K­
12
4
15130:
attend
college
or
trade
school
5
15140:
attend
adult
education
and
special
training
6
15200:
attend
other
classes
51
51
51
15300:
do
homework
7
54
54,
549
54
15400:
use
library
55
55
55
15500:
other
education
56
56
56
<
16>
Entertainment/
Social
Activities
16000:
general
entertainment/
social
activities
16100:
attend
sports
events
21
70
70
70
Table
B­
1.
CHAD
Activity
Codes
Code
Denver
Washington
Cincinnati
California
1988
California
1990
NHAPS
36
16200:
participate
in
social,
political,
or
religious
activities
11
15
16210:
practice
religion
18
65
65
16300:
view
movie
72
72
72
16400:
attend
theater
73
73
73
16500:
visit
museums
24
74
74
74
16600:
visit
26
75
75
75
16700:
attend
a
party
25
76
76
76
16800:
go
to
bar/
lounge
23
77
77
77
16900:
other
entertainment/
social
events
22
78,71
71,78
71,
78
<
17>
Leisure
17000:
leisure,
general
9,
29
17100:
participate
in
sports
and
active
leisure
14,
34
17110:
participate
in
sports
28
80
80
80
17111:
hunting,
fishing,
hiking
29
17112:
golf
801
801
17113:
bowling/
pool/
ping
pong/

pinball
802
802
17114:
yoga
803
803
Table
B­
1.
CHAD
Activity
Codes
Code
Denver
Washington
Cincinnati
California
1988
California
1990
NHAPS
37
17120:
participate
in
outdoor
leisure
81
17121:
play,
unspecified
811
17122:
passive,
sitting
81
17130:
exercise
82
17131:
walk,
bike,
or
jog
(
not
in
transit)
13
30,
31
82
82
17140:
create
art,
music,
participate
in
hobbies
32
17141:
participate
in
hobbies
83
83
83
17142:
create
domestic
crafts
84
84
84
17143:
create
art
85
85
85
17144:
perform
music/
drama/
dance
86
86
86
17150:
play
games
87
87
87
17160:
use
of
computer
88
88
17170:
participate
in
recess
and
physical
education
27
17180:
other
sports
and
active
leisure
33
17200:
participate
in
passive
leisure
17210:
watch
17211:
watch
adult
at
work
8
Table
B­
1.
CHAD
Activity
Codes
Code
Denver
Washington
Cincinnati
California
1988
California
1990
NHAPS
38
17212:
watch
someone
provide
child
care
28
17213:
watch
personal
care
48
17214:
watch
education
58
17215:
watch
organizational
activities
68
17216:
watch
recreation
88
17220:
listen
to
radio/
listen
to
recorded
music/
watch
t.
v.
35
17221:
listen
to
radio
90
90
90
17222:
listen
to
recordedmusic
92
92
92
17223:
watch
t.
v.
91,
914
91,
914,
915
91
17230:
read,
general
34
17231:
read
books
93,
939
93,
934,
937,

938
93
17232:
read
magazine/
not
ascertained
94
94,
944
94
17233:
read
newspaper
95,
954
95,
954
95
17240:
converse/
write
36
17241:
converse
96
96
96
17242:
write
for
leisure/
pleasure/

paperwork
97,
971
97,
971
97
Table
B­
1.
CHAD
Activity
Codes
Code
Denver
Washington
Cincinnati
California
1988
California
1990
NHAPS
39
17250:
think
and
relax
37
98
98
17260:
other
passive
leisure
38
17300:
other
leisure
14
98,
875,
877,

879
<
18>
Travel
18000:
travel,
general
1
1,
21
1
18100:
travel
during
work
3
3
03
18200:
travel
to/
from
work
9
9
09
18300:
travel
for
child
care
29
29
29
18400:
travel
for
goods
and
services
39
39
39
18500:
travel
for
personal
care
49
49
49
18600:
travel
for
education
59
59
59
18700:
travel
for
organizational
activity
69
69
69
18800:
travel
for
event/
social
activity
79
79
79
18900:
travel
for
leisure
18910:
travel
for
active
leisure
89
89
89
18920:
travel
for
passive
leisure
99
99
99
40
Appendix
C
CHAD
Location
Codes
41
Table
C­
1.
CHAD
Location
Codes
Code
Denver
Washington
Cincinnati
California
1988
California
1990
NHAPS
<
30>
Home
30000:
residence,
general
30010:
your
residence
30020:
other's
residence
32
30100:
residence,
indoor
2
0200
30120:
your
residence,
indoor
3
30121:
kitchen
1
1
101
30122:
living
room/
family
room
2
2
102
30123:
dining
room
3
3
103
30124:
bathroom
4
4
104
30125:
bedroom
5
5
105
30126:
study/
office
6
6
106
30127:
basement
8
8
108
30128:
utility
room/
laundry
room
9
9
110
30129:
other
indoor
30130:
other's
residence,
indoor
4
30131:
kitchen
3201
201
30132:
living
room/
family
room
3202
202
30133:
dining
room
3203
203
30134:
bathroom
3204
204
30135:
bedroom
3205
205
Table
C­
1.
CHAD
Location
Codes
Code
Denver
Washington
Cincinnati
California
1988
California
1990
NHAPS
42
30136:
study/
office
3206
206
30137:
basement
3208
208
30138:
utility
room/
laundry
room
3209
210
30139:
other
indoor
30200:
residence,
outdoor
76
0883
60
30210:
your
residence,
outdoor
30211:
pool,
spa
10
10
111
30219:
other
outdoor
11
11
112
30220:
other's
residence,
outdoor
30221:
pool,
spa
3210
211
30229:
other
outdoor
3211
212
30300:
garage
30310:
indoor
garage
38
30320:
outdoor
garage
55
30330:
your
garage
7
7
107
30331:
indoor
garage
51
0661
30332:
outdoor
garage
71
0881
30340:
other's
garage
3207
207
30341:
indoor
garage
30342:
outdoor
garage
30400:
other,
residence
13
Table
C­
1.
CHAD
Location
Codes
Code
Denver
Washington
Cincinnati
California
1988
California
1990
NHAPS
43
<
31>
Travel
31000:
travel,
general
1
0100
31100:
motorized
travel
31110:
car
1
51
51
301
31120:
truck
28
31121:
truck
(
pickup
or
van)
30
52
52
302
31122:
truck
(
other
than
pickup
or
van)
57
57
303
31130:
motorcycle/
moped/
motorized
scooter
31
60
60
304
31140:
bus
29
55
55
305
31150:
train/
subway/
rapid
transit
32
56
56,
69,
70
310
31160:
airplane
33
58
58
311
31170:
boat
312
31171:
motorized
boat
34
31172:
unmotorized
boat
35
31200:
non­
motorized
travel
31210:
walk
53
53
306
31220:
bicycle/
skateboard/

rollerskates
59
59
307
31230:
in
a
stroller
or
carried
by
an
adult
63
308
31300:
waiting
Table
C­
1.
CHAD
Location
Codes
Code
Denver
Washington
Cincinnati
California
1988
California
1990
NHAPS
44
31310:
wait
for
bus,
train,
ride
(
at
stop)
54
54
313
31320:
wait
for
travel,
indoors
314
31900:
travel,
other
61
300,
320
31910:
other
vehicle
36
<
32­
34>
Other
Indoor
32000:
other
indoor,
general
6,
50
32100:
office
building/
bank/
post
office
3
0300
5
21
21
401
32200:
industrial
plant/
factory/
warehouse
53
6
22
22
402
32300:
grocery
store/
convenience
store
4
0400
8
23
23
403
32400:
shopping
mall/
non­
grocery
store
58
0664
48
24
24
404
32500:
bar/
night
club/
bowling
alley
405
32510:
bar/
night
club
52
29
29
32520:
bowling
alley
51
32600:
repair
shop
32610:
auto
repair
shop/
gas
station
54
40
33
33
406
32620:
other
repair
shop
55
41
418
32700:
indoor
gym/
sports
or
health
club
42
31
31
407
32800:
childcare
facility
32810:
childcare
facility,
house
253
32820:
childcare
facility,
commercial
255
32900:
public
building/
library/
museum/
theater
26
26
408
Table
C­
1.
CHAD
Location
Codes
Code
Denver
Washington
Cincinnati
California
1988
California
1990
NHAPS
45
32910:
auditorium,
sport's
arena,
concert
hall
56
0662
43
32920:
library,
courtroom,
museum,

theater
61
44,
50,
45
33100:
Laundromat
409
33200:
hospital/
health
care
facility/
doctor's
office
59
0666
49
27
27
410
33300:
beauty
parlor/
barber
shop/
hair
dresser's
37
37
411
33400:
at
work:
no
specific
location,
moving
among
locations
38
412
33500:
school
60
0665
7
25
251
413
33600:
restaurant
5
0500
46
28
28
414
33700:
church
57
0663
47
30
30
415
33800:
hotel/
motel
35
35
416
33900:
dry
cleaners
36
36
417
34100:
parking
garage
52
0661
39
419
34200:
laboratory
0667
34200:
other,
indoor
(
specify)
62
0669
53,
9
39
39
400,
420
<
35­
36>
Other
Outdoor
35000:
other
outdoor,
general
8,
70
34,
11
35100:
sidewalk/
street/
neighborhood
501
35110:
within
10
yards
of
street
7
0700
10
35200:
public
garage/
parking
lot
0881
35210:
public
garage
72
56
Table
C­
1.
CHAD
Location
Codes
Code
Denver
Washington
Cincinnati
California
1988
California
1990
NHAPS
46
35220:
parking
lot
73
57
502
35300:
service
station/
gas
station
74
0885
58
503
35400:
construction
site
75
0882
59
504
35500:
amusement
park
38
35600:
school
grounds/
playground
505
35610:
school
grounds
77
61
35620:
playground
62
34
35700:
sports
stadium
and
amphitheater
78
63
506
35800:
park/
golf
course
79
0884
64
507
35810:
park
35820:
golf
course
35900:
pool,
river,
lake
508
36100:
restaurant,
picnic
510
36200:
farm
511
36300:
other
outdoor
(
specify)
0889
67,
11
40
40
500,
520
47
Appendix
D
Occupational
Codes
Used
in
CHAD
(
Based
on
U.
S.
Census
Bureau
occupational
code
categories)

005
Administrators
and
officials,
public
administration
007
Financial
managers
008
Personnel
and
labor
relations
managers
009
Purchasing
managers
013
Managers,
marketing,
advertising,
and
public
relations
014
Administrators,
education
015
Managers,
medicine
and
health
016
Managers,
properties
and
real
estate
019
Managers
and
administrators,
not
elsewhere
coded(
n.
e.
c.)
023
Accountants
and
auditors
025
Other
financial
officers
026
Management
analysts
027
Personnel,
training,
and
labor
relations
specialists
029
Buyers,
wholesale
and
retail,
except
farm
products
037
Management­
related
occupations,
n.
e.
c.
055
Electrical
and
electronic
engineers
059
Engineers,
n.
e.
c.
064
Computer
systems
analysts
ans
scientists
065
Operations
and
systems
researchers
and
analysts
067
Statistician
075
Geologists
and
geodesists
077
Agricultural
and
food
scientists
078
Biological
and
life
scientists
084
Physicians
095
Registered
nurses
096
Pharmacists
097
Dietitians
124
Political
science
teachers
137
Art,
drama,
and
music
teachers
138
Physical
education
teachers
139
Education
teachers
143
English
teachers
148
Trade
and
industrial
teachers
155
Teachers,
prekindergarden
and
kindergarden
156
Teachers,
elementary
school
157
Teachers,
secondary
school
158
Teachers,
special
education
159
Teachers,
n.
e.
c.
163
Counselors,
educational
and
vocational
164
Librarians
48
173
Urban
planners
174
Social
workers
176
Clergy
178
Lawyers
185
Designers
186
Musicians
and
composers
189
Photographers
195
Editors
and
reporters
198
Announcers
203
Clinical
lab
technologists
and
technicians
207
Licensed
practical
nurses
208
Health
technologists
and
technicians,
n.
e.
c.
213
Electrical
and
electronic
technicians
216
Engineering
technicians,
n.
e.
c.
228
Broadcast
equipment
operators
229
Computer
programmers
235
Technicians,
n.
e.
c.
243
Supervisors
and
proprietors,
sales
occupations
253
Insurance
sales
occupations
254
Real
estate
sales
occupations
255
Securities
and
financial
services
sales
occupations
257
Sales
occupations,
other
business
services
259
Sales
representatives,
mining,
manufacturing
and
wholesale
263
Sales
workers,
motor
vehicles
and
boats
268
Sales
workers,
hardware
and
building
supplies
274
Sales
workers,
other
other
commodities
275
Sales
counter
clerks
276
Cashiers
277
Street
and
door­
to­
door
sales
workers
278
News
vendors
303
Supervisors,
general
office
309
Peripheral
equipment
operators
313
Secretaries
315
Typists
319
Receptionists
323
Information
clerks,
n.
e.
c.
336
Records
clerks
337
Bookkeepers,
accounting,
and
auditing
clerks
354
Postal
clerks,
except
mail
carriers
357
Messengers
375
Insurance
adjusters,
examiners,
and
insvestigators
376
Investigators
and
adjusters,
except
insurance
377
Eligibility
clerks,
social
welfare
378
Bill
and
account
collectors
379
General
office
clerks
385
Data­
entry
keyers
406
Child
care
workers,
private
household
49
418
Police
detectives,
public
services
426
Guards
and
police,
except
public
service
427
Protective
services
occupations,
n.
e.
c.
434
Bartenders
435
Waiters
and
waitresses
436
Cooks,
except
short
order
444
Miscellaneous
food
preparation
occupations
446
Heath
aides,
except
nursing
447
Nursing
aides,
orderlies,
and
attendants
449
Maids
and
housemen
453
Janitors
and
cleaners
467
Homemaker
468
Child
care
workers
495
Forestry
workers,
except
logging
503
Supervisors,
mechanics,
and
repairers
506
Auto
mechanic
apprentices
507
Bus,
truck,
and
stationery
engine
mechanics
508
Aircraft
engine
mechanics
509
Small
engine
repairers
514
Auto
body
and
related
repairers
516
Heavy
equipment
mechanics
518
Industrial
machinery
repairers
529
Telephone
installers
and
repairers
534
Heating,
air
conditioning,
and
refregeration
mechanics
535
Camera,
watch,
and
musical
instrument
repairers
536
Locksmiths
and
safe
repairers
539
Mechanical
controls
and
valve
repairers
555
Supervisors,
electricians
and
power
transmission
installers
558
Supervisors,
construction
occupations,
n.
e.
c.
567
Carpenters
577
Electrical
power
installers
and
repairers
588
Concrete
and
terrazzo
finishers
599
Construction
trades,
n.
e.
c.
617
Mining
occupations,
n.
e.
c.
633
Supervisors,
production
occupations
637
Machinists
644
Precision
grinders,
fitters,
and
tool
sharpeners
657
Cabinet
makers
and
bench
carpenters
677
Optical
goods
workers
678
Dental
lab
and
medical
appliance
technicians
696
Stationary
engineers
734
Printing
machine
operators
736
Typesetters
and
compositors
743
Textile
cutting
machine
operators
744
Textile
sewing
machine
operators
763
Roasting
and
baking
machine
operators,
food
765
Folding
machine
operators
50
774
Photographic
process
machine
operators
777
Miscellaneous
machine
operators,
n.
e.
c.
779
Machine
operators,
not
specified
787
Hand
molding,
casting,
and
forming
occupations
796
Production
inspectors,
checkers,
and
examiners
804
Truck
drivers,
heavy
805
Truck
drivers,
light
856
Industrial
truck
and
tractor
equipment
operators
869
Construction
laborers
877
Stock
handlers
and
baggers
885
Garage
and
service
station
related
occupations
888
Hand
packers
and
packagers
889
Laborers,
except
construction
998
Not
employed
999
Retired
51
Appendix
E
CHAD
Questionnaire
Information
1.
Which
of
the
following
best
describes
your
living
quarters?

2.
How
are
your
living
quarters
heated?

3.
Which
fuel
or
energy
source
is
used
most
for
heating
in
your
living
quarters?

4.
Does
your
house
have
an
attached
garage?

5.
Do
you
use
a
gas
stove
in
your
house?

6a.
Are
you
employed
outside
the
home?

6b.
If
employed,
do
you
work
part­
time
or
full­
time?

7.
Are
you
attending
school?

8.
Do
you
have
asthma?

9.
Shave
you
been
around
pesticides
lately?

10.
Which
U.
S.
Bureau
of
Census
Occupational
Group
does
your
main
occupation
fall
under?

11.
What
is
your
household's
total
income
before
taxes?

12.
Do
you
smoke?

13.
Do
you
have
air
conditioning?

14.
State
15.
What
is
your
zipcode?

16.
County
52
Appendix
F
Program
for
Recoding
NHAPS
Variables
into
CHAD
Format
*
Program
for
reading
diary
data
and
translating
into
CHAD
form
;
*
Original
PROGRAM
NAME=
C:\
WA50\
MCCURDY\
PROGRAMS\
TESTACTIV
8­
22­
97
;
*
This
program
operates
on
the
NHAPS
data
;
*
Latest
changes
made
by
Graham
Glen
on
December
29,
1997.
;

RUN;
OPTIONS
OBS=
MAX;
LIBNAME
OUT
'
E:\
WA50\
SASDATA';
TITLE
'
NHAPS
DATA';

*
Read
questionnaire
data
(
only
the
variables
needed
for
CHAD);

DATA
QST1;
INFILE
'
D:\
NHAPS\
DATA\
MAIN.
DAT'
PAD;

LENGTH
CHADID
$
9
OCCUP
$
7
PID
$
6
COUNTY
$
20
STATE
$
14
AGEGROUP
JOBHOURS
ZIPCODE
DUMMY5
$
5
WRAPTIME
DUMMY4
$
4
AGE
WEIGHT
TEMPERAT
PREF
$
3
HOUSING
DAYTYPE
HEATING
FUEL
EDUCAT
INCOME
$
2
GARAGE
GASSTOVE
GENDER
RACE
STUDENT
ASTHMA
HEARTLUN
SMOKER
SMOKER2
PESTICID
EMPLOYED
FULLTIME
AIRCOND
$
1;

KEEP
CHADID
PID
DAYNUM
NDAYS
AGEGROUP
JOBHOURS
ZIPCODE
AGE
HOUSING
DAYTYPE
HEATING
FUEL
EDUCAT
WRAPTIME
GARAGE
GASSTOVE
GENDER
RACE
STUDENT
ASTHMA
HEARTLUN
SMOKER
SMOKER2
PESTICID
EMPLOYED
FULLTIME
OCCUP
INCOME
STATE
AIRCOND
WEIGHT
TEMPERAT
COUNTY;

INPUT
#
1
@
1
PID
$
CHAR6.
AB
11
STATEN
25­
26
#
2
SMOK2
20­
21
SMOKY
23
CIGS
25
CIGARS
27
#
3
BIRTH
47­
48
EMPL
50
UNEMP
52
WORKH
58­
59
RACEN
73
HISP
75
GRADE
77­
78
SEX
80
#
4
ASTH
10
ANGINA
12
RESP
14
ZIP
47­
51
HOUSE
56
ATTGAR
68
GSTOV
76
#
5
HEAT
10
FUELN
12
DAYTYP
42
OTHSMOK
60
#
6
PEST
22
#
7
FURNY
49
FUELY
51
WSTOVE
55
KERO
57
FIRE
61
#
20;

*
Use
prefix
NHA
for
NHAPS
Air
questionnaire,
and
NHW
for
Water
questionnaire;

DUMMY5=
10000+_
N_;
IF
(
AB=
1)
THEN
PREF='
NHA';
ELSE
IF
(
AB=
2)
THEN
PREF='
NHW';
ELSE
PREF='
NHX';
CHADID=
PREF
||
DUMMY5
||
'
A';

PID=
TRANSLATE(
PID,'
0','
');

DAYNUM=
1;
NDAYS
=
1;

IF
(
DAYTYP=
1)
THEN
DAYTYPE='
WD';
ELSE
53
IF
(
DAYTYP=
2)
THEN
DAYTYPE='
WE';
ELSE
DAYTYPE='
X'
;

*
All
NHAPS
diaries
start
at
midnight;
WRAPTIME='
0000';

IF
(
HOUSE=
1)
THEN
HOUSING='
AP';
ELSE
IF
(
HOUSE=
2)
THEN
HOUSING='
SF';
ELSE
IF
(
HOUSE=
3)
THEN
HOUSING='
MF';
ELSE
IF
(
HOUSE=
4)
THEN
HOUSING='
O'
;
ELSE
HOUSING='
X'
;

IF
(
AB=
1)
THEN
DO;
IF
(
FURNY=
1)
THEN
HEATING='
VB';
ELSE
IF
(
FUELY=
2)
THEN
HEATING='
E'
;
ELSE
IF
(
WSTOVE=
1)
THEN
HEATING='
UV';
ELSE
IF
(
KERO=
1)
THEN
HEATING='
UV';
ELSE
IF
(
FIRE=
1)
THEN
HEATING='
UV';
ELSE
HEATING='
X'
;
END;
IF
(
AB=
2)
THEN
DO;
IF
(
HEAT=
1)
THEN
HEATING='
VB';
ELSE
IF
(
HEAT=
2)
THEN
HEATING='
S'
;
ELSE
IF
(
HEAT=
3)
THEN
HEATING='
UV';
ELSE
IF
(
HEAT=
7)
THEN
DO;
IF
(
FUELN=
1)
THEN
HEATING='
UV';
ELSE
IF
(
FUELN=
2)
THEN
HEATING='
E'
;
ELSE
IF
(
FUELN=
3)
THEN
HEATING='
UV';
ELSE
IF
(
FUELN=
4)
THEN
HEATING='
UV';
ELSE
IF
(
FUELN=
5)
THEN
HEATING='
UV';
ELSE
IF
(
FUELN=
6)
THEN
HEATING='
UV';
ELSE
IF
(
FUELN=
7)
THEN
HEATING='
E'
;
ELSE
HEATING='
O'
;
END;
ELSE
HEATING='
X';
END;

IF
(
AB=
1)
THEN
DO;
IF
(
FUELY=
1)
THEN
FUEL='
G'
;
ELSE
IF
(
FUELY=
2)
THEN
FUEL='
ES';
ELSE
IF
(
FUELY=
3)
THEN
FUEL='
OK';
ELSE
IF
(
FUELY=
4)
THEN
FUEL='
C'
;
ELSE
IF
(
FUELY=
5)
THEN
FUEL='
OK';
ELSE
IF
(
FUELY=
6)
THEN
FUEL='
W'
;
ELSE
IF
(
FUELY=
7)
THEN
FUEL='
ES';
ELSE
IF
(
WSTOVE=
1)
THEN
FUEL='
W'
;
ELSE
IF
(
KERO=
1)
THEN
FUEL='
OK';
ELSE
FUEL='
X'
;
END;
IF
(
AB=
2)
THEN
DO;
IF
(
FUELN=
1)
THEN
FUEL='
G'
;
ELSE
IF
(
FUELN=
2)
THEN
FUEL='
ES';
ELSE
IF
(
FUELN=
3)
THEN
FUEL='
OK';
ELSE
IF
(
FUELN=
4)
THEN
FUEL='
OK';
ELSE
IF
(
FUELN=
5)
THEN
FUEL='
C'
;
ELSE
IF
(
FUELN=
6)
THEN
FUEL='
W'
;
ELSE
IF
(
FUELN=
7)
THEN
FUEL='
ES';
ELSE
IF
(
FUELN=
0)
THEN
FUEL='
O'
;
ELSE
FUEL='
X'
;
END;

IF
(
ATTGAR=
0)
THEN
GARAGE='
N';
ELSE
IF
(
ATTGAR=
1)
THEN
GARAGE='
Y';
ELSE
GARAGE='
X';
54
IF
(
GSTOV=
0)
THEN
GASSTOVE='
N';
ELSE
IF
(
GSTOV=
1)
THEN
GASSTOVE='
Y';
ELSE
GASSTOVE='
X';

IF
(
SEX=
1)
THEN
GENDER='
M';
ELSE
IF
(
SEX=
2)
THEN
GENDER='
F';
ELSE
GENDER='
X';

IF
(
RACEN=
1)
THEN
RACE='
W';
ELSE
IF
(
RACEN=
2)
THEN
RACE='
B';
ELSE
IF
(
RACEN=
3)
THEN
RACE='
A';
ELSE
IF
(
RACEN=
4)
THEN
RACE='
O';
ELSE
IF
(
RACEN=
5)
THEN
RACE='
H';
ELSE
IF
(
HISP=
1)
THEN
RACE='
H';
ELSE
RACE='
X';

IF
(
GRADE<
1)
THEN
EDUCAT='
N'
;
ELSE
IF
(
GRADE<
8)
THEN
EDUCAT='
SE';
ELSE
IF
(
GRADE=
8)
THEN
EDUCAT='
E'
;
ELSE
IF
(
GRADE<
12)
THEN
EDUCAT='
SH';
ELSE
IF
(
GRADE=
12)
THEN
EDUCAT='
H'
;
ELSE
IF
(
GRADE<
16)
THEN
EDUCAT='
SC';
ELSE
IF
(
GRADE=
16)
THEN
EDUCAT='
C'
;
ELSE
IF
(
GRADE=
17)
THEN
EDUCAT='
SG';
ELSE
IF
(
GRADE=
18)
THEN
EDUCAT='
G'
;
ELSE
EDUCAT='
X'
;

*
Only
the
year
of
birth
was
recorded.
Age
is
calculated
depending
on
the
;
*
quarter
in
which
the
survey
was
conducted.
;
Q=
SUBSTR(
PID,
6,1);
IF
(
BIRTH=
99)
THEN
AGE='
X';
ELSE
DO;
NAGE=(
92­
BIRTH)+
FLOOR(
Q/
4);
IF
(
NAGE>=
0)
THEN
AGE=
NAGE;
ELSE
AGE='
0';
END;

IF
(
NAGE=.)
THEN
AGEGROUP='
X'
;
ELSE
IF
(
NAGE<
5)
THEN
AGEGROUP='
0_
4'
;
ELSE
IF
(
NAGE<
15)
THEN
AGEGROUP='
5_
14'
;
ELSE
IF
(
NAGE<
19)
THEN
AGEGROUP='
15_
18';
ELSE
IF
(
NAGE<
65)
THEN
AGEGROUP='
19_
64';
ELSE
AGEGROUP='>=
65'
;

IF
(
EMPL=
1)
THEN
EMPLOYED='
Y';
ELSE
IF
(
EMPL=
2)
THEN
EMPLOYED='
Y';
ELSE
IF
(
EMPL=
3)
THEN
EMPLOYED='
N';
ELSE
EMPLOYED='
X';

IF
(
EMPL=
1)
THEN
FULLTIME='
Y';
ELSE
IF
(
EMPL=
2)
THEN
FULLTIME='
N';
ELSE
IF
(
EMPL=
3)
THEN
FULLTIME='
N';
ELSE
FULLTIME='
X';

IF
(
EMPL
=
3
)
THEN
JOBHOURS='
0'
;
ELSE
IF
(
WORKH<
10)
THEN
JOBHOURS='
0_
9'
;
ELSE
IF
(
WORKH<
20)
THEN
JOBHOURS='
10_
19';
ELSE
IF
(
WORKH<
30)
THEN
JOBHOURS='
20_
29';
ELSE
IF
(
WORKH<
40)
THEN
JOBHOURS='
30_
39';
ELSE
IF
(
WORKH<
50)
THEN
JOBHOURS='
40_
49';
ELSE
IF
(
WORKH<
80)
THEN
JOBHOURS='
50_
79';
ELSE
IF
(
WORKH<
87)
THEN
JOBHOURS='
80_'
;
ELSE
JOBHOURS='
X'
;
55
IF
(
AGEGROUP='
5_
14')
THEN
STUDENT='
Y';
ELSE
IF
(
UNEMP=
4)
THEN
STUDENT='
Y';
ELSE
IF
(
UNEMP<
8)
THEN
STUDENT='
N';
ELSE
IF
(
FULLTIME='
Y')
THEN
STUDENT='
N';
ELSE
STUDENT='
X';

OCCUP='
X';

IF
(
ASTH=
0)
THEN
ASTHMA='
N';
ELSE
IF
(
ASTH=
1)
THEN
ASTHMA='
Y';
ELSE
ASTHMA='
X';

IF
(
ASTH=
0
AND
ANGINA=
0
AND
RESP=
0)
THEN
HEARTLUN='
N';
ELSE
IF
(
ASTH=
1
OR
ANGINA=
1
OR
RESP=
1)
THEN
HEARTLUN='
Y';
ELSE
HEARTLUN='
X';

IF
(
SMOKY=
0
AND
CIGARS=
0)
THEN
SMOKER='
N';
ELSE
IF
(
SMOKY=
1
OR
CIGARS=
1)
THEN
SMOKER='
Y';
ELSE
SMOKER='
X';

IF
(
AB=
1)
THEN
DO;
IF
(
OTHSMOK=
0)
THEN
SMOKER2='
N';
ELSE
IF
(
OTHSMOK=
1)
THEN
SMOKER2='
Y';
ELSE
SMOKER2='
X';
END;
IF
(
AB=
2)
THEN
DO;
IF
(
SMOK2=
0
)
THEN
SMOKER2='
N';
ELSE
IF
(
SMOK2<
11)
THEN
SMOKER2='
Y';
ELSE
SMOKER2='
X';
END;

INCOME='
X';

IF
(
ZIP=.
)
THEN
ZIPCODE='
X';
ELSE
IF
(
ZIP=
99999)
THEN
ZIPCODE='
X';
ELSE
IF
(
ZIP>
9999
)
THEN
ZIPCODE=
ZIP;
ELSE
DO;
DUMMY4=
ZIP;
ZIPCODE='
0'
||
DUMMY4;
END;

IF
(
PEST=
0)
THEN
PESTICID='
N';
ELSE
IF
(
PEST=
1)
THEN
PESTICID='
Y';
ELSE
PESTICID='
X';

IF
(
STATEN=
1)
THEN
STATE='
Alabama'
;
ELSE
IF
(
STATEN=
3)
THEN
STATE='
Arizona'
;
ELSE
IF
(
STATEN=
4)
THEN
STATE='
Arkansas'
;
ELSE
IF
(
STATEN=
5)
THEN
STATE='
California'
;
ELSE
IF
(
STATEN=
6)
THEN
STATE='
Colorado'
;
ELSE
IF
(
STATEN=
7)
THEN
STATE='
Connecticut'
;
ELSE
IF
(
STATEN=
8)
THEN
STATE='
Delaware'
;
ELSE
IF
(
STATEN=
9)
THEN
STATE='
Dist.
Columbia';
ELSE
IF
(
STATEN=
10)
THEN
STATE='
Florida'
;
ELSE
IF
(
STATEN=
11)
THEN
STATE='
Georgia'
;
ELSE
IF
(
STATEN=
13)
THEN
STATE='
Idaho'
;
ELSE
IF
(
STATEN=
14)
THEN
STATE='
Illinois'
;
ELSE
IF
(
STATEN=
15)
THEN
STATE='
Indiana'
;
ELSE
IF
(
STATEN=
16)
THEN
STATE='
Iowa'
;
ELSE
IF
(
STATEN=
17)
THEN
STATE='
Kansas'
;
ELSE
IF
(
STATEN=
18)
THEN
STATE='
Kentucky'
;
ELSE
IF
(
STATEN=
19)
THEN
STATE='
Louisiana'
;
ELSE
IF
(
STATEN=
20)
THEN
STATE='
Maine'
;
ELSE
IF
(
STATEN=
21)
THEN
STATE='
Maryland'
;
ELSE
56
IF
(
STATEN=
22)
THEN
STATE='
Massachusetts'
;
ELSE
IF
(
STATEN=
23)
THEN
STATE='
Michigan'
;
ELSE
IF
(
STATEN=
24)
THEN
STATE='
Minnesota'
;
ELSE
IF
(
STATEN=
25)
THEN
STATE='
Mississippi'
;
ELSE
IF
(
STATEN=
26)
THEN
STATE='
Missouri'
;
ELSE
IF
(
STATEN=
27)
THEN
STATE='
Montana'
;
ELSE
IF
(
STATEN=
28)
THEN
STATE='
Nebraska'
;
ELSE
IF
(
STATEN=
29)
THEN
STATE='
Nevada'
;
ELSE
IF
(
STATEN=
30)
THEN
STATE='
New
Hampshire'
;
ELSE
IF
(
STATEN=
31)
THEN
STATE='
New
Jersey'
;
ELSE
IF
(
STATEN=
32)
THEN
STATE='
New
Mexico'
;
ELSE
IF
(
STATEN=
33)
THEN
STATE='
New
York'
;
ELSE
IF
(
STATEN=
34)
THEN
STATE='
North
Carolina';
ELSE
IF
(
STATEN=
35)
THEN
STATE='
North
Dakota'
;
ELSE
IF
(
STATEN=
36)
THEN
STATE='
Ohio'
;
ELSE
IF
(
STATEN=
37)
THEN
STATE='
Oklahoma'
;
ELSE
IF
(
STATEN=
38)
THEN
STATE='
Oregon'
;
ELSE
IF
(
STATEN=
39)
THEN
STATE='
Pennsylvania'
;
ELSE
IF
(
STATEN=
41)
THEN
STATE='
Rhode
Island'
;
ELSE
IF
(
STATEN=
42)
THEN
STATE='
South
Carolina';
ELSE
IF
(
STATEN=
43)
THEN
STATE='
South
Dakota'
;
ELSE
IF
(
STATEN=
44)
THEN
STATE='
Tennessee'
;
ELSE
IF
(
STATEN=
45)
THEN
STATE='
Texas'
;
ELSE
IF
(
STATEN=
46)
THEN
STATE='
Utah'
;
ELSE
IF
(
STATEN=
47)
THEN
STATE='
Vermont'
;
ELSE
IF
(
STATEN=
49)
THEN
STATE='
Virginia'
;
ELSE
IF
(
STATEN=
50)
THEN
STATE='
Washington'
;
ELSE
IF
(
STATEN=
51)
THEN
STATE='
West
Virginia'
;
ELSE
IF
(
STATEN=
52)
THEN
STATE='
Wisconsin'
;
ELSE
IF
(
STATEN=
53)
THEN
STATE='
Wyoming'
;
ELSE
STATE='
X'
;

AIRCOND
='
X';
WEIGHT
='
X';
TEMPERAT='
X';
COUNTY
='
X';

RUN;

PROC
SORT
DATA=
QST1;
BY
PID;
RUN;

*
Now
process
the
diary
data.;

DATA
NHAPS1;
INFILE
'
D:\
NHAPS\
DATA\
DIARY.
DAT'
PAD;

LENGTH
PID
$
6
STARTIME
ENDTIME
A4
L4
$
4
ACTTEXT
DESCRIP
$
49
ACTIVITY
LOCATION
A5
L5
$
5
WEEKDAY
A3
L3
$
3
A2
L2
$
2
GASUSE
SMOKING2
HEAVYBR
QFACTLOC
QFINFER
QFMETAB
$
1;

KEEP
PID
STARTIME
DURATION
ENDTIME
ACTIVITY
LOCATION
DESCRIP
GASUSE
SMOKING2
HEAVYBR
QFACTLOC
QFINFER
QFMETAB
MONTH
DAY
YEAR
WEEKDAY
HOUR
MINUTE
ENDHOUR
ENDMIN;

INPUT
@
1
PID
$
CHAR6.
MONTH
9­
10
DAY
11­
12
YEAR
13­
14
STARTIME
15­
18
ENDTIME
19­
22
BREATH
25
ACTTEXT
26­
74
ACTIV
75­
77
LOCATE
78­
80
SMOKING
81
DURATION
82­
85
WKDAY
87
HOUR
15­
16
MINUTE
17­
18
ENDHOUR
19­
20
ENDMIN
21­
22;

IF
YEAR=.
THEN
DELETE;
IF
MONTH=.
THEN
DELETE;
IF
DAY=.
THEN
DELETE;
IF
HOUR=.
THEN
DELETE;
57
IF
MINUTE=.
THEN
DELETE;
IF
DURATION=
0
THEN
DELETE;

*
Convert
to
4
digit
representation
for
YEAR;
YEAR=
YEAR+
1900;

IF
WKDAY=
1
THEN
WEEKDAY='
MON';
ELSE
IF
WKDAY=
2
THEN
WEEKDAY='
TUE';
ELSE
IF
WKDAY=
3
THEN
WEEKDAY='
WED';
ELSE
IF
WKDAY=
4
THEN
WEEKDAY='
THU';
ELSE
IF
WKDAY=
5
THEN
WEEKDAY='
FRI';
ELSE
IF
WKDAY=
6
THEN
WEEKDAY='
SAT';
ELSE
IF
WKDAY=
7
THEN
WEEKDAY='
SUN';
ELSE
WEEKDAY='
X';

*
Recode
activities
using
new
CHAD
codes;
IF
ACTIV=
60
THEN
ACTIVITY='
10111';
ELSE
IF
ACTIV=
61
THEN
ACTIVITY='
10112';
ELSE
IF
ACTIV=
62
THEN
ACTIVITY='
10113';
ELSE
IF
ACTIV=
63
THEN
ACTIVITY='
10114';
ELSE
IF
ACTIV=
64
THEN
ACTIVITY='
10115';
ELSE
IF
ACTIV=
66
THEN
ACTIVITY='
10116';
ELSE
IF
ACTIV=
67
THEN
ACTIVITY='
10117';
ELSE
IF
ACTIV=
68
THEN
ACTIVITY='
10118';
ELSE
IF
ACTIV=
01
THEN
ACTIVITY='
10120';
ELSE
IF
ACTIV=
05
THEN
ACTIVITY='
10120';
ELSE
IF
ACTIV=
02
THEN
ACTIVITY='
10200';
ELSE
IF
ACTIV=
08
THEN
ACTIVITY='
10300';
ELSE
IF
ACTIV=
10
THEN
ACTIVITY='
11100';
ELSE
IF
ACTIV=
11
THEN
ACTIVITY='
11210';
ELSE
IF
ACTIV=
12
THEN
ACTIVITY='
11220';
ELSE
IF
ACTIV=
13
THEN
ACTIVITY='
11300';
ELSE
IF
ACTIV=
14
THEN
ACTIVITY='
11400';
ELSE
IF
ACTIV=
15
THEN
ACTIVITY='
11630';
ELSE
IF
ACTIV=
16
THEN
ACTIVITY='
11650';
ELSE
IF
ACTIV=
17
THEN
ACTIVITY='
11700';
ELSE
IF
ACTIV=
18
THEN
ACTIVITY='
11800';
ELSE
IF
ACTIV=
19
THEN
ACTIVITY='
11900';
ELSE
IF
ACTIV=
20
THEN
ACTIVITY='
12100';
ELSE
IF
ACTIV=
21
THEN
ACTIVITY='
12200';
ELSE
IF
ACTIV=
22
THEN
ACTIVITY='
12300';
ELSE
IF
ACTIV=
23
THEN
ACTIVITY='
12400';
ELSE
IF
ACTIV=
24
THEN
ACTIVITY='
12500';
ELSE
IF
ACTIV=
25
THEN
ACTIVITY='
12600';
ELSE
IF
ACTIV=
26
THEN
ACTIVITY='
12700';
ELSE
IF
ACTIV=
27
THEN
ACTIVITY='
12800';
ELSE
IF
ACTIV=
28
THEN
ACTIVITY='
13100';
ELSE
IF
ACTIV=
30
THEN
ACTIVITY='
13210';
ELSE
IF
ACTIV=
31
THEN
ACTIVITY='
13220';
ELSE
IF
ACTIV=
38
THEN
ACTIVITY='
13230';
ELSE
IF
ACTIV=
32
THEN
ACTIVITY='
13300';
ELSE
IF
ACTIV=
33
THEN
ACTIVITY='
13400';
ELSE
IF
ACTIV=
34
THEN
ACTIVITY='
13500';
ELSE
IF
ACTIV=
35
THEN
ACTIVITY='
13600';
ELSE
IF
ACTIV=
36
THEN
ACTIVITY='
13700';
ELSE
IF
ACTIV=
37
THEN
ACTIVITY='
13800';
ELSE
IF
ACTIV=
40
THEN
ACTIVITY='
14110';
ELSE
IF
ACTIV=
44
THEN
ACTIVITY='
14120';
ELSE
IF
ACTIV=
41
THEN
ACTIVITY='
14200';
ELSE
IF
ACTIV=
42
THEN
ACTIVITY='
14300';
ELSE
IF
ACTIV=
43
THEN
ACTIVITY='
14400';
ELSE
IF
ACTIV=
45
THEN
ACTIVITY='
14500';
ELSE
IF
ACTIV=
47
THEN
ACTIVITY='
14600';
ELSE
IF
ACTIV=
48
THEN
ACTIVITY='
14700';
ELSE
58
IF
ACTIV=
50
THEN
ACTIVITY='
15100';
ELSE
IF
ACTIV=
51
THEN
ACTIVITY='
15200';
ELSE
IF
ACTIV=
54
THEN
ACTIVITY='
15300';
ELSE
IF
ACTIV=
55
THEN
ACTIVITY='
15400';
ELSE
IF
ACTIV=
56
THEN
ACTIVITY='
15500';
ELSE
IF
ACTIV=
70
THEN
ACTIVITY='
16100';
ELSE
IF
ACTIV=
65
THEN
ACTIVITY='
16210';
ELSE
IF
ACTIV=
72
THEN
ACTIVITY='
16300';
ELSE
IF
ACTIV=
73
THEN
ACTIVITY='
16400';
ELSE
IF
ACTIV=
74
THEN
ACTIVITY='
16500';
ELSE
IF
ACTIV=
75
THEN
ACTIVITY='
16600';
ELSE
IF
ACTIV=
76
THEN
ACTIVITY='
16700';
ELSE
IF
ACTIV=
77
THEN
ACTIVITY='
16800';
ELSE
IF
ACTIV=
71
THEN
ACTIVITY='
16900';
ELSE
IF
ACTIV=
78
THEN
ACTIVITY='
16900';
ELSE
IF
ACTIV=
80
THEN
ACTIVITY='
17110';
ELSE
IF
ACTIV=
81
THEN
ACTIVITY='
17122';
ELSE
IF
ACTIV=
82
THEN
ACTIVITY='
17130';
ELSE
IF
ACTIV=
83
THEN
ACTIVITY='
17141';
ELSE
IF
ACTIV=
84
THEN
ACTIVITY='
17142';
ELSE
IF
ACTIV=
85
THEN
ACTIVITY='
17143';
ELSE
IF
ACTIV=
86
THEN
ACTIVITY='
17144';
ELSE
IF
ACTIV=
87
THEN
ACTIVITY='
17150';
ELSE
IF
ACTIV=
88
THEN
ACTIVITY='
17160';
ELSE
IF
ACTIV=
90
THEN
ACTIVITY='
17221';
ELSE
IF
ACTIV=
92
THEN
ACTIVITY='
17222';
ELSE
IF
ACTIV=
91
THEN
ACTIVITY='
17223';
ELSE
IF
ACTIV=
93
THEN
ACTIVITY='
17231';
ELSE
IF
ACTIV=
94
THEN
ACTIVITY='
17232';
ELSE
IF
ACTIV=
95
THEN
ACTIVITY='
17233';
ELSE
IF
ACTIV=
96
THEN
ACTIVITY='
17241';
ELSE
IF
ACTIV=
97
THEN
ACTIVITY='
17242';
ELSE
IF
ACTIV=
98
THEN
ACTIVITY='
17250';
ELSE
IF
ACTIV=
03
THEN
ACTIVITY='
18100';
ELSE
IF
ACTIV=
09
THEN
ACTIVITY='
18200';
ELSE
IF
ACTIV=
29
THEN
ACTIVITY='
18300';
ELSE
IF
ACTIV=
39
THEN
ACTIVITY='
18400';
ELSE
IF
ACTIV=
49
THEN
ACTIVITY='
18500';
ELSE
IF
ACTIV=
59
THEN
ACTIVITY='
18600';
ELSE
IF
ACTIV=
69
THEN
ACTIVITY='
18700';
ELSE
IF
ACTIV=
79
THEN
ACTIVITY='
18800';
ELSE
IF
ACTIV=
89
THEN
ACTIVITY='
18910';
ELSE
IF
ACTIV=
99
THEN
ACTIVITY='
18920';
ELSE
ACTIVITY='
X';

*
Recode
locations
(
microenvironments)
into
CHAD
codes;
IF
LOCATE=
100
THEN
LOCATION='
30010';
ELSE
IF
LOCATE=
114
THEN
LOCATION='
30010';
ELSE
IF
LOCATE=
120
THEN
LOCATION='
30010';
ELSE
IF
LOCATE=
199
THEN
LOCATION='
30010';
ELSE
IF
LOCATE=
200
THEN
LOCATION='
30020';
ELSE
IF
LOCATE=
214
THEN
LOCATION='
30020';
ELSE
IF
LOCATE=
220
THEN
LOCATION='
30020';
ELSE
IF
LOCATE=
299
THEN
LOCATION='
30020';
ELSE
IF
LOCATE=
113
THEN
LOCATION='
30120';
ELSE
IF
LOCATE=
101
THEN
LOCATION='
30121';
ELSE
IF
LOCATE=
102
THEN
LOCATION='
30122';
ELSE
IF
LOCATE=
103
THEN
LOCATION='
30123';
ELSE
IF
LOCATE=
104
THEN
LOCATION='
30124';
ELSE
IF
LOCATE=
105
THEN
LOCATION='
30125';
ELSE
IF
LOCATE=
106
THEN
LOCATION='
30126';
ELSE
IF
LOCATE=
108
THEN
LOCATION='
30127';
ELSE
IF
LOCATE=
110
THEN
LOCATION='
30128';
ELSE
IF
LOCATE=
213
THEN
LOCATION='
30130';
ELSE
59
IF
LOCATE=
201
THEN
LOCATION='
30131';
ELSE
IF
LOCATE=
202
THEN
LOCATION='
30132';
ELSE
IF
LOCATE=
203
THEN
LOCATION='
30133';
ELSE
IF
LOCATE=
204
THEN
LOCATION='
30134';
ELSE
IF
LOCATE=
205
THEN
LOCATION='
30135';
ELSE
IF
LOCATE=
206
THEN
LOCATION='
30136';
ELSE
IF
LOCATE=
208
THEN
LOCATION='
30137';
ELSE
IF
LOCATE=
210
THEN
LOCATION='
30138';
ELSE
IF
LOCATE=
111
THEN
LOCATION='
30211';
ELSE
IF
LOCATE=
112
THEN
LOCATION='
30219';
ELSE
IF
LOCATE=
211
THEN
LOCATION='
30221';
ELSE
IF
LOCATE=
212
THEN
LOCATION='
30229';
ELSE
IF
LOCATE=
107
THEN
LOCATION='
30330';
ELSE
IF
LOCATE=
207
THEN
LOCATION='
30340';
ELSE
IF
LOCATE=
399
THEN
LOCATION='
31000';
ELSE
IF
LOCATE=
301
THEN
LOCATION='
31110';
ELSE
IF
LOCATE=
302
THEN
LOCATION='
31121';
ELSE
IF
LOCATE=
303
THEN
LOCATION='
31122';
ELSE
IF
LOCATE=
304
THEN
LOCATION='
31130';
ELSE
IF
LOCATE=
305
THEN
LOCATION='
31140';
ELSE
IF
LOCATE=
310
THEN
LOCATION='
31150';
ELSE
IF
LOCATE=
311
THEN
LOCATION='
31160';
ELSE
IF
LOCATE=
312
THEN
LOCATION='
31170';
ELSE
IF
LOCATE=
306
THEN
LOCATION='
31210';
ELSE
IF
LOCATE=
307
THEN
LOCATION='
31220';
ELSE
IF
LOCATE=
308
THEN
LOCATION='
31230';
ELSE
IF
LOCATE=
313
THEN
LOCATION='
31310';
ELSE
IF
LOCATE=
314
THEN
LOCATION='
31320';
ELSE
IF
LOCATE=
300
THEN
LOCATION='
31900';
ELSE
IF
LOCATE=
320
THEN
LOCATION='
31900';
ELSE
IF
LOCATE=
499
THEN
LOCATION='
32000';
ELSE
IF
LOCATE=
401
THEN
LOCATION='
32100';
ELSE
IF
LOCATE=
402
THEN
LOCATION='
32200';
ELSE
IF
LOCATE=
403
THEN
LOCATION='
32300';
ELSE
IF
LOCATE=
404
THEN
LOCATION='
32400';
ELSE
IF
LOCATE=
405
THEN
LOCATION='
32500';
ELSE
IF
LOCATE=
406
THEN
LOCATION='
32610';
ELSE
IF
LOCATE=
418
THEN
LOCATION='
32620';
ELSE
IF
LOCATE=
407
THEN
LOCATION='
32700';
ELSE
IF
LOCATE=
408
THEN
LOCATION='
32900';
ELSE
IF
LOCATE=
409
THEN
LOCATION='
33100';
ELSE
IF
LOCATE=
410
THEN
LOCATION='
33200';
ELSE
IF
LOCATE=
411
THEN
LOCATION='
33300';
ELSE
IF
LOCATE=
412
THEN
LOCATION='
33400';
ELSE
IF
LOCATE=
413
THEN
LOCATION='
33500';
ELSE
IF
LOCATE=
414
THEN
LOCATION='
33600';
ELSE
IF
LOCATE=
415
THEN
LOCATION='
33700';
ELSE
IF
LOCATE=
416
THEN
LOCATION='
33800';
ELSE
IF
LOCATE=
417
THEN
LOCATION='
33900';
ELSE
IF
LOCATE=
419
THEN
LOCATION='
34100';
ELSE
IF
LOCATE=
400
THEN
LOCATION='
34200';
ELSE
IF
LOCATE=
420
THEN
LOCATION='
34200';
ELSE
IF
LOCATE=
599
THEN
LOCATION='
35000';
ELSE
IF
LOCATE=
501
THEN
LOCATION='
35100';
ELSE
IF
LOCATE=
502
THEN
LOCATION='
35220';
ELSE
IF
LOCATE=
503
THEN
LOCATION='
35300';
ELSE
IF
LOCATE=
504
THEN
LOCATION='
35400';
ELSE
IF
LOCATE=
505
THEN
LOCATION='
35600';
ELSE
IF
LOCATE=
506
THEN
LOCATION='
35700';
ELSE
IF
LOCATE=
507
THEN
LOCATION='
35800';
ELSE
IF
LOCATE=
508
THEN
LOCATION='
35900';
ELSE
IF
LOCATE=
510
THEN
LOCATION='
36100';
ELSE
IF
LOCATE=
511
THEN
LOCATION='
36200';
ELSE
IF
LOCATE=
500
THEN
LOCATION='
36300';
ELSE
60
IF
LOCATE=
520
THEN
LOCATION='
36300';
ELSE
LOCATION='
X';

*
Create
substrings
of
activity
and
location
codes.
For
example,
A2
is
the
;
*
first
(
leftmost)
two
digits
of
the
activity
code.
If
A2='
18'
then
the
;
*
activity
is
some
form
of
travel.
The
coding
system
is
tree­
based
so
that
;
*
this
type
of
grouping
becomes
fairly
natural.
;

A5=
ACTIVITY;
L5=
LOCATION;
A4=
SUBSTR(
A5,1,4);
L4=
SUBSTR(
L5,1,4);
A3=
SUBSTR(
A4,1,3);
L3=
SUBSTR(
L4,1,3);
A2=
SUBSTR(
A3,1,2);
L2=
SUBSTR(
L3,1,2);

*
Set
the
quality
flag
QFACTLOC
to
identify
suspect
activity­
location
pairs;
QFACTLOC='
0';
IF
NOT
(
A2='
X'
OR
L2='
X')
THEN
DO;

IF
(
A3='
110'
&
L2>'
30')
THEN
QFACTLOC='
1';
IF
(
A3='
111'
&
(
L5='
30124'
OR
L5='
30134'))
THEN
QFACTLOC='
1';
IF
(
A3='
112'
&
(
L2>'
30'
OR
L3='
302'))
THEN
QFACTLOC='
1';
IF
(
A3='
113'
&
L3='
301')
THEN
QFACTLOC='
1';
IF
(
A3='
113'
&
L2>'
30'
&
L3
NE
'
351')
THEN
QFACTLOC='
1';
IF
(
A3='
119'
&
L2>'
30'
&
L3
NE
'
351'
&
L3
NE
'
311')
THEN
QFACTLOC='
1';
IF
(
A3='
125'
&
(
L3='
302'
OR
L2='
35'))
THEN
QFACTLOC='
1';
IF
(
A3='
126'
&
(
L3='
301'
OR
L2='
32'))
THEN
QFACTLOC='
1';
IF
(
A3='
131'
&
L3
NE
'
339')
THEN
QFACTLOC='
1';
IF
(
A3='
134'
&
L3
NE
'
332')
THEN
QFACTLOC='
1';
IF
(
A3='
135'
&
NOT
(
L3
IN
('
321','
329','
311','
324')))
THEN
QFACTLOC='
1';
IF
(
A3='
136'
&
L4
NE
'
3261'
&
L3
NE
'
353')
THEN
QFACTLOC='
1';
IF
(
A3='
1410'
&
L2>'
30'
&
NOT
(
L3
IN
('
327','
338','
359')))
THEN
QFACTLOC='
1';
IF
(
A4='
1410'
&
L2='
30'
&
NOT
(
L5
IN
('
30124','
30125','
30134')))
THEN
QFACTLOC='
1';
IF
(
A3='
1411'
&
L2>'
30'
&
NOT
(
L3
IN
('
327','
338','
359')))
THEN
QFACTLOC='
1';
IF
(
A4='
1411'
&
L2='
30'
&
NOT
(
L5
IN
('
30124','
30125','
30134')))
THEN
QFACTLOC='
1';
IF
(
A4='
1510'
&
NOT
(
L3
IN
('
335','
320','
329','
301','
356')))
THEN
QFACTLOC='
1';
IF
(
A4='
1510'
&
L3='
301'
&
L4
NE
'
3013')
THEN
QFACTLOC='
1';
IF
(
A4='
1511'
&
NOT
(
L3
IN
('
328','
300','
301')))
THEN
QFACTLOC='
1';
IF
(
A4='
1512'
&
NOT
(
L3
IN
('
335','
356','
329')))
THEN
QFACTLOC='
1';
IF
(
A4='
1513'
&
NOT
(
L3
IN
('
335','
320','
329')))
THEN
QFACTLOC='
1';
IF
(
A4='
1514'
&
NOT
(
L3
IN
('
335','
320','
329')))
THEN
QFACTLOC='
1';
IF
(
A3='
154'
&
NOT
(
L3
IN
('
320','
335','
329')))
THEN
QFACTLOC='
1';
IF
(
A2='
16'
&
L2='
31')
THEN
QFACTLOC='
1';
IF
(
A3='
161'
&
L3<'
325'
&
L3
NE
'
320')
THEN
QFACTLOC='
1';
IF
(
A3='
161'
&
(
L3='
326'
OR
L3='
328'))
THEN
QFACTLOC='
1';
IF
(
A3='
161'
&
L2='
33'
&
L3
NE
'
335')
THEN
QFACTLOC='
1';
IF
(
A3='
161'
&
L2='
35'
&
L3<'
354')
THEN
QFACTLOC='
1';
IF
(
A3='
164'
&
NOT
(
L3='
329'
OR
L3='
320'))
THEN
QFACTLOC='
1';
IF
(
A3='
165'
&
NOT
(
L3='
329'
OR
L3='
320'))
THEN
QFACTLOC='
1';
IF
(
A3='
168'
&
NOT
(
L3
IN
('
320','
325','
336')))
THEN
QFACTLOC='
1';
IF
(
A2='
17'
&
A4
NE
'
1772'
&
L2='
31'
&
L3
NE
'
312')
THEN
QFACTLOC='
1';
IF
(
A3='
171'
&
L2='
31'
&
L3
NE
'
312')
THEN
QFACTLOC='
1';
IF
(
A5='
17111'
&
NOT
(
L3
IN
('
350','
359','
363')))
THEN
QFACTLOC='
1';
IF
(
A5='
17112'
&
NOT
(
L3='
350'
OR
L3='
358'))
THEN
QFACTLOC='
1';
IF
(
A5='
17113'
&
(
L2='
31'
OR
L2>='
33'))
THEN
QFACTLOC='
1';
IF
(
A5='
17113'
&
L2='
32'
&
NOT
(
L3
IN
('
320','
325','
327')))
THEN
QFACTLOC='
1';
IF
(
A4='
1712'
&
NOT
(
L3='
302'
OR
L2='
35'))
THEN
QFACTLOC='
1';
IF
(
A5='
17131'
&
L2<'
32'
&
L3
NE
'
302')
THEN
QFACTLOC='
1';
IF
(
A5='
17131'
&
L2='
32'
&
NOT
(
L3='
327'
OR
L3='
320'))
THEN
QFACTLOC='
1';
IF
(
A4='
1717'
&
NOT
(
L3
IN
('
320','
335','
350','
356')))
THEN
QFACTLOC='
1';
IF
(
A4='
1718'
&
L2='
30'
&
NOT
(
L3='
302'
OR
L3='
301'))
THEN
QFACTLOC='
1';
IF
(
A4='
1718'
&
L2='
32'
&
NOT
(
L3='
327'
OR
L3='
320'))
THEN
QFACTLOC='
1';
IF
(
A5='
17223'
&
(
L3='
302'
OR
L2='
35'))
THEN
QFACTLOC='
1';
IF
(
A2='
18'
&
L2
NE
'
31'
&
L5
NE
'
35110')
THEN
QFACTLOC='
1';
END;
61
*
Remove
extraneous
characters
from
the
text
description
of
the
activity;
DESCRIP=
COMPRESS(
ACTTEXT,'
0123456789');

IF
(
SMOKING>
0
AND
SMOKING<
8)
THEN
SMOKING2='
Y';
ELSE
IF
(
SMOKING=
0)
THEN
SMOKING2='
N';
ELSE
SMOKING2='
X';

IF
(
BREATH=
1)
THEN
HEAVYBR='
Y';
ELSE
IF
(
BREATH=
0)
THEN
HEAVYBR='
N';
ELSE
HEAVYBR='
X';

GASUSE
='
X';
QFINFER='
0';
QFMETAB='
0';

RUN;

PROC
SORT
DATA=
NHAPS1;
BY
PID;
RUN;

DATA
OCCUPS;
SET
QST1;
BY
PID;
KEEP
PID
OCCUP;
RUN;

DATA
NHAPS1A;
MERGE
NHAPS1(
IN=
IN1)
OCCUPS
(
IN=
IN2);
BY
PID;
IF
(
IN1=
0)
THEN
DELETE;
RUN;

PROC
SORT
DATA=
NHAPS1A;
BY
PID;
RUN;

*
Check
to
see
if
all
diary
days
have
exactly
1440
minutes;

PROC
MEANS
DATA=
NHAPS1
NOPRINT;
BY
PID;
VAR
DURATION;
OUTPUT
OUT=
DURS
SUM=
DURSUM;
RUN;

PROC
MEANS
DATA=
DURS;
VAR
DURSUM;
RUN;

*
Output
records
that
do
not
meet
standard
assumptions
about
time;

DATA
BADRECS;
SET
NHAPS1;
BY
PID;
LENGTH
LASTEND
$
4;
RETAIN
LASTEND
'
0000';
IF
(
ACTIVITY='
X'
OR
LOCATION='
X')
THEN
OUTPUT;
IF
(
FIRST.
PID
AND
STARTIME
NE
'
0000')
THEN
OUTPUT;
62
IF
(
FIRST.
PID)
THEN
LASTEND=
STARTIME;
IF
(
STARTIME
NE
LASTEND)
THEN
OUTPUT;
LASTEND=
ENDTIME;
IF
(
HOUR<
0
OR
HOUR>
23
OR
MINUTE<
0
OR
MINUTE>
59)
THEN
OUTPUT;
IF
((
60*
HOUR+
MINUTE+
DURATION)
NE
(
60*
ENDHOUR+
ENDMIN))
THEN
OUTPUT;
RUN;

*
Break
activities
at
hour
boundaries;

DATA
NHAPS2;
SET
NHAPS1A;
BY
PID;
LENGTH
ENDT
$
4
ENDHR
ENDMN
$
2
DUMMY1
QFMETAB
$
1;
DROP
ENDH
ENDM
DUMMY1
ENDHR
ENDMN
ENDT;
RETAIN
RECNUM
0;

QFMETAB='
0';
IF
ACTIVITY
IN
('
11000','
11200','
11220','
11300','
11310','
11600','
11610',
'
11620','
11630','
11640','
11650','
11700','
11800','
11900','
12000',
'
12100','
12200','
12600','
12700','
12800','
13000','
13100','
13200',
'
13210','
13220','
13230','
13300','
13400','
13500','
13600','
13700',
'
13800','
14110','
14300','
16700','
16800','
16900','
17113','
17114',
'
17170','
17100','
17110','
17111','
17112','
17120','
17121','
17130',
'
17131','
17140','
17144','
17180')
THEN
QFMETAB='
1';
IF
(
ACTIVITY>='
10000'
&
ACTIVITY<'
10200')
THEN
IF
OCCUP
IN
('
PROTECT','
FARM','
PREC','
MACH','
TRANS','
TECH')
THEN
QFMETAB='
1';

IF
(
FIRST.
PID)
THEN
RECNUM=
0;
chop:
IF
(
ENDHOUR=
HOUR
OR
(
ENDHOUR=
HOUR+
1
AND
ENDMIN=
0))
THEN
DO;
RECNUM=
RECNUM+
1;
OUTPUT;
END;
ELSE
DO;
ENDH=
ENDHOUR;
ENDM=
ENDMIN;
ENDT=
ENDTIME;
ENDHOUR=
HOUR+
1;
ENDMIN=
0;
DURATION=
60­
MINUTE;
IF
(
ENDMIN>
9)
THEN
ENDMN=
ENDMIN;
ELSE
DO;
DUMMY1=
ENDMIN;
ENDMN='
0'
||
DUMMY1;
END;
IF
(
ENDHOUR>
9)
THEN
ENDHR=
ENDHOUR;
ELSE
DO;
DUMMY1=
ENDHOUR;
ENDHR='
0'
||
DUMMY1;
END;
ENDTIME=
ENDHR
||
ENDMN;
RECNUM=
RECNUM+
1;
OUTPUT;

STARTIME=
ENDTIME;
63
HOUR=
ENDHOUR;
MINUTE=
ENDMIN;
DURATION=
60*(
ENDH­
HOUR)+(
ENDM­
MINUTE);
ENDHOUR=
ENDH;
ENDMIN=
ENDM;
ENDTIME=
ENDT;
GO
TO
CHOP;
END;

RUN;

PROC
SORT
DATA=
NHAPS2;
BY
PID
RECNUM;
RUN;

*
Check
the
new
data
set
(
with
breaks
at
each
hour)
to
ensure
that
no
new
;
*
problems
have
arisen.
;

DATA
BADRECS;
SET
NHAPS2;
BY
PID;
LENGTH
LASTEND
$
4;
RETAIN
LASTEND
'
0000';
IF
(
ACTIVITY='
X'
OR
LOCATION='
X')
THEN
OUTPUT;
IF
(
FIRST.
PID
AND
STARTIME
NE
'
0000')
THEN
OUTPUT;
IF
(
FIRST.
PID)
THEN
LASTEND=
STARTIME;
IF
(
STARTIME
NE
LASTEND)
THEN
OUTPUT;
LASTEND=
ENDTIME;
IF
(
HOUR<
0
OR
HOUR>
23
OR
MINUTE<
0
OR
MINUTE>
59)
THEN
OUTPUT;
IF
((
60*
HOUR+
MINUTE+
DURATION)
NE
(
60*
ENDHOUR+
ENDMIN))
THEN
OUTPUT;
IF
(
DURATION>
60)
THEN
OUTPUT;
IF
(
ENDHOUR
NE
HOUR)
THEN
DO;
IF
((
ENDHOUR
NE
HOUR+
1)
AND
(
ENDMIN
NE
0))
THEN
OUTPUT;
END;
RUN;

*
Sum
information
across
each
diary
day
to
get
daily
totals;

DATA
SUM;
SET
NHAPS2;
BY
PID;

LENGTH
LASTACT
LASTLOC
$
5;

KEEP
PID
RECCOUNT
QCACTLOC
QCHEAVY
QCINFER
QCMETAB
QCMISS
QCSLEEP
TRAVMOR
TRAVEVE
QFTRAVEL
YEAR
MONTH
DAY
WEEKDAY
QCMEALS
QCEATIME
QCLONG;

RETAIN
QCACTLOC
QCHEAVY
QCINFER
QCMETAB
QCMISS
QCSLEEP
TRAVMOR
TRAVEVE
RECCOUNT
QCMEALS
QCEATIME
LASTEAT
QCLONG
CURDUR
0
LASTACT
LASTLOC;

IF
(
FIRST.
PID)
THEN
DO;
RECCOUNT
=
0;
QCACTLOC=
0;
QCHEAVY
=
0;
QCINFER
=
0;
64
QCMETAB
=
0;
QCMISS
=
0;
QCSLEEP
=
0;
TRAVMOR
=
0;
TRAVEVE
=
0;
QCEATIME=
0;
QCMEALS
=
0;
LASTEAT
=
0;
QCLONG
=
0;
CURDUR
=
0;
LASTACT
='
0';
LASTLOC
='
0';
END;
RECCOUNT=
RECCOUNT+
1;
IF
(
QFACTLOC='
1')
THEN
QCACTLOC=
QCACTLOC+
DURATION;
IF
(
HEAVYBR
='
Y')
THEN
QCHEAVY
=
QCHEAVY
+
DURATION;
IF
(
QFINFER
='
1')
THEN
QCINFER
=
QCINFER
+
DURATION;
IF
(
QFMETAB
='
1')
THEN
QCMETAB
=
QCMETAB
+
DURATION;
IF
(
ACTIVITY='
14500')
THEN
QCSLEEP
=
QCSLEEP
+
DURATION;
IF
(
ACTIVITY='
X'
OR
LOCATION='
X')
THEN
QCMISS
=
QCMISS
+
DURATION;
IF
(
ACTIVITY='
14400')
THEN
DO;
QCEATIME=
QCEATIME+
DURATION;
IF
(
LASTEAT=
0)
THEN
QCMEALS=
QCMEALS+
1;
LASTEAT=
1;
END;
ELSE
LASTEAT=
0;
IF
(
ACTIVITY='
18000'
OR
ACTIVITY='
18200')
THEN
DO;
IF
(
HOUR
IN
(
6,7,8))
THEN
TRAVMOR
=
TRAVMOR
+
DURATION;
IF
(
HOUR
IN
(
16,17,18))
THEN
TRAVEVE
=
TRAVEVE
+
DURATION;
END;
IF
((
ACTIVITY
NE
LASTACT)
OR
(
LOCATION
NE
LASTLOC))
THEN
CURDUR=
0;
CURDUR=
CURDUR+
DURATION;
IF
(
QCLONG<
CURDUR)
THEN
QCLONG=
CURDUR;
LASTACT=
ACTIVITY;
LASTLOC=
LOCATION;

IF
(
LAST.
PID)
THEN
DO;
QCSLEEP=
FLOOR((
QCSLEEP+
30)/
60);
QFTRAVEL=
0;
IF
((
TRAVMOR>
2*
TRAVEVE)
&
(
TRAVMOR>
30))
THEN
QFTRAVEL=
1;
IF
((
TRAVEVE>
2*
TRAVMOR)
&
(
TRAVEVE>
30))
THEN
QFTRAVEL=
1;
IF
(
WEEKDAY='
SAT'
OR
WEEKDAY='
SUN')
THEN
QFTRAVEL=
0;
OUTPUT;
END;
RUN;

PROC
SORT
DATA=
SUM;
BY
PID;
RUN;

PROC
CONTENTS
DATA=
SUM;
RUN;
PROC
CONTENTS
DATA=
QST1;
RUN;

*
Merge
the
data
from
the
NHAPS
questionnaire
with
the
summary
data
calculated
;
*
from
the
diaries.
;

DATA
QST2;
65
MERGE
SUM(
IN=
IN1)
QST1(
IN=
IN2);
BY
PID;
LENGTH
NEWTYPE
$
2;
DROP
NEWTYPE;
IF
(
IN1=
0)
THEN
DELETE;
IF
(
IN2=
0)
THEN
DELETE;
NEWTYPE='
WD';
IF
(
WEEKDAY='
SAT'
OR
WEEKDAY='
SUN')
THEN
NEWTYPE='
WE';
IF
(
NEWTYPE
NE
DAYTYPE)
THEN
PUT
'
BAD
DAYTYPE
'
PID
DAYTYPE
NEWTYPE;
RUN;

*
The
CHADID
was
determined
from
the
questionnaire
data
(
depending
on
which
;
*
type
(
air
or
water)
was
administered.
However,
the
CHADID
must
be
written
;
*
back
onto
the
diaries.
;

DATA
CHADID;
SET
QST1;
KEEP
CHADID
PID;
RUN;

PROC
SORT
DATA=
CHADID;
BY
PID;
RUN;

*
The
CHAD
version
of
the
diaries
is
stored
in
the
SAS
dataset
NHPDIARY
;
*
and
is
also
written
to
an
ASCII
file
for
input
into
the
database.
;

DATA
OUT.
NHPDIARY;
MERGE
NHAPS2
CHADID;
BY
PID;
DROP
HOUR
MINUTE
ENDHOUR
ENDMIN
YEAR
MONTH
DAY
WEEKDAY;

FILE
'\\
METISRV0\
AEAR\
WA50\
CHADDATA\
NHPDIARY.
TXT';

SEQ=
RECNUM;
IF
(
ENDTIME='
2400')
THEN
ENDTIME='
0000';

PUT
CHADID
1­
9
RECNUM
11­
13
SEQ
15­
17
STARTIME
19­
22
DURATION
24­
25
ENDTIME
27­
30
ACTIVITY
32­
36
LOCATION
38­
42
GASUSE
44
SMOKING2
46
QFACTLOC
48
QFINFER
50
QFMETAB
52
HEAVYBR
54
DESCRIP
56­
104;

RUN;

*
The
CHAD
questionnaire
data
is
written
in
both
SAS
and
ASCII
form
to
two
files
;
*
named
NHPSURV
(
one
in
SASDATA
directory
and
the
other
in
CHADDATA).
;

DATA
OUT.
NHPSURV;
SET
QST2;

FILE
'\\
METISRV0\
AEAR\
WA50\
CHADDATA\
NHPSURV.
TXT';
66
PUT
CHADID
1­
9
PID
15­
21
DAYNUM
23
NDAYS
25
YEAR
27­
30
MONTH
32­
33
DAY
35­
36
WEEKDAY
38­
40
RECCOUNT
42­
44
WRAPTIME
46­
49
QCSLEEP
51­
52
QFTRAVEL
54
QCMISS
56­
59
HOUSING
61­
62
HEATING
64­
65
FUEL
67­
68
GARAGE
70
GASSTOVE
72
GENDER
74
RACE
76
EDUCAT
78­
80
AGE
82­
84
AGEGROUP
86­
90
EMPLOYED
92
FULLTIME
94­
95
JOBHOURS
97­
101
STUDENT
103
ASTHMA
105
SMOKER
107
SMOKER2
109
INCOME
111­
112
ZIPCODE
114­
118
PESTICID
120
AIRCOND
122
HEARTLUN
124
WEIGHT
130­
132
TEMPERAT
134­
136
DAYTYPE
138­
139
OCCUP
141­
147
QCHEAVY
149­
152
QCINFER
154­
157
QCMETAB
159­
162
QCACTLOC
164­
167
STATE
169­
182
COUNTY
184­
203
QCMEALS
205­
206
QCEATIME
208­
211
QCLONG
213­
216;

RUN;

PROC
FREQ
DATA=
NHAPS2;
TABLES
QFACTLOC
QFMETAB
QFINFER;
RUN;

DATA
STATS;
SET
QST2;
KEEP
QCA
QCMT
QCI
QFT
QCS
QCML
QCE
QCH
QCMS;
IF
(
QCACTLOC>
0)
THEN
QCA
=
1;
ELSE
QCA
=
0;
IF
(
QCMETAB
>
0)
THEN
QCMT=
1;
ELSE
QCMT=
0;
IF
(
QCINFER
>
0)
THEN
QCI
=
1;
ELSE
QCI
=
0;
IF
(
QCSLEEP
>
0)
THEN
QCS
=
1;
ELSE
QCS
=
0;
IF
(
QCMEALS
>
0)
THEN
QCML=
1;
ELSE
QCML=
0;
IF
(
QCEATIME>
0)
THEN
QCE
=
1;
ELSE
QCE
=
0;
IF
(
QCHEAVY
>
0)
THEN
QCH
=
1;
ELSE
QCH
=
0;
IF
(
QCMISS
>
0)
THEN
QCMS=
1;
ELSE
QCMS=
0;
QFT=
QFTRAVEL;
RUN;

PROC
FREQ
DATA=
STATS;
TABLES
QCA
QCMT
QCI
QFT
QCS
QCML
QCE
QCH
QCMS;
RUN;

*
End
of
processing
of
NHAPS
data;
67
Appendix
G
Modeling
Multi­
Route
Intake
Dose
in
an
Internally
Consistent
Manner
C:\
1MCCURDY\
PAPERS\
BMR.
CONS
T.
McCurdy:
October
23,
1997
MODELING
MULTI­
ROUTE
INTAKE
DOSE
IN
AN
INTERNALLY
CONSISTENT
MANNER
Fundamental
Approach
The
focus
of
a
risk
assessment
to
environmental
contaminants
should
be
an
individual,
since
each
individual
differs
widely
in
his
or
her
activities,
health
status,
metabolism,
distribution,
storage,
and
elimination
of
xenobiotic
substances.
Human
health
effects
arise
when
the
amount
and
pattern
of
dose
received
at
a
target
organ­­
the
dose
rate
over
time­­
exceed's
that
organ's
ability
to
"
defend"
itself
during
some
defined
time
period.
In
general,
timeweighted
dose
metrics
do
not
provide
relevant
information
necessary
to
understand
or
predict
these
effects.

The
amount
and
pattern
of
dose
received
at
a
target
organ
is
related
in
a
complex
biochemical
manner
to
the
amount/
pattern
of
dose
taken
into
the
body
(
intake
dose)
via
the
respiratory
tract,
the
gastrointestinal
system,
and
through
the
skin.
Thus,
in
a
total
human
health
risk
assessment,
it
is
essential
to
understand
the
intake
dose
pattern
associated
with
breathing,
ingestion­­
both
food
and
liquids­­
and
dermal
absorption.
The
various
media­­
air,
liquids,
and
physical
surfaces­­
interact
with
the
main
entry
pathways
into
the
body
so
it
is
necessary
to
assess
multi­
media,
multi­
pathway
"
total"
exposure
in
order
to
fully
understand
human
health
impact
associated
with
environmental
pollutants.

Understanding
the
time
pattern
of
dose
received
implies
that
the
time
series
of
dose
and
exposure
be
evaluated.
This
leads
directly
to
an
activity­
specific
approach
that
"
tracks"
individuals
in
time
and
space
as
they
go
through
their
daily
activities.
Since
mediaspecific
exposure
and
dose
assessments
do
not
maintain
the
correlated
nature
of
an
individual's
bodily
processes,
it
is
68
necessary
to
use
a
metric
suitable
for
multi­
pathway
modeling.
Such
a
metric
is
a
MET.

MET:
Metabolic
Equivalent
of
Work
A
MET
is
defined
to
be
the
ratio
of
an
activity­
specific
metabolic
rate
to
a
person's
resting
metabolic
rate
(
RMR).
Since
a
MET
is
a
ratio,
it
is
unitless.

There
are
scores
of
articles
providing
estimates
of
METS
for
specific
activities
(
METa).
These
will
be
provided
in
NERL's
"
Consolidated
Human
Activity
Database"
(
CHAD)
in
the
most
disaggregated
way
possible;
at
the
minimum­­
and
especially
for
low­
MET
activities­­
a
deterministic
"
best
estimate"
will
be
provided.
Otherwise,
an
estimated
METa
range
or
distribution­
parameters
of
the
METa
will
be
provided.

CHAD
will
also
(
in
early
1998)
provide
post­
high
exercise­
adjusted
METS
to
mimic
the
"
oxygen
debt"
phenomenon
(
higher­
than­
expected
oxygen
consumption
after
exercise
to
compensate
for
the
oxygen
used
during
anaerobic
breakdown
of
adenosine
triphosphate
(
ATP).

Modeling
Inhalation
Intake
Dose
Using
METS
1.
For
a
particular
individual
sampled
from
CHAD
to
represent
a
specific
age/
gender
cohort,
obtain
an
estimate
of
his
or
her
body
weight
(
body
mass:
BM)
in
kg.
Said
estimates
will
be
provided
in
CHAD,
but
could
be
ignored
by
the
user
and
obtained
some
other
way.

2.
Predict
RMR
(
in
units
of
kcal
min­
1
kg­
1)
using
a
regression
approach
using
two
or
more
anthropogenic
variables
(
age,
gender,
BM)
as
independent
variables.
There
are
over
100
relevant
equations
found
in
the
literature.
Add
variability
for
a
particular
individual
by
adding
on
an
"
error
term"
(
e:
x
)=
0,
F,
from
the
literature)
using
Monte
Carlo
sampling.
Multiply
RMR
by
1,440
and
by
BM
to
obtain
daily
resting
metabolism
(
RMD):

RMD
=
(
RMR
+
e
)
*
1,440
*
BM
3.
Obtain
activity­
specific
estimates
of
metabolism­­
energy
expenditure
(
EE)­­
as:
METa
*
RMR
*
BM;
we
will
call
this
EEa.
Its
unit
is
kcal
min­
1.

4.
Obtain
activity­
specific
oxygen
consumption
by
converting
the
EEa
to
mL
O
2
using
a
distribution
of
energy­
oxygen
conversion
factors
found
in
the
literature.
(
The
age/
gender
specific
estimate
of
BM
obtained
above
minimizes
the
spread
of
this
distribution
but
does
not
eliminate
it.)
The
converted
variable
is
V
0
O2a.
69
5.
Convert
V
0
O2a
to
V
0
Ea
using
a
distribution
of
V
0
E­
to­
V
0
O
2
ratios
found
in
the
literature.
(
The
pNEM
model
uses
these
now.)
A
summary
of
such
ratios
specific
to
age/
gender
classes
is
available.
V
0
Ea,
when
multiplied
by
the
activity­
specific
pollutant
concentration
in
the
microenvironment
provides
the
estimate
of
inhalation
intake
dose
for
the
activity.
Steps
4
and
5
easily
could
be
conflated
into
a
single
conversion
distribution
if
so
desired.

6.
Obtain
total
daily
energy
expenditure
(
EED)
by
simply
summing
all
EEa's
for
the
day.

Modeling
Daily
Fluid
&
Water
Ingestion
Intake
Dose
It
is
assumed
here
that
there
is
an
estimate
of
the
amount
of
xenobiotic
substance
of
interest
in
fluids
and
water
consumed
by
the
individual.
The
estimate
could
be
either
microenvironmentspecific
or
on
a
space­
free
basis.
If
it
is
the
former,
then
EEa
would
be
used
to
develop
the
fluid/
water
intake
dose
estimate.
The
following
assumes
that
only
daily
estimates
of
the
xenobiotic
substance
are
available.

7.
Convert
EED
to
daily
fluid
and
water
consumption
(
FWID)
using
relationships
found
in
the
literature;
the
relationship
is
fairly
linear
for
an
individual.
FWID
is
the
estimate
of
daily
fluid/
water
intake
needed
for
fluid
ingestion
modeling.

Modeling
Daily
Food
Ingestion
Intake
Dose
As
above,
it
is
assumed
that
there
is
an
estimate
of
the
xenobiotic
substance
in
the
individual's
daily
food
intake­­
either
specific
to
a
food
type
or
to
a
daily
aggregate
(
or
proportion)
of
food
types.

8.
Convert
EED
to
daily
food
consumption
(
in
kcal)
directly,
assuming
that
the
individual's
daily
energy
intake
(
EID)
equals
his
or
her
daily
energy
expenditure.
This
implies
that
there
is
no
daily
loss
or
gain
in
body
mass­­
weight.
If
body
mass
change
is
considered
in
the
assessment,
EID
must
be
adjusted
for
the
change.

Use
age­,
gender­,
and
cultural­
specific
nutritional
proportions
found
in
the
dietary
literature
to
disaggregate
EID
into
the
amount
of
protein,
carbohydrates,
and
fat
consumed.
Alternatively,
use
a
"
food
basket"
approach
to
estimate
the
amount
and
type
of
food
consumed
given
EID.
These
approaches
will
provide
an
estimate
of
food
ingestion
needed
for
risk
assessment
modeling.

Modeling
Dermal
Absorption
Intake
Dose
70
The
only
other
major
route
of
multi­
pathway
modeling
usually
discussed
is
dermal
absorption.
This
route
is
not
well­
handled
using
an
energy
expenditure
approach,
although
for
a
particular
activity,
dermal
absorption
probably
is
monotonically
related
to
expenditure
(
as
more
activity
a
priori
increases
the
surface
area
of
skin
available
for
contact
given
a
steady­
state
concentration
of
the
substance
in
the
microenvironment).
This
subject
needs
more
research
in
order
to
develop
usable
metabolic­
or
energy
expenditure­
to­
dermal
uptake
relationships.

Classifying
Individuals
by
Activity
Levels
Divide
EED
by
RMD
to
obtain
what
clinical
nutritionists
call
a
person's
Physical
Activity
Index
(
PAI;
it
is
unitless).
The
PAI
can
be
used
to
classify
individuals
into
activity
classes
for
separate
exposure
modeling
assessment
using
the
following
PAI
categories:

Sedentary
<
1.50
Normal
1.50
­
1.75
Active
>
1.75
For
example,
the
PAI
could
be
used
to
identify
active
children
or
exercising
adults.
CHAD
will
also
include
(
in
1998)
an
estimate
of
a
person's
PAI,
but
obviously
it
could
be
ignored
by
a
user.
71
Appendix
H
METS
Distributions
for
CHAD
Activities
Table
H­
1.
METS
Distributions
for
CHAD
Activities
ACTIVITY
AGE
OCC.
DN
DL
MEAN
MED
SD
MIN
MAX
FLAG
10...
X
ADMIN
4
L
1.8
1.7
0.1
1.4
2.7
0
10...
X
PROF
4
L
1.8
1.7
0.1
1.4
2.7
0
10...
X
ADMSUP
4
L
1.8
1.7
0.1
1.4
2.7
0
10...
X
TECH
4
L
3.9
3.0
0.3
1.3
8.4
1
10...
X
TRANS
4
L
3.9
3.0
0.3
1.3
8.4
1
10...
X
SALE
5
T
2.9
2.7
1.0
1.2
5.6
0
10...
X
SERV
5
T
2.9
2.7
1.0
1.2
5.6
0
10...
X
HSHLD
5
T
2.9
2.7
1.0
1.2
5.6
0
10...
X
PROTECT
5
T
3.3
3.3
0.4
2.5
4.5
1
10...
X
PREC
4
L
3.6
3.5
0.0
2.5
6.0
1
10...
X
MACH
5
T
5.2
5.3
1.4
1.6
8.4
1
10...
X
FARM
5
T
8.5
8.4
2.1
3.6
13.8
1
17100
1
X
4
L
.
5.0
.
1.4
16.0
1
17100
2
X
1
N
5.0
5.0
2.0
1.0
9.0
1
17100
3
X
1
N
4.5
4.5
1.4
1.7
7.3
1
17110
1
X
4
L
.
3.2
.
1.4
10.0
1
17110
2
X
4
L
.
3.2
.
1.4
10.0
1
17110
3
X
4
L
.
3.0
.
1.4
9.0
1
17111
1
X
1
N
5.6
5.6
2.1
1.4
9.8
1
17111
2
X
1
N
5.8
5.8
2.4
1.0
10.6
1
17111
3
X
1
N
4.7
4.7
1.8
1.1
8.3
1
17112
1
X
2
U
3.8
3.8
1.0
2.0
5.5
1
17112
2
X
2
U
3.8
3.8
1.0
2.0
5.5
1
17112
3
X
2
U
3.5
3.5
0.9
2.0
5.0
1
17120
1
X
4
L
.
3.9
.
2.0
9.0
1
17120
2
X
4
L
.
3.9
.
2.0
9.0
1
17120
3
X
6
P
3.5
3.5
.
.
.
1
17121
1
X
4
L
.
3.9
.
2.0
9.0
1
17121
2
X
4
L
.
3.9
.
2.0
9.0
1
17121
3
X
6
P
3.5
3.5
.
.
.
1
17130
1
X
4
L
.
5.5
.
1.8
11.3
1
17130
2
X
1
N
5.7
5.7
1.8
2.1
9.3
1
17130
3
X
1
N
4.7
4.7
1.2
2.3
7.1
1
17131
1
X
4
L
.
5.5
.
1.8
11.3
1
17131
2
X
1
N
5.7
5.7
1.8
2.1
9.3
1
17131
3
X
1
N
4.7
4.7
1.2
2.3
7.1
1
17140
1
X
1
N
5.3
5.3
1.8
1.7
8.9
1
17140
2
X
1
N
5.2
5.2
1.7
1.7
8.9
1
72
17140
3
X
1
N
3.8
3.8
1.0
1.8
5.8
1
17144
1
X
1
N
5.3
5.3
1.8
1.7
8.9
1
17144
2
X
1
N
5.2
5.2
1.7
1.7
8.9
1
17144
3
X
1
N
3.8
3.8
1.0
1.8
5.8
1
17180
1
X
4
L
.
5.9
.
2.0
17.4
1
17180
2
X
1
N
6.0
6.0
2.0
2.0
10.0
1
17180
3
X
1
N
4.8
4.8
1.4
2.0
7.6
1
10200
X
X
2
U
1.8
1.8
0.4
1.0
2.5
0
10300
X
X
2
U
1.8
1.8
0.4
1.0
2.5
0
11000
X
X
5
T
4.7
4.6
1.3
1.5
8.0
1
11100
X
X
4
L
.
2.5
.
2.0
4.0
0
11110
X
X
3
E
2.8
2.5
0.9
1.9
4.0
0
11200
X
X
3
E
3.4
3.0
1.4
2.0
5.0
1
11210
X
X
2
U
2.5
2.5
0.1
2.3
2.7
0
11220
X
X
3
E
4.1
3.5
1.9
2.2
5.0
1
11300
X
X
1
N
5.0
5.0
1.0
2.0
7.0
1
11310
X
X
3
E
5.3
4.5
2.7
2.6
6.0
1
11400
X
X
3
E
2.2
2.0
0.7
1.5
4.0
0
11410
X
X
6
P
2.0
2.0
.
.
.
0
11500
X
X
6
P
2.0
2.0
.
.
.
0
11600
X
X
1
N
4.5
4.5
1.5
2.0
8.0
1
11610
X
X
6
P
4.5
4.5
.
.
.
1
11620
X
X
3
E
4.9
4.5
1.4
3.5
6.0
1
11630
X
X
5
T
3.5
3.4
0.4
3.0
4.5
1
11640
X
X
3
E
4.7
4.5
0.7
4.0
6.0
1
11650
X
X
2
U
4.5
4.5
1.4
2.0
7.0
1
11700
X
X
2
U
3.5
3.5
0.9
2.0
5.0
1
11800
X
X
2
U
3.3
3.3
0.1
3.0
3.5
1
11900
X
X
3
E
6.6
5.5
3.6
3.0
9.0
1
12000
X
X
4
L
.
3.0
.
2.5
5.0
1
12100
X
X
2
U
3.3
3.3
0.1
3.0
3.5
1
12200
X
X
2
U
3.3
3.3
0.1
3.0
3.5
1
12300
X
X
2
U
2.8
2.8
0.1
2.5
3.0
0
12400
X
X
2
U
2.8
2.8
0.1
2.5
3.0
0
12500
X
X
2
U
2.8
2.8
0.1
2.5
3.0
0
12600
X
X
2
U
4.5
4.5
0.3
4.0
5.0
1
12700
X
X
2
U
3.2
3.2
0.1
3.0
3.3
1
12800
X
X
2
U
3.0
3.0
0.3
2.5
3.5
1
13000
X
X
5
T
3.8
3.7
0.8
2.0
6.0
1
13100
X
X
2
U
3.3
3.3
0.4
2.5
4.0
1
13200
X
X
5
T
3.7
3.6
0.8
2.0
6.0
1
13210
X
X
5
T
3.9
3.8
0.8
2.2
6.0
1
13220
X
X
2
U
3.4
3.4
0.6
2.3
4.5
1
13230
X
X
2
U
3.5
3.5
0.6
2.5
4.5
1
13300
X
X
2
U
3.5
3.5
0.6
2.5
4.5
1
13400
X
X
2
U
3.5
3.5
0.6
2.5
4.5
1
13500
X
X
2
U
3.5
3.5
0.6
2.5
4.5
1
13600
X
X
2
U
3.5
3.5
0.6
2.5
4.5
1
13700
X
X
2
U
3.5
3.5
0.6
2.5
4.5
1
13800
X
X
2
U
3.5
3.5
0.6
2.5
4.5
1
14000
X
X
2
U
2.0
2.0
0.6
1.0
3.0
0
14100
X
X
1
N
2.0
2.0
0.3
1.0
4.0
0
14110
X
X
2
U
3.0
3.0
0.6
2.0
4.0
1
14120
X
X
2
U
1.8
1.8
0.4
1.0
2.5
0
14200
X
X
2
U
1.8
1.8
0.4
1.0
2.5
0
14300
X
X
4
L
.
3.0
.
2.5
5.0
1
14400
X
X
2
U
1.8
1.8
0.1
1.5
2.0
0
73
14500
X
X
4
L
.
0.9
.
0.8
1.1
0
14600
X
X
6
P
2.5
2.5
.
.
.
0
14700
X
X
5
T
2.0
2.0
0.4
1.0
2.9
0
15000
X
X
4
L
.
1.8
.
1.4
4.0
0
15100
X
X
2
U
2.1
2.1
0.4
1.4
2.8
0
15110
X
X
2
U
2.3
2.3
0.4
1.5
3.0
0
15120
X
X
2
U
2.1
2.1
0.4
1.4
2.8
0
15130
X
X
2
U
2.0
2.0
0.3
1.4
2.5
0
15140
X
X
2
U
1.8
1.8
0.2
1.4
2.2
0
15200
X
X
2
U
2.2
2.2
0.5
1.4
3.0
0
15300
X
X
6
P
1.8
1.8
.
.
.
0
15400
X
X
2
U
2.3
2.3
0.4
1.5
3.0
0
15500
X
X
2
U
2.8
2.8
0.7
1.5
4.0
0
16000
X
X
4
L
.
2.0
.
1.0
6.0
0
16100
X
X
2
U
2.7
2.7
0.8
1.4
4.0
0
16200
X
X
2
U
1.7
1.7
0.2
1.4
2.0
0
16210
X
X
2
U
1.7
1.7
0.2
1.4
2.0
0
16300
X
X
2
U
1.3
1.3
0.2
1.0
1.6
0
16400
X
X
2
U
1.7
1.7
0.4
1.0
2.3
0
16500
X
X
2
U
2.5
2.5
0.3
2.0
2.9
0
16600
X
X
2
U
1.5
1.5
0.3
1.0
1.9
0
16700
X
X
4
L
.
3.0
.
1.5
8.0
1
16800
X
X
4
L
.
3.0
.
1.5
8.0
1
16900
X
X
2
U
3.8
3.8
1.3
1.5
6.0
1
17113
X
X
2
U
3.0
3.0
0.6
2.0
4.0
1
17114
X
X
5
T
3.1
3.2
0.6
1.4
4.0
1
17122
X
X
2
U
1.5
1.5
0.2
1.2
1.8
0
17141
X
X
5
T
2.8
2.7
0.8
1.5
5.0
0
17142
X
X
5
T
2.0
1.9
0.4
1.5
3.0
0
17143
X
X
2
U
2.5
2.5
0.3
2.0
3.0
0
17150
X
X
5
T
3.3
3.2
0.6
2.4
5.0
1
17160
X
X
2
U
1.6
1.6
0.2
1.2
2.0
0
17170
X
X
2
U
5.0
5.0
1.7
2.0
8.0
1
17200
X
X
4
L
.
1.3
.
1.0
2.3
0
17210
X
X
2
U
1.5
1.5
0.2
1.2
1.8
0
17211
X
X
2
U
.
.
.
1.2
.
0
17212
X
X
2
U
.
.
.
1.2
.
0
17213
X
X
2
U
.
.
.
1.2
.
0
17214
X
X
2
U
.
.
.
1.2
.
0
17215
X
X
2
U
.
.
.
1.2
.
0
17216
X
X
2
U
2.7
2.7
0.8
1.4
4.0
0
17220
X
X
4
L
.
1.2
.
0.9
2.3
0
17221
X
X
2
U
1.2
1.2
0.1
1.0
1.3
0
17222
X
X
2
U
1.9
1.9
0.2
1.5
2.3
0
17223
X
X
6
P
1.0
1.0
.
.
.
0
17230
X
X
2
U
1.3
1.3
0.2
1.0
1.6
0
17231
X
X
2
U
1.3
1.3
0.2
1.0
1.6
0
17232
X
X
2
U
1.3
1.3
0.2
1.0
1.6
0
17233
X
X
2
U
1.3
1.3
0.2
1.0
1.6
0
17240
X
X
2
U
1.4
1.4
0.2
1.0
1.8
0
17241
X
X
2
U
1.4
1.4
0.2
1.0
1.8
0
17242
X
X
2
U
1.4
1.4
0.2
1.0
1.8
0
17250
X
X
2
U
1.2
1.2
0.1
1.0
1.3
0
17260
X
X
2
U
1.9
1.9
0.2
1.5
2.3
0
17300
X
X
2
U
1.5
1.5
0.2
1.2
1.8
0
18000
X
X
4
L
.
2.0
.
1.0
7.0
0
18100
X
X
4
L
.
2.0
.
1.0
7.0
0
74
18200
X
X
4
L
.
2.0
.
1.0
7.0
0
18300
X
X
4
L
.
2.0
.
1.0
7.0
0
18400
X
X
4
L
.
2.0
.
1.0
7.0
0
18500
X
X
4
L
.
2.0
.
1.0
7.0
0
18600
X
X
4
L
.
2.0
.
1.0
7.0
0
18700
X
X
4
L
.
2.0
.
1.0
7.0
0
18800
X
X
4
L
.
2.0
.
1.0
7.0
0
18900
X
X
4
L
.
2.0
.
1.0
7.0
0
18910
X
X
4
L
.
2.0
.
1.0
7.0
0
18920
X
X
4
L
.
2.0
.
1.8
7.0
0
Key
for
activity
distribution
file.

ACTIVITY
=
activity
code
(
activity
10...
refers
to
a
set
of
activity
values,
rather
than
a
single
value)
AGE
=
age
group
(
1=
<
25;
2=
25­
39;
3=
>
40;
X
=
missing)
OCC.
=
occupation
(
X=
missing)
DN
=
indicator
for
distribution
type
(
1
­
normal;
2
­
uniform;
3
­
exponential;
4
­
lognormal;
5
­
triangular;
6
­
point)
DL
=
indicator
for
distribution
type
(
first
letter
of
distribution
name)
MEAN
=
mean
value
MED
=
median
value
SD
=
standard
deviation
MIN
=
lower
bound
MAX
=
upper
bound
FLAG
=
indicator
of
high
median
(
0
if
MED
<
3;
1,
otherwise)

Missing
numeric
values
are
indicated
by
a
single
decimal
point.

Further
information
about
the
activity
codes
can
be
found
in
Table
in
Appendix
.
