Nationwide Truck Commodity Flow Study - Data Collection and Processing

Nationwide Truck Commodity Flow Study - Data Collection and Processing - May 1976

Click HERE for graphic.

NATIONWIDE TRUCK COMMODITY FLOW STUDY

DATA COLLECTION AND PROCESSING

Reported by: Frances H. Anderson
Program Management Division
Office of Highway Planning

May 1976

NATIONWIDE TRUCK COMMODITY FLOW STUDY

Data Collection and Processing

TABLE OF CONTENTS

I. Purpose of the Study

II. Sample Design

III. Sampling and Data Collection Problems

A. Annual Change of License Tags
B. Invalid Truck Registrations
C. Grouping of Registration Numbers
D. Unforeseen Changes in Registration number Series
E. Poor Cooperation by a Motor Vehicle Department
F. Companies with Large Fleets
G. Rental Trucks
H. Efforts Expended by State Highway Agencies
I. Questionnaire Design
J. Telephone Followup
K. Mailing Losses

IV. Coding and Editing Problems

A. Outline of FHWA Procedures
B. Outline of CDC's Procedures
C. Microfilming
D. Training
E. Review of Edited Questionnaires
F. Commodity Coding
G. Geocoding
H. Quality Control

V. Computer Editing Problems

VI. Commodity Resolution Program

VII. File Conversion Program

VIII. Completion Delays

IX. Recommendations

X. Appendix

NATIONWIDE TRUCK COMMODITY FLOW STUDY

I. Purpose of the Study

The nationwide Truck Commodity Flow Study (NTCFS) was designed to
bridge the gap in information on the total universe of commodity
movement by truck. If adequate data were available such as are
easily obtained for other transportation modes, these data would
assist the States and the Department of Transportation in better
defining their policy positions in such areas as highway
development, intermodal competition and emergency transportation
planning.

The need is very great for reliable, current, national and regional
statistics on the characteristics of commodity movements over the
Nation's streets and highways. A discussion of the inadequacies of
present sources of information on goods movements over highways
follows.

Reports prepared by regulated carriers that are required by the
Interstate Commerce Commission provide one source of information on
goods movement over highways. These carriers transport about 40
percent of the ton-mileage of commodities moved over highways.
Forty percent is obviously an important but an incomplete portion
of the picture. In addition, the routes of regulated carriers are
largely restricted to the principal arterial system. Therefore,
the reports to the Interstate Commerce Commission provide almost no
input to measure the service performed by all other functional
systems.

A second source of information is the commodity data collected by
State highway departments as part of their annual summer truck
weight studies. It is evident that the kind of commodities carried
by trucks can vary markedly by season on many portions of the
highway network. Moreover, most locations at which States collect
truck weight data are on the rural principal arterial system.
Estimates of commodity movements on lower order rural functional
systems and urban functional systems are subject to greater
sampling variability than the estimates of commodity movements on
rural principal arterials.

Another source, the Census of Transportation, conducted every 5
years by the Bureau of the Census, provides data on manufacturers'
shipments of commodities and the major mode used in shipping. It
omits raw materials and Any product not put through a manufacturing
process (estimated to be about 60 to 70 percent). Firms with fewer
than 20 employees were not included in the last survey.

The preceding discussion has outlined the lack of reliable
information on the totality of highway goods movements for input
into modal analysis. Presently available sources do not satisfy
national needs. The NTCFS will not completely remedy the situation
but it will supply a good deal of what is presently lacking. The
study has two major objectives. The first is to get the origin and
destination movements of privately owned trucks

classified by type of place, such as railroad siding or truck
terminal. The second major objective is to get the commodities
carried on the trucks. Additional information has been obtained on
characteristics of the trucks and of their operation.

The NTCFS sample design includes all carriers, regulated and
nonregulated; all highway systems, not just principal arterial
systems; all commodities moved by trucks, whether raw or
manufactured; all truck movements, whether or not truck was the
major mode used in shipping; and all businesses, regardless of size
of business operation.

The sample selected should provide adequate estimates of the
national picture of commodity movement and, in some cases, a good
regional picture. The national sample size of 1 percent of the
total registered trucks was assumed sufficient to provide with 95
percent confidence, that the true value of ton-miles will be within
10 percent of its NTCFS sample estimate. The sample size for an
individual State is generally not large enough to provide a
reliable State picture. However, three States-Connecticut,
Kentucky, and Wisconsin--requested that their sample sizes be
increased so that State summaries could be made.

II. Sample Design

The NTCFS was based on a probability sample of registration
numbers, either issued or available for issuance during the period,
July 1, 1972--June 30, 1973, to registrants of privately owned
trucks and tractors in the 50 States, the District of Columbia, and
Puerto Rico. Henceforth, in this report, the term, "States" will
include the District of Columbia and Puerto Rico.

The States were assigned to four strata on the basis of the number
of tractors registered during 1969. The components of each stratum
are indicated in Table I.

A separate sample of registration numbers was selected for each
State for each of the 12 months. The selected numbers were
preassigned to the seven days of a one-week period in each month.
The one-week period was the week containing the fifteenth day of
the month. In many States, after the registered weight and other
information were determined for the selected numbers, they were
stratified by registered weight and subsampled to obtain the final
sample for the month. The registrant of a vehicle assigned a
registration number in the final sample was asked to supply data on
the usage of that vehicle during a specified 24-hour period and
also to supply data on other vehicle characteristics. The term
"vehicle" refers only to the power unit bearing the license number
selected in the sample.

4
Table I: - States Stratified by the Number of Tractors in 1969

-------------------------------------------------------------------

Stratum 1:- Stratum 2.- Stratum 3:- Stratum 4:-
Fewer than 5,000 5,000-14,999 15,000-29,999 30,000 or more
Tractors Tractors Tractors Tractors
-------------------------------------------------------------------

Alaska Arizona Georgia Alabama

Delaware Arkansas Iowa California

District of Columbia Colorado Kansas Florida

Hawaii Connecticut Louisiana Illinois

Idaho1/ Kentucky Minnesota Indiana

Maine Maryland Missouri Michigan

Montana Massachusetts New Jersey New York

Nevada Mississippi North Carolina Ohio

New Hampshire Nebraska Tennessee Pennsylvania

New Mexico Oklahoma Washington Texas

North Dakota Oregon Wisconsin

Puerto Rico South Carolina

Rhode Island Virginia

South Dakota West Virginia

Utah

Vermont

Wyoming

1/Idaho was shifted to stratum 1 because its report includes trucks
registered in other States which receive only a validation sticker
from Idaho upon payment of full registration fee.

The desired sample size for each month for the States in the four
strata is presented in Table II. The sample of registration numbers
was selected by the Washington Headquarters of FHWA three months months
before the start of data collection in July 1972. Which numbers would
actually be assigned to registrants during the data collection period
could not be known during the selection phase. Therefore the sample of
numbers were increased to reflect the expected proportion of unissued
registrations.

Table II:-Expected Number of Sold Plates Per Month in the Sample
Classified by Stratum and Vehicle Category

----------------------------------------------------------------------
Sold Plates Desired Total Number Total Total for
Stratum Per Month For Per of Per the
Light Medium Heavy State States Month Year
Trucks Trucks Trucks
----------------------------------------------------------------------

1 50 50 50 150 17 2,250 27,000

2 70 70 70 210 14 2,940 35,280

3 100 100 100 300 11 3,300 39,600

4 130 130 130 390 10 3,900 46,800
-----------------------------------------------------------------------

Total 52 12,390 148,680
------------------------------------------------------------------------

For sampling purposes, all States provided lists of the registration
number series that would be available during the one-year data collec-
tion period. The States indicated whether each series had been designated
for issuance to a particular weight class, to a particular operation class,
or to a particular vehicle type class. The States also estimated how many
registration numbers would be issued by the end of each quarter year.

For 15 States, it was possible to stratify all the registration
numbers into at least the three categories indicated in Table II.
A final sample of numbers was selected from each stratum by the Federal
Highway Administration (FHWA) and sent to the States.

In 17 States, registration numbers were not predesignated for issuance
to any category of vehicle. It was not possible to assign any of the
number series to one of the categories indicated in Table II. A first
stage sample of license numbers was selected from the entire number file
of each State at a rate expected to yield the number of heavy trucks
desired for that State. This oversampled light and medium trucks. Each
month, the States determined which of the selected numbers had been issued

to registrants and the registered weights of the vehicles. All
numbers assigned to vehicles registered above a prespecified weight were
included in the final sample. All other numbers were subsampled to yield
the final sample. The subsampling rates and random starts were specified
by the Washington Headquarters of FHWA.

In each of the remaining 20 States, one or more of the number series
were designated for special categories, such as farm vehicles or prorated
vehicles.2/ Most of the vehicles registered in the special categories
could be assumed to be one of the categories indicated in Table II. A
final sample of registration numbers was selected from such designated
series. A first stage sample was selected for each State from the
remaining number series. Each month, those numbers that were found to
have been assigned to vehicles were stratified on the basis of registered
weight, and a final sample was selected from each stratum.

The State highway agencies were sent computer listings of the license
numbers for the entire 12-month data collection period and subsampling
instruction, if required. The monthly computer listings were referred
to as the Registration Number Register (RNR). 3/ These were used to
record the status of each license number in the final sample throughout
the data collection phase. According to the State's preference, each
received either a deck of punched cards or a magnetic tape containing the
selected license numbers. The license numbers were then matched to the
registration files to get, for each issued license: the registrant's name
and address, the vehicle's registered weight and the registration basis
used by the State.

Approximately three weeks before the data week in each month, the
truck registration files were searched to determine which of the selected
license numbers had actually been issued to registrants. The name and
address of each registrant and the registered weight of his vehicle were
obtained. If required, numbers assigned to vehicles in specified weight
classes were subsampled.

Four to seven days before an assigned travel day, a questionnaire, an
explanatory letter, and a postpaid return envelope were mailed by each
State to registrants assigned license numbers in the final sample.
A reminder letter was mailed to each registrant to reach him hopefully
on the travel day or on the following day. If no response was received by
the end of a one-week period following the travel day, a follow-up
questionnaire was mailed to the registrant. Returned questionnaires were
reviewed by State personnel and completed, in many cases, by telephone
follow-up.

2/ Prorated vehicles are registered in each of the several States in
which they operate. Registration fees are paid proportional to the amount
of operation in each State.

3/ See Figure 1, Appendix.

Two States, Connecticut and Wisconsin, requested that their sample
rates be increased to yield somewhat more than twice the number originally,
planned, and to provide a better basis for detailed State analysis.
Kentucky also requested a larger sample size so that it could do a State
analysis based on the combined samples of NTCFS and the Kentucky Statewide
Travel Survey which was conducted concurrently with the NTCFS. The FHWA
staff selected both samples but Kentucky processed the Kentucky Statewide
Travel Survey questionnaires.

Connecticut registration files are automated. It was agreed that the
State could select its own sample from the current files each month
instead of using the numbers preselected in Washington, D.C. Connecticut
Department of Transportation personnel calculated a sample rate for each
of three weight classes to be applied, with random starts to the file in
each of three months composing a quarter year.

III. Sampling and Data Collection Problems

This section presents the problems encountered in implementing the
study design and details the steps taken, if any, to resolve them.

A. Annual Change of License Tags

Most States issue new license tags each registration year, Updating
the passenger car registration files is given priority over updating the
truck registration files.

It was anticipated that two months would provide sufficient time for
motor vehicle departments to update the truck registration files. The
"Procedures Guide" 4/ for the study prepared by the Federal Highway
Administration suggested that the States might use registration
information from the old year's files during the two-month update period.
However, in some States, three months were required before the current
year's truck registration files were sufficiently updated to be a useful
source of information. Whenever this situation occurred, the States
obtained registration information for the study from the old file during
the first three months of the new registration year.

Extension of the period for making use of the old registration files
was agreed to reluctantly. The longer the period, the greater is the non-
response rate. At least three causes for this increase in nonresponse
may be cited: registrants move and the new addresses are unknown;
registrants who learn that their current numbers are not available
believe they do not have to return the questionnaire; and registrants
with two or more trucks forget which truck was assigned the selected
license number the previous registration year and do not complete the
questionnaire.

In three States, Louisiana, Illinois, and Florida, special situations
were encountered. In Louisiana, the registration year starts April 1,
with a three-month grace period ending June 30. The effective
registration year starts, therefore, at July 1. During July 1972, the
motor vehicle department was engaged in automating the current registration
data. Most of the truck registration applications were awaiting conversion
to a form suitable for machine processing. The State's study supervisor did
not consider it feasible to obtain registration information from the old
files. As a result, no data were obtained for July. Therefore, Louisiana
contributed only 11 months of data to the National study. The June
1973, data were duplicated and substituted for the July 1972, sample.

4/
See Appendix, page 7.

In Illinois, the registration year extends from July 1 to June 30.
Registration application data are prepared for computer processing
with priority given to passenger car registrations. For July and August,
the State obtained registration information from the old files. In
September, registration information from the new files was obtained
only for a very small proportion of licenses in the sample. The State's
study supervisor did not attempt to obtain data from the old file even
though that was suggested to him. He was convinced that the truck files
would be updated in time for the October samples. On the contrary, it took
Illinois six months to update its truck files. As a result, no data were
received from Illinois for October through December. No data from
Illinois was included for the survey months of September, October, and
November; January data were duplicated and substituted for December.

Florida's truck registration year starts July I with a grace period
ending August 21. At the start of data collection, the motor vehicle
department was changing from a manual operation to a computerized
operation. It was anticipated that current year data would not be
available for four months. It was agreed that Florida would obtain
truck registration information from the previous year's files for the
July-October samples. It was further agreed that the State could benefit
from the newly automated files by selecting the samples for each month
after October from the sold numbers in the file instead of using the
numbers preselected in Washington, D.C., from the complete series of
possible numbers.

B. Invalid Truck Registrations

Analysis in Washington of the returns for the first two months showed
that a significant proportion of the nonresponses in some States with a
high nonresponse rate was assignable to registrants who had moved or
whose addresses were unknown. A check by one State, California,
showed that most of such nonresponses were for registrants in
some former year who had not renewed their applications during
the intervening years. California is one of a growing number of
States that issues validation stickers to reregistrants. The State's
truck registration files were not well purged when a registrant failed
to file for renewal with the start of a new registration year. All
States were asked to check whether their sample
numbers represented currently valid registrations. Invalid
registrations
were then considered in the same class as numbers never issued. This
procedure reduced the nonresponse rate in several States.

C. Grouping of Registration Numbers

In any given State one of two procedures was programmed for computer
assignment of each selected registration number to a day of the week
for which usage data would be required. In the first procedure, if
numbers were to be selected from a given registration number series,
the largest integer, I, less than or equal to N/7 was determined. The
first group of I selected numbers was assigned to Monday, the second

group of I selected numbers was assigned to Tuesday, etc. In the
second procedure, the first selected number was assigned to Monday,
the second selected number was assigned to Tuesday, etc. The first
procedure was simpler than the second to program. It was
adopted whenever there was no indication that the selection procedure
would result in undesirable clustering. During the first month of data
collection it was found that undesirable clustering had occurred in some
States. In such cases, the selected sample numbers for succeeding months
were reassigned using the second procedure.

Clustering was evidenced in one of two ways. In one way, registration
numbers had been issued in numerical sequence. Not all possible
numbers had been issued. As a result, most if not all the numbers
assigned to Saturday and Sunday were unsold. No usage data would have
been obtained. In the second way, license number series had been
assigned to district offices within the State. Under the first
procedure for assigning the data date, all of the license numbers in
the same series were likely to be assigned the same data day. As a
result, all of the trucks sampled in one section of the S@ate were
scheduled to report for the same data day. No truck usage data for
that section of the State would have been obtained for the other data
days. The second procedure for assigning data days should be adopted
for all States in any future study.

D. Unforeseen Changes in Registration Number Series

Before sample selection, each State highway agency sent in a list of
the series of truck registration numbers that would be valid during
the data collection period. The source of each list was the State's
motor vehicle department. In three States, Missouri, Oklahoma, and
the District of Columbia, the motor vehicle departments instituted
changes during the data collection period that were not shown on the
lists submitted as a basis for sampling. In these States, the addition
of a suitable constant to the numbers originally selected produced
registration numbers within the range of the new series. However,
the Missouri State Highway Commission was not notified of the
changes for a three-month period. As a result, the samples
for March, April, and May 1973, did not include any trucks registered
for more than 24,000 pounds.

E. Poor Cooperation by a Motor Vehicle Department

In Massachusetts, the computer registration files do not contain the
registered weights of trucks. Registered weights are available in a
card file requiring manual search. The motor vehicle department in
Massachusetts would not supply the registered weights and would allow
highway department personnel to search the card file only on Sundays.
The search showed that the card file had two faults: it contained
out-of-date data; and many current records were missing from the file.

F. Companies with Large Fleets

Large trucking companies presented several problems.
1. These companies determine which portions of their fleets
will be registered in each State. Since registration
numbers change annually in many States, the large trucking
companies find it more convenient to identify each vehicle
by a constant serial number rather than a changing regis-
tration number. Such companies found it difficult to
complete questionnaires for vehicles identified only by
registration number. All States require a vehicle identifi-
cation number on the truck registration application. The
States were asked to supply such identification number as
well as the registration numbers when sending study
questionnaires to large trucking companies.

2. During October and November 1972, several States were
disturbed because a number of fleet operators who had
completed questionnaires for sample vehicles during the first
two months were now returning questionnaires for vehicles in
their fleets indicating no vehicle travel on the specified days.
All States were to get in touch with fleet operators and with
trucking associations to elicit better or continued cooperation
by truckers. These contacts were in addition to the contacts
State highway agencies were asked to make before the start of
data collection in July 1972.

3. In many fleets, records of vehicle operations are maintained
at some central location. Questionnaires mailed to a local
address by the State were often transmitted to the central
location which might be 2,000 miles away. A time lag
resulted under the best of circumstances. Sometimes the
staff of the central office postponed the job of answering
the questionnaires, especially since they received one or
two questionnaires from several States in each month. The
assistance of the American Trucking Association was obtained
in eliciting better cooperation in such cases from its members.

G. Rental Trucks

During the first few months of data collection, several States
reported that the local offices of truck rental companies were unable
to supply truck usage information for vehicles assigned selected license
numbers. It was decided to try to obtain data on rented truck usage by
mail through the cooperation of four rental companies - Avis, Hertz, Ryder,
and U-Haul. Selected components of their field organizations were asked
to hand a packet containing a questionnaire with a postpaid, pre-
addressed envelope and an explanatory letter to each renter returning
a truck on four specified days, two in May and two in June, 1973.

Of the four companies, only Avis agreed to supply the names and
addresses of truck renters who received questionnaires. The infor-
mation would have been used to mail follow-up questionnaires to non-
respondents. No follow-up was attempted.

Each shipment of packets to a distributor also contained an explanatory
letter plus a preaddressed, postpaid envelope for the distributor.
He was asked to send a count of the number of packs distributed. Not
all responded.

Section A of the standard study questionnaire was suitably modified
for response by truck renters. As with the standard study questionnaire,
respondents were asked to supply truck usage data for a 24-hour
period. For rented trucks, that period was the 24 hours ending when
the truck was returned.

Approximately 11,000 packets were sent to the selected components of
the field organizations of the rental companies. We received only 77
questionnaires with data for rented trucks plus 19 questionnaires for
leased vehicles. Other methods must be devised to obtain data on the
usage of rental trucks.

H. Efforts Expended by State Highway Agencies

The need of the Department of Transportation for data on commodities
carried by trucks is so pressing that the FHWA was asked to undertake
a national study as soon as possible based upon the results of a small
scale pilot study that had previously been completed in three States.
The NTCFS was designed by the FHWA with no chance for input from the
States. As a result, some highway agencies were reluctant to take on
another study at a time when resources were being stretched very thin
by ongoing requirements.

Most States cooperated and assigned manpower to carry out the study.
A creditable job was done by almost all States. Some did an outstanding
job. Wisconsin may be mentioned as one of the States in which study
personnel expended extra effort to get questionnaires that were completed
with consistent data.

On the other hand, Oklahoma at first refused to cooperate because of a
State moratorium on staffing. When an agreement was finally reached
to carry out the study, the effort expended failed to meet minimum
response requirements. During the first six months of data collection,
no follow-up questionnaires were mailed to nonrespondents, and no
telephone calls were made to correct or complete returned questionnaires.
The lack of follow-up effort may have been a result of FHWA's underestimate
of the staffing requirement for the data collection phase. The follow-up
problem was aggravated by the Oklahoma Post Office's failure to forward or
return questionnaires even though they were sent by first-class mail.
More effort was expended during the last six months of data collection
in Oklahoma.

While Massachusetts did not refuse to participate in the study, the
planning agency put almost no effort into doing an acceptable job.
At first two employees were available part-time to conduct the study.
However, their tasks were compounded by the need to manually
transcribe the registered weight from the DMV files before selecting
the license numbers for the final sample. Then because return envelopes
and postpaid mailing envelopes were not available, the mail-out procedure
was even more tedious. No auxiliary help was provided and the follow-up
and review phases suffered.

Puerto Rico produced a Spanish translation of the study questionnaire
and mailed that to registrants. Answers on returned questionnaires were
translated, entered onto English language forms, and submitted.
Despite the Spanish language questionnaire, the response rate in Puerto
Rico was very low.

I. Questionnaire Design 5/

The first month of data collection showed three weaknesses in the
questionnaire design. The States were asked to correct two in an FHWA
Notice dated July 25, 1972. The weaknesses and the corrective
measures
follow.

Some registrants, whose vehicles were not driven during the specified
24-hour period, correctly checked "No" in question 6, Section A, but did
not answer the preceding questions. Their justification was the wording
of the first sentence in question 1: "Please indicate the description of
your vehicle as it was driven during the entire 24-hour period noted
above." The States were asked to cross out the last portion of that
sentence beginning with "as" for all succeeding mailings.

Some respondents provided trip information only on load-carrying
trips. This occurred most frequently for repetitive trips with loaded
trucks out and empty trucks back, as in gravel hauling. The States were
asked to attach a preprinted message to questionnaires mailed in the future
with the message, "Please report all stops, even those when the truck is not
loaded." Some States stamped this message on Section B of the questionnaire.

Some registrants, especially registrants of pickup trucks, returned
unanswered questionnaires with the message that they obviously were
not involved since their vehicles were used for personal transportation
and not to haul commodities. Other registrants returned blank question-
naires because their trucks were not for hire and therefore they believed
they were not required to answer. No measures were taken to solve this
problem. It is believed that had the study title been "Nationwide Truck
Usage Study" the problem would have been minimized.

5/ See Figures 2A and 2B, Appendix

As the study progressed, it was realized that question 6, Section A,
should have been the last data question in Section A. When the
respondent checked "No" for question 6, he was instructed to return
the questionnaire. If he followed instructions, no data were made
available on type of business or on fleet size. An FHWA Notice dated
December 12, 1972, asked the States to cross out the instruction in
question 6: "(if "no" please return questionnaire)." Acceptable data
on type of business and fleet size should be available for the last
half year of the study.

J. Telephone Followup

The "Procedures Guide" requested that States review returned question-
naires for completeness and consistency of information. Priority was
given to questionnaires reporting commodity movement. State survey
employees were to telephone a respondent when the review of his
questionnaire indicated problems. Ideally, the accomplishment of
these tasks required that somebody be available who was capable of
detecting incomplete or inconsistent responses. During telephone follow-
up, this person should exhibit a good telephone personality, and be
able to elicit information. These capabilities were not made
available in every State.

If a similarly designed study is ever undertaken again, personnel
assigned to these tasks should be tested and trained before the
start of data collection. Their output should be reviewed periodically.
Good data must be obtained at the source. Later processing cannot
always correct bad data or supply missing data.

K. Mailing Losses

The States were asked to mail in the returned questionnaires that had
passed review each Friday. One mailing of 68 questionnaires from
Idaho was lost by the Postal Service. These cases will be treated as
nonresponses.

Massachusetts shipped all questionnaires for one month's sample at the
end of the cut-off period for that month instead of sending in a portion
each Friday. The data for June, the last month, never arrived in this
office. The questionnaires for May were duplicated and substituted for
the June sample.

The original request for mailing each Friday was based on the
expectation that all processing would be accomplished by FHWA personnel.
This did not occur. In any case, when States are asked in the future
to mail in irreplaceable questionnaires, they should be asked to send
them by registered mail.

IV. Coding and Editing Problems

Control Data Corporation (CDC), obtained the job to code and edit the
data and to supply the FHWA, with 12 monthly tapes and one summary
tape suitable for tabulation. The CDC set-up a schedule to prepare
for the operation during September and October, 1972, and to start
processing in November. The schedule called for CDC to finish the
job in July of 1973.

A. Outline of FHWA Procedures

1. Record the number of questionnaires received at weekly
intervals from each State.

2. Verify for each questionnaire received its identifying
information on the accompanying Shipment Record with that
in the RNR.

3. Indicate in the RNR the response status of each license
number in the final sample.

4. Bundle the questionnaires and Shipment Records by State for
each data month and deliver them to CDC.

B. Outline of CDC's Procedure

1. Group the questionnaires from one State for one month into
work batches of approximately 100 forms and assign serial
numbers to each form.

2. In a pre-code edit operation, review the contents of each
batch for consistency and legibility. Make corrections and
improvements in accordance with instructions provided in
the editor's manual. Assign codes to several of the items
on the questionnaire.

3. Punch questionnaires in a flexible format. Codes were
punched for items which were self-coded or which had
been precoded. Punch commodity name and geographical
location in alphabetic format. In general, several cards
were punched for each questionnaire.

4. Verify the keypunching of questionnaire data by an
independent keycoding of the same data. Perform a computer
comparison of the two outputs; select the better and enter
it onto a monthly master file.

5. Code, punch and store the data on the Shipment Records as
a control file in the computer. Match the questionnaire
identifying data in the master file against the corresponding
control file data and insert registration information in
each master file record. Flag any omissions, duplications,
or errors in the identification fields and assure that each
questionnaire was accounted for in the master file.

6. Pass the data on the master file through a computerized
edit which checks for accuracy, completeness and consistency.

7. Copy geographic information, punched in alphabetic format,
onto a tape. Deliver the tape to Boeing Computer Services
(Boeing) where a computer program assigned geocodes to at
least 90 percent of the geographic data. Boeing staff
manually coded the remainder. Insert the geocoded data back
into the appropriate master file locations.

8. List alphabetic commodity names and code them according to
the codes in the "Commodity Classification for Transportation
Statistics." Build a computerized library of commodity
codes so that a large proportion of the commodity codes could
be assigned by computer when processing succeeding months of
data.

9. Convert information in the master file into a specified
final format on tape for each data month. Deliver each tape
to FHWA.

C. Microfilming

As a safety measure, CDC sent each batch of questionnaires to a
subcontractor to be microfilmed before processing. When FHWA
personnel inspected the microfilm delivered by CDC, FHWA found that
the subcontractor failed to maintain good control of his operation.
Both sides of a questionnaire, were not always filmed, and control
numbers were not always correctly indicated on the microfilm
containers. Finally, the subcontractor lost one batch of question-
naires from the September sample for Massachusetts. The microfilming
of questionnaires was terminated.

The fact that CDC supervisory personnel did not discover the problem
of bad control in the microfilm processing was the first sign of a long-
term weakness--uncritical acceptance of outputs from the different
stages of the operation. At least three causes may be advanced for
this:

1. Insufficient time spent by the top supervisor on the
project.

2. Inexperience of the processing managers assigned to the
project.

3. Too great a reliance on computer processing to yield an
acceptable final product.

D. Training

The CDC prepared training manuals for the precode editors and
submitted them to FHWA for comments. The FHWA personnel reviewed
the training manuals, and participated in the training sessions.
At the close of the training sessions, revisions and additions to
manuals were planned.

The CDC revised the training material and sent revised copies to FHWA
staff for approval in December 1972, after which, final manuals were
prepared. It was not until August of 1973 that FHWA realized that the

CDC editors had never been issued the revised instructions. Many of
the recurring problems found in FHWA review of the editing in the
interim had been covered in the revisions. The editors could hardly
have been faulted for improper handling of these cases.

When specific problems occurred that were not covered in the revised
version of the edit instructions, FHWA staff prepared brief summaries
of the proper handling of these situations. Again, it was discovered
that the summaries were not distributed to the editors.

E. Review of Edited Questionnaires

By means of a remote terminal installed at FHWA by CDC, the editing
and keypunching operations could be reviewed as the work progressed
without the disturbance of a personal visit. Edited questionnaire
data in the master file could be displayed and compared with the
microfilm records of the original respondent's entries. Weaknesses
in both processes were discovered in the FHWA review of the July data.

1. As each batch of questionnaires was edited, the editor
indicated on "Exception Sheets" the items that he or she
could not resolve. No steps had been taken by supervisors
to resolve the problems thus noted before the batch was
punched. Therefore incomplete or inaccurate data were
entered in the master file. After some prodding, the
obvious corrective measure was adopted; and the problems
were cleared up before the keypunch step.

2. Editors were not always converting quantities of
commodities to commodity weights. Greater emphasis was
placed on this problem in retraining editors. Supervisory
personnel were prodded to expand the table containing
conversion factors for use by the editors.

3. Editors misunderstood when the term "mixed freight" was
acceptable. Rules were clarified and additional training
was given.

4. Editors had trouble determining the weight on the truck at
the start of the day when the truck made more than the
reported nine stops.

5. Punchers seemed to have problems adhering to the strict
punch format. Problems occurred when field delimiters were
omitted.

6. Not punching the data as coded on the questionnaire was
evidenced by errors such as missing data, transposed stop data,
missing commodity weights and misspelled geocode data.

The FHWA rejected the processed July data because of the poor quality
of the editing and the keypunching. After retraining, the editors did
an acceptable job analyzing questionnaires to determine whether the
data were mutually consistent or whether a missing entry could be imputed
from other information on the questionnaire. The FHWA personnel periodi-
cally checked editors' output to assure that acceptable standards were
being maintained.

Keypunchers were also retrained, however, the effect of the retraining
could not easily be measured because of simultaneous changes to the
verification procedure. Keycoding to verify keypunching was
discontinued because CDC considered it too expensive and time-consuming.
It was succeeded by one unverified punching of the data and an eyeball
scan of the listed punch card images. Only the most glaring errors were
found in this step. The scanning function was soon incorporated in a
computer consistency edit program. In each of the three procedures, it
was only as a result of FHWA initiative and insistence that any efforts
were made to assure quality in the keypunching procedure. For the last
six months, data to be keypunched were transcribed onto punch card format
sheets. There was finally, a measurable improvement in keypunch quality.

F. Commodity Coding

Commodity coding was performed by selected editors. For each data
month, uncoded commodity names were punched, sorted and listed
alphabetically. Commodity editors coded directly on the listing
opposite the commodity names.

Keypunchers then punched the identification fields and the commodity
codes for insertion in the corresponding master file records.

Weaknesses were found in the procedure. At times, the commodity names
on the listing did not reflect all the detail provided on the
questionnaires. For example, "corrugated containers" might be listed
as "containers." Commodity editors might then code the item as
"containers returned empty", or as some class of metal container, or
as a paper product, or leave it as a problem to be resolved. In many
instances, commodity editors might have found leads to the correct codes
from data on the questionnaire, such as business type of the registrant
or type of place to which delivery was made. Eventually it was decided
to code commodity on the questionnaire instead of coding from a computer
listing. In other instances, if commodity names were unfamiliar, they
were merely coded unknown. The FHWA outlawed this course and insisted
that editors refer unresolved commodity names to FHWA for coding.

Sometimes instead of commodity names, gibberish was found on the
listing. This seemed to indicate bad punching before the commodity
listing was made. At first, CDC merely deleted these records. Later
at FHWA insistence, these cases were investigated and corrections made
to the master file record and proper commodity names were retrieved and
coded.

The FHWA review again uncovered the fact that insufficient effort was
made to ensure quality at the end of the commodity coding or punching
steps. Instances were found where commodity editors recorded improper
codes; where commodity names went uncoded; where codes were inconsistent
e.g., "lead pipe" and "pipe, lead" had different codes; and where
commodity codes were not punched. Each record referred for commodity
coding did not have a corresponding record of coded data for insertion
in the master file. As a result, incorrect, inconsistent, and incomplete
data were entered in the master file and passed on to succeeding steps.

It was anticipated that a computerized dictionary of commodity codes
could be established from the July and August questionnaires. Use of
this dictionary could decrease the number of commodities requiring
manual coding during the processing of subsequent data months. Two
versions of the commodity dictionary were prepared. One was in alphabe-
tical order of commodity names; the other, in numerical order of
commodity code. Early versions of the commodity dictionary included
names that were too ambiguous or general to permit the computer to
automatically assign codes. The FHWA and CDC worked together to
prepare a computer dictionary of names that could automatically be
coded. Other names were referred to a manual lookup procedure which
included analysis of other entries on the source questionnaire.

G. Geocoding

Boeing was assigned the task of geocoding the stop locations reported
on the questionnaires. Location data were punched in a strict
alphabetic format and included in the master file records for each
questionnaire. In a separate operation, CDC copied the location data
and identifying information onto a tape which was delivered to Boeing.
Boeing then used this tape as input to a computer program, developed by
the Bureau of the Census, which automatically supplied geocodes for all
locations included in the program's reference file. All other locations
were coded manually by Boeing staff using either a computer listing of
the reference file or the 1970 Census of Population and Housing
Geographic Identification Code Scheme (GICS). The geocoded location
data was returned to CDC where it was inserted in the corresponding
master file records. After processing each data month, Boeing staff
expanded the geocode program reference file to include any additional
location names or variations in spelling of location names which were
considered likely to occur frequently in succeeding data months.

20
Few problems arose with the geocoding operation at Boeing. These
concerned codes in the reference file. At first the reference file
contained 1960 data rather than data for 1970. As a result, place
size changes were not reflected; county designations were incorrect;
and newly-defined census places and Standard Metropolitan Statistical
Area (SMSA) designations were not included. These problems were solved
when 1970 data were inserted in the reference file.

In New England where SMSA boundaries split some counties, all places
within the county were given an unknown SMSA code rather than the
applicable code. The FHWA manually researched the appropriate codes
and Boeing corrected the reference file.

H. Quality Control

According to requirements specified by FHWA, CDC developed a plan for
reviewing a sample of each editor's workload. The acceptability of an
editor's output was determined by comparing the number of errors found
on the questionnaires examined with a predetermined tolerance level.
A tally was made of the number of errors in each data field of the
questionnaires. If the tally was within tolerance for each data field,
the batch was considered acceptable.

The FHWA reviewed some of the batches of July questionnaires that had
passed CDC's quality control. The review showed so many errors that
all of the work for July was rejected. It was determined that most
of the errors occurred on questionnaires with vehicle travel, especially
those with commodity information. It was also determined that the small
random sample selected from a work batch most often contained too few
questionnaires with travel to properly estimate the batch's acceptability.
The FHWA decided that review for quality control of editor output could be
based solely on questionnaires reporting travel. In addition, shortly
after this modification was adopted, CDC was asked to supply FHWA with
batches of questionnaires periodically so that FHWA personnel could
continually review editor output that had passed quality control.

All of CDC quality control efforts were concentrated on the manual
editing of questionnaires. Scarcely ever were the results of
succeeding operations reviewed. Specific instances of this failure
will be discussed later in this report.

21
V. Computer Editing Problems

The CDC developed computer procedures for performing a consistency
check on the data that had been manually edited and keypunched. The
computer consistency edit was based upon specifications prepared by
FHWA with some modification to make optimum use of the CDC computer
capabilities. This edit program was eventually augmented and revised
as experience indicated the need. One such set of additions was
incorporated to check the keypunching quality.

The major functions of the consistency edit were to detect whether:

A. The code in a particular field was within the permissible range
of values for that field.

B. The code in a particular field was consistent with the code or
lack of code in another field.

C. The code in a particular field of one type of record was
consistent with the code in a specific field of another type
of record prepared for the same vehicle.

The FHWA was to be provided with listings showing the number of times
each field was erroneously coded or had no entry. The purpose of the
listing was to provide an early indicator of the quality of the manual
edit and of the completeness of the questionnaire response. No listing
was provided until after FHWA conducted an independent review of the
data in the master file and challenged the high frequency of codes
indicating that data were unknown. Investigation of the problem showed
that unknown codes were supplied automatically whenever a questionnaire
field was left blank whether or not a response was required. Editors
and keypunchers were given revised instructions and the edit program
was changed accordingly.

The benefit of the consistency edit as an early warning signal was
nullified because records flagged for containing errors were passed on
unchanged to later steps. Rejected records from this edit were
changed only if a correction card were being prepared for some other
reason. Therefore, incorrect records often remained in the master
file even after errors had been flagged. The rationale for this
procedure was that some processing error must be tolerated. Major
changes would be made only if the number of errors were excessive.
The CDC staff did not analyze the errors flagged by the consistency
edit. Apparently, no attempt was made to determine the source of the
errors by reviewing the questionnaire. Had CDC prepared the specified
listings showing error rates and response rates for particular fields,
they would have been equipped to evaluate the merit of their position
on the tolerance of the error rates. After investigating the causes
of the errors, CDC realized the need for review and correction
procedures for the edit rejects.

It may be well to note a particular instance that resulted from the
incorrect programming of some edit specifications. One of the
interfield edits is performed on stop purpose and the commodity
fields. The edit states that unless the stop purpose is to pickup
or deliver there should not be commodity data. The CDC's program
handled any conflicting cases by blanking out the commodity data.
No error message was provided and no investigation of the cause of
the conflicting codes was made. The problem was compounded later
in commodity resolution when the blank commodity fields were
encountered.

Another problem surfaced with computer editing after CDC began
reviewing rejects. When rejects were analyzed, editors were
prompted to code questionnaires to pass the edit checks. For
example, when many questionnaires were rejected because of missing
vehicle type codes, editors were prodded to supply a vehicle type
code even when there was insufficient information given on the
questionnaire. In a meeting with CDC staff, FHWA worked up
guidelines for supplying missing data and required all exceptions
to the guidelines to be cleared through FHWA.

In another instance editors were instructed to rewrite commodity names
to match those already in the commodity dictionary so that fewer
commodity names would require manual lookup. This was particularly
undesirable because specific commodity names were changed to more
general names and the purpose of the 5-digit level of commodity coding
was thwarted. The rewriting was terminated and specific commodity
names restored.

While edit rejects were being corrected, records that passed edit were
processed through succeeding steps. By the time the corrections were
added to the master file, the corrected records had missed other phases
and lagged behind the other records in the master file in correctness
and completeness. As a result the final master file would contain
records with no geocodes, no commodity codes, and no calculated entries.

23
VI. Commodity Resolution Program

The commodity resolution program was designed to calculate the effect
of each commodity handled during the travel day and to provide
appropriately coded data for inclusion in the master file. This
program assigned a unique number to each commodity handled at a
stop, determined the weight of the carried load, coded the stop
when a commodity was first reported on the truck, and calculated
the distance a delivered commodity had been carried.

The FHWA had provided detailed coding and editing specifications for
this program. However, in spite of this assistance, serious problems
with the procedures and products of this program were encountered, as
follows:

A. Commodity number was incorrectly assigned by the computer.
All commodities delivered and/or picked up within a stop
were supposed to be assigned a number starting with 01 for
each stop, with any delivered commodities being numbered
first. The first few submittals of outputs showed that the
computer was assigning commodity numbers in continuous
fashion for all stops instead of starting with 01 for each
stop.

B. Duplicate commodity records, the results of uncorrected
keypunch errors, remained undetected throughout commodity
resolution.

C. Commodities were coded as delivered at a stop even though
the carried load to that stop was zero. The most frequent
cause of this problem was the reversal of action codes for
pickup and delivery records.

D. Impossible commodity codes were in the field. No provision
had been made to check the keypunching of commodity codes.

E. Mispunched commodity weights were passed on to be resolved
often resulting in excessive weights of carried load. Yet
no corrective or investigative action was taken.

F. The record for each commodity delivered at a stop is
required to contain a code to show the stop when the
commodity was first reported on the truck. The CDC had
difficulty programming the computer to enter the appropriate
code.

G. The distance a commodity was carried was incorrectly coded
as unknown even though the mileage was available. This
occurred especially when commodity weight was unknown.

The CDC was notified of this long list of problems with the commodity
resolution program. Requests for modification of the program were for
a while met with resistance because of rewrite problems and additional
costs. Eventually the program was rewritten so that along with
required capabilities, the computer could also resolve unknown weights
for commodities under specified conditions, treat cases of alternating
pickups and deliveries, and extend the first in-first out principle to
a delivery matched against multiple pickups. These changes were based
upon suggestions and decision tables prepared by FHWA.

VII. File Conversion Program

The purpose of this program was to prepare the questionnaire data in
the specified FHWA format. By this point in the processing procedures,
all data should have been edited and-coded and ready for final output
format. Most of the serious problems occurred at this phase. Many of
the problems were serious only because the processing procedure made
no provision for backtracking to correct errors detected in the final
stages of the processing. Some examples follow:

A. The program at first did not produce the type 2 record
required when the respondent had indicated vehicle travel
but had not provided any information on trips made.
Correction of this error required backing up the processing
to the editing stage.

B. No type 3 records were produced for trips started on the
travel day but not completed on the travel day. This
problem needed revised editing instructions and complete
reprocessing.

C. Questionnaire data in the control file had no match in the
master file. Apparently, some questionnaire records had
been deleted for correction and not replaced. A second
control file/master file match was instituted before file
conversion to prevent this from recurring.

D. License numbers were sometimes omitted and incorrect. This
indicated a failure to review the results of the control
file/master file match. This error is inexcusable because
license number is critical by the very nature of the survey
design.

E. Vehicle type was inconsistent with body type. For example,
the record for a single unit vehicle should not contain a
body type code for a trailer. A decision table was
developed by FHWA to enable the computer edit to detect
any inconsistencies between vehicle type and body type codes.

F. One peculiar situation was discovered in the file
conversion program. A delivery record had been incorrectly
coded as a pickup. The next record was for a pickup of
60,000 pounds. The commodity weight in the error record
was also 60,000 pounds. Therefore, the carried load to the
third stop should have been 120,000 pounds. Only five digits
had been allowed for the carried load field then the computer
attempted to enter the 120,000 pounds in the five-digit field,
the left-most digit was merely truncated, no error message
was printed! One hundred thousand pounds of commodity had
been lost!

Each record stored in the computer master file had been assigned a
random access location number. Adding or deleting records altered all
of the random access numbers and made matching the location numbers at
subsequent stages of the processing impossible.

Matching procedures with the control file, commodity coding and
geocoding had to be redone each time a change was made in the master
file.

26
VIII. Completion Delays

The FHWA efforts to evaluate the quality of the individual phases of
the processing system were often frustrated by the operating procedures
employed.

Review of questionnaire editing revealed that batches of questionnairs
were passing quality control containing errors, omissions and
unresolved problems far above the tolerance levels. In some cases
unresolved problems we're noted on exception sheets. However,
these notations had been ignored and batches were sent on to be
keypunched. Review of keypunching revealed such errors as mispunches,
omissions and transposition of stop data. Yet, the punched cards were
fed into computer programs which were ill-equipped to recognize and
flag errors. Edit programs attempting to evaluate the bad source data
compounded the errors by deleting conflicting entries rather than first
flagging the edit rejects. Edit rejects were never reviewed to trace
the reason for the edit failure. Data from the computer edit was then
used to calculate entries in final format. Incorrect, incomplete,
inconsistent, and incomprehensible data found its way onto the final
master file.

The CDC presented the master file in this incredible condition for
FHWA approval. Subsequently, FHWA reviewed, rewrote and revamped
the edit procedures, quality control, flow charts, decision tables
and computer processing. When the system had been harnessed to the
task, bad housekeeping caused further delays. Deliverables to FHWA
were not checked before being submitted. Listings were delivered with
blank and transposed data fields. Tapes were submitted with improper
format, tape marks, and translation gibberish.

The problem of bad housekeeping was eliminated; and CDC, finally having
a suitable processing system, was put on a strict delivery time
schedule. Then FHWA regularly received processable tapes with a
tolerable error level in the data file. During the FHWA manual
processing of questionnaires requiring imputation of total stops made
by multistep vehicles, another problem surfaced. Some questionnaire
data that was supplied by the respondents had been deleted by CDC when
rejected by the edit procedures. Apparently, it was more expedient for
CDC to delete these entries than to correct them and risk missing the
deadline.

The impact of the data that had to be restored was estimated at
approximately 10 percent of the total questionnaires showing commodity
movement. The FHWA felt this was significant enough to require
retrieval since only 20 percent of the acceptable questionnaires
showed commodity movement. One casualty directly attributed to the
long delay in getting acceptable questionnaire data was a tabulation
of the ton miles and vehicle miles of commodity movement by location
at previous stop and by the destination stop. This table had been
programmed at FHWA by a former FHWA programmer whose documentation
can no longer be deciphered after a 2-year wait for testing
with live data. The table and associated subtables will have to be
reprogrammed.

The long time period required to make the data for NTCFS ready for
publication and analysis is the result of the aforementioned problems.

IX. Recommendations for Future Nationwide Truck Commodity Flow Study

Sample Design

Consideration should be given to the following revisions to the
procedure for selecting the license numbers in the sample.

A. In addition to asking States what license numbers would be
available for sale; ask what is the highest number they
expect to issue for each series.

B. In States where license numbers are not distributed
according to a specified weight class, type of operation,
or use; obtain State data on the distribution by weight
class of the vehicles registered.

C. In States with continuous registration, select the sample
from the State file of issued registrations each month.
Sampling rates and random starts would be specified by
Headquarters FHWA; supervision of the sampling process
would be provided by Division Office personnel.

Sampling and Data Collection

Whenever possible FHWA should encourage the individual State motor
vehicle agencies to improve the efficiency of the truck registration
procedures. Special efforts should be made to encourage States to
reduce the length of the changeover period at the end of the regis-
tration year. In States using validation stickers for re-registrants
it is imperative that registration files be thoroughly purged of non-
reregistrants.

Standardization of information contained in registration files should
be encouraged. Most States collect data on the vehicle weight, vehicle
description, the manufacturer's serial number and the operating class.
If all of this information were consolidated in the registration files,
the sampling process could be simplified.

Rental Trucks

All NCTFS efforts to collect data on commodities moved by rental trucks
failed miserably. Voluntary response to the questionnaires was
practically nil. Rental company estimates of an average day's rental
volume were very far from the number of questionnaires actually
distributed to renters on the specified day. A procedure must be
devised to obtain data on the usage of rental trucks for commodity
movement. Personal interviewing may be warranted in spite of the
associated cost and danger.

Efforts in State Highway Agencies

States should be given the opportunity to provide input to the study
design by indicating any individual data requirements and interests.
Personnel assigned to conduct the State's data collection activities
should be tested and trained by the Headquarters staff in order to
improve the response rate and the effectiveness of the followup.

Realistic manpower estimates for conducting NTCFS should be made by
State and FHWA staff so that NTCFS staff are not spread so thin that
all procedures cannot be adequately covered.

Mailing Losses

The response rates for at least three data months were adversely
affected by mailing losses. Questionnaire data could not be
recollected due to the time lapse, and lost questionnaires were
handled as nonresponses. In the future when States are requested
to mail in irreplaceable questionnaires they should be sent by
registered mail.

Contracted Services

It is recommended that contractors not be used to process data for
this type of survey. There was a significant time lag in discovering
problems, retraining, and effecting corrective measures. This lag
could have been reduced if the coding, editing, and quality control
operations had been performed inhouse with an augmented clerical staff.
Yielding direct control over the processing operation was not
compensated for by a savings in time or money. Computer programming
required of the contractor could have been performed by the FHWA
programmers who produced the summary tabluation programs.

Frequency of Future NTCFS

Because the utility, economic impact and operation of light trucks do
not change significantly over the years, it may be worthwhile to
consider including light trucks in such a survey only every five or
six years. Consideration should be given to sampling fleets from ICC
files and giving fleet owners prior notice of trucks in sample for
upcoming survey months. This would provide sufficient time for fleet
owners to locate the trucks in the monthly sample and forward their
respective questionnaires.

APPENDIX

TRUCK COMMODITY FLOW STUDY

PROCEDURES GUIDE

TABLE OF CONTENTS

Page
GENERAL INFORMATION

A. Purpose of Study . . . . . . . . . . . . . .. . . . . .1

B. Purpose of this Guide. . . . . . . . . . . . . . . . .1

C. Description of the Study . . . . . . . . . . . . . . .1

II. DETAILED PROCEDURES. . . . . . . . . . . . . . . . . . . . .3

A. Determination of Registrants . . . . . . . . . . . . . 3

1. Procedure when not Required to Subsample. . . . . 3

a. Computerized Search of the Registration Files. 3

b. Manual Search of the Registration Files. . . . 4

2. Procedure when Required to Subsample from the
Entire List . . . . . . . . . . . . . . . . . . . 5

a. Computerized Search of the Registration Files 5

b. Manual Search of the Registration Files . . . 6

3. Procedure when Required to Subsample Part of
the List. . . . . . . . . . . . . . . . . . . . . 7

4. Procedure at the End of the Registration Year . . 7

a. First Condition. . . . . . . . . . . . . . . 7

b. Second Condition . . . . . . . . . . . . . . 7

c. Third Condition. . . . . . . . . . . . . . . 8

B. Forms to be Used . . . . . . . . . . . . . . . . . . . 8

C. Mailing Procedures . . . . . . . . . . . . . . . . . . 9

D. Special Handling of Some Returns . . . . . . . . . . .10

1. Leased Vehicles . . . . . . . . . . . . . . . . .10

2. Multistop Vehicles. . . . . . . . . . . . . . . .10

3. Miscellaneous Situations. . . . . . . . . . . . .11

E. Review of Responses. . . . . . . . . . . . . . . . . .12

F. Shipping Procedures . . . . . . . . . . . . . . . . .13

LIST OF FIGURES
Figure
Page

1 Registration Number Register . . . . . . . . . . . . . 15

2A Questionnaire, Section A . . . . . . . . . . . . . . . 16

2B Questionnaire, Section B . . . . . . . . . . . . . . . 17

3 Model Letter of Explanation. . . . . . . . . . . . . . 18

4 Model Letter to be Sent to Registrants During
Changeover Registration Period . . . . . . . . . . . . 19

5 Model Reminder Letter. . . . . . . . . . . . . . . . . 20

6 Model Follow-up Letter . . . . . . . . . . . . . . . . 21

7 Model Letter of Explanation to Lessees . . . . . . . . 22

8 Model Letter to Accompany Supplemental
Questionnaire(s) . . . . . . . . . . . . . . . . . . . 23

9 Shipment Record. . . . . . . . . . . . . . . . . . . . 24

TRUCK COMMODITY FLOW STUDY
PROCEDURES GUIDE

I. GENERAL INFORMATION

A. Purpose of Study

This study was designed primarily to collect information on types
of commodities carried on highways and on destinations of truck
movements. These data together with other information, such as
characteristics of vehicles and of weights of carried load, will
provide a basis for measuring an important economic service provided
by highways and for comparing highway usage with commodity flow over
other modes of transportation. Such information is urgently needed
or national planning and decisionmaking as to the relative needs of
the various transportation modes to make possible an economical
expenditure of the Nation's resources.

B. Purpose of this Guide

This guide explains the general procedures for conducting the
Truck Commodity Flow Study and presents the forms to be used.

C. Description of the Study

The Truck Commodity Flow Study is based upon a sample of truck
registration numbers for nonpublicly owned trucks in each of the 50
States, Puerto Rico, and the District of Columbia. The sample
registration numbers were randomly preselected in Washington without
any foreknowledge of which, if any, had been assigned to registrants.
An independent sample was selected for each month from the entire file
of registration numbers of each State. The registration numbers in the
sample for each of the survey months have been uniformly distributed
among the 7 days of the data week. The data week is that week in the
month containing the fifteenth day. The survey will run from July
1972 through June 1973.

No later than May 31, 1972, each State will receive a listing on
a form titled "Registration Number Register" (see figure 1) showing
selected numbers either for all 12 months of the survey year or for
some of the months. Most States will also receive either a magnetic
tape or card deck containing the listed information. Partial listings
and tapes or cards will be supplemented when sample selection is
completed.

In some States, specific series of registration numbers are
issued to trucks in declared weight classes. In such cases, the entire
registration list was divided into at least three groups--heavy,
intermediate, and light trucks. An independent sample of numbers was
selected from each group. For these States, this will be the final

sample. In some States, when this was not possible, one sample of
numbers was selected from the entire file to yield a sufficient sample of
heavy trucks and an oversample of the lighter trucks. In such instances,
these States will receive instructions on subsampling those numbers
found to have been issued to. the lighter trucks. In the remaining
States, it was possible to divide the registration numbers into two
groups such that all the numbers selected from one group are included
in the final sample but the numbers selected from the other group must
be subsampled. A State that must subsample will also receive a supply
of "Registration Number Register" sheets with no license numbers listed.

No later than 3 weeks before the data week in a given month, it
should be determined which of that month's sample numbers have been
issued (sold) to registrants. Preferably, this will be a computerized
procedure making use of the data on the magnetic tape or card deck
received from the Washington office. Otherwise, a manual search will
be required. In States requiring subsampling, the sold registration
numbers should be listed in specified weight groups and subsampled to
produce the final sample for the month. In these States, the names and
addresses of the registrants in the final sample should be recorded by
hand on the supply of blank Registration Number Register sheets. The
final sample of registration numbers for States not requiring sub-
sampling will be preprinted on the Registration Number Register pages
received from the Washington office. The names and addresses of the
registrants should be recorded opposite their respective license numbers
in the Registration Number Register. Progress of each operation
thereafter should be controlled by entries on the sheets of the
Registration Number Register containing the final sample.

Approximately four days before the assigned travel day,
registrants should be mailed questionnaires. One to two days before
the assigned travel day, reminder letters should be sent to the
registrants. One week after the assigned travel day, nonrespondents
should be mailed follow-up questionnaires.

Returned questionnaires should be given a preliminary acceptance
review and corrected by telephone in specified cases. For the first 2
months of the survey, all of the respondents who indicate that their
vehicles made more than 9 stops on the assigned travel day will be
mailed supplementary questionnaires requesting information for the
additional stops. This procedure is designed to verify the assumption
that all of the stops of a multistep vehicle have the same character-
istics.

Each Friday, questionnaires that have passed the review operation
should be mailed to the Washington office together with a summary
report on Form TCFS-3, Shipment Record (figure 9).

II. DETAILED PROCEDURES

A. Determination of Registrants

Each State will receive a listing showing the selected
registration numbers, the month for which those numbers were selected
and the day of the data week for which information is desired. The
listing form is the Registration Number Register. Each registration
number will be associated on the listing with a unique four digit
sequence number. That association must be maintained at all stages
of the study.

To facilitate the search of computerized Department of Motor
Vehicles registration files, each State will also receive a magnetic
tape or card deck with the same data as the listing. in addition,
States that must select a subsample from the listed numbers will receive
a supply of Registration Number Register forms on which no registration
numbers are listed. A. supply of Registration Number Register pages
will be provided for each of the 7 days of the data week.

The format of the tape or the card deck is as follows:

Number of
Column Characters Description

1-2 2 State Code
3-4 2 Month Code (January = 01, etc.)
5 1 Day Code (Sunday = 1, etc.)
6-10 5 Sequence Number (right justified)
11-20 10 License Number (left justified)

When this manual went to press, the license numbers were recorded
with no special characters such as dashes. For example, AX-1987 was
recorded as AX1987. However, a memorandum was sent out requesting
information on any special format which should be incorporated in the
records. If possible, we will revise the formats before the tapes or
card decks are sent out.

1. Procedure when not Required to Subsample

No later than three weeks before the data week for a given month,
the names and addresses of the registrants and the registered weights of
the vehicles should be determined.

a. Computerized Search of the Registration Files

If the motor vehicle registrations are kept in a
computerized file, the Department of Motor Vehicles should be
asked to provide a computer listing of the names and addresses of
the registrants whose license numbers are on the sample listing.

The Department of Motor Vehicles can use the magnetic tape or
card deck as input to a search program. For each license number on
the tape or on the cards, the Department of Motor Vehicles should
provide a listing of the name and address of the registrant and the
registered weight of the vehicle. The Department of Motor Vehicles
listing should also indicate which, if any, of the registration numbers
have not been sold. The Department of Motor Vehicles should also print
out the State Code, the Month Code, the Day Code and the Sequence Number
from the input tape or card deck, in addition to the registrant's name
and address and the vehicle's registered weight.

Transcribe the names and addresses and weights from the
Department of Motor Vehicles computer listing to the Registration
Number Register, matching license numbers. Note any unsold
registration numbers by writing "UNSOLD" instead of the name and
address in the Registration Number Register.

If only one registration basis is applied for all truck
numbers listed on a page, write the description of the registration
basis in the space provided in the upper right of each such Registra-
tion Number Register page. Otherwise, record the coded registration
basis for each license number on that page in red above its registered
weight. Use the following codes:

Code Registration Basis

A Gross weight of vehicle
B Empty weight
C Chassis weight
D Gross weight per load carrying axle
E Owner's declared capacity

b. Manual Search of the Registration Files

If computerized motor vehicle registration files are not
available, a manual search of the files will be necessary. There
may be two sets of files maintained at the Department of Motor
Vehicles. One may be by name and address, the other by license number.
Use the latter file. Look up each license number in the sample and
copy the registered names and addresses and weights directly into the
Registration Number Register. Note any unsold registration numbers by
writing "UNSOLD" instead of the name and address in the Registration
Number Register opposite that license number.

If only one registration basis is applied for all truck numbers
listed on a page, write the description of the registration basis in
the space provided in the upper right of each such Registration Number
Register page. Otherwise, record the coded registration basis for each
license number in red above its registered weight. Use the following
codes:

Code Registration Basis

A Gross weight of vehicle
B Empty weight
C Chassis weight
D Gross weight per load carrying axle
E Owner's declared capacity

2. Procedure when Required to Subsample from the Entire List

No later than three weeks before the data week for a given month,
determine the names and addresses of the registrants and the
registered weights of the trucks.

a. Computerized Search of the Registration Files

The Department of Motor Vehicles can use the magnetic tape
copy of the listing or the card deck as input to a search program.
For each license number on the tape or cards, the Department of Motor
Vehicles should provide a listing of the name and address of the
registrants; the registered weight of the vehicle; and the corresponding
State Code, Month Code, Day Code, and Sequence Number from the input
tape or card deck, Registration numbers that have not yet been sold
should be identified on the Department of Motor Vehicles listing.
The Department of Motor Vehicles should list the registration data
separately by weight groups and by day code within weight group as
specified in the subsampling instructions sent to the State.

Subsample each weight class on the Department of Motor
Vehicles listing according to the instructions that will be sent.
For each registration number selected in the subsample, record the
following information on a blank Registration Number Register page,
matching the day code with the day of the data week in the heading of
the Registration Number Register page.

(1) Sequence number
(2) License number
(3) Sampling fraction for that weight class: for example,
1/10
(4) Registered weight
(5) Name and address of registrant

If only one registration basis is applied for all truck
numbers listed on a page, copy the description of the registration basis
in the space provided in the upper right of the Registration

Number Register pages for the final sample, otherwise, record the
coded registration basis for each license number in red above its
registered weight, Use the following codes:

Code Registration Basis

A Gross weight of vehicle
B Empty weight
C Chassis weight
D Gross weight per load carrying axle
E Owner's declared capacity

b. Manual Search of the Registration Files

If computerized motor vehicle registration files are not
available, a manual search of the files will be necessary. Look up
each license number on the listing received from the Washington office.
Copy the registered weight of each vehicle. Write "UNSOLD" if a
license number has not been assigned to a registrant. Assign a code
to each registered weight according to its weight class as indicated
in the following example. The weight classes for each State asked to
subsample will be specified in the subsampling instructions sent from
the Washington office.

Weight Class Code

0 - 12,000 pounds 1
12,001 - 36,000 pounds 2
36,001 pounds or more 3

Write the appropriate code for the weight class in red above each
registered weight on the listing. Work with each weight class
separately. Find the sampling fraction for the weight class and
select the license numbers for the final sample counting from the
random start number for that class. The random start number will be
supplied with the other subsampling instructions. Copy the registration
information for each license number selected in the subsample to one
of the blank pages of the Registration Number Register, matching its
day code with the day of the data week, which is preprinted in the
heading of the Registration Number Register page. Write in the sampling
fraction for that weight class in the appropriate column.

If only one registration basis is applied for all truck numbers
listed on a page, copy the description of the registration basis in
the space provided in the upper right of each such Registration Number
Register page for the final sample. Otherwise, record the coded
registration basis for each license number in red above its registered
weight in the Registration Number Register. Use the following codes:

Code Registration Basis

A Gross weight of vehicle
B Empty weight
C Chassis weight
D Gross weight per load carrying axle
E Owner's declared capacity

3. Procedure when Required to Subsample Part of the List

In some States, the sample design requires that only part of the
registration numbers on the listings received from Washington be
subsampled. The registration series sequence that require subsampling
will be distinguishable by their assigned 4-digit sequence numbers.
Specific instruction, identifying the series sequences to be sub-
sampled, and the procedures to use will be sent from the Washington
office.

For those series sequences that do not require subsampling,
follow the instructions in Section II.A.1 of this manual.

4. Procedures at the End of the Registration Year

The year of data collection for this study will,in general, not
coincide with the truck registration year of most States. This will
create a problem in determining the names and addresses of registrants
except in States where reregistrants receive a validating sticker to
place on their old license plates. For all other States, the following
procedures have been devised.

a. First Condition

The data period falls within the grace period when trucks
may or may not carry the licenses for the new registration year.

Instead of the model Letter of Explanation, shown in figure
3, enclose with a questionnaire the model letter shown in figure 4.
Mail these to the registrant who was issued the license tags for the
old year. Do not try to locate the new owner of the license number
since the registration files may not be up to date because of the
crush of new registrations.

b. Second Condition

The data period falls within the first full month of the
new registration year.

During this first month, it is expected that updating of
the files has not caught up with the mass of new registrations. Use
the same procedure as for the first condition.

c. Third Condition

The data period falls within the second full month of the
new registration year.

It is expected that updating of the files has eliminated
the backlog. Use the model letter shown in figure 3. Send it together
with a questionnaire to each selected registrant shown in the new file.

B. Forms to be Used

In addition to the Registration Number Register (figure 1) and
the tape or card deck, the Federal Highway Administration will send each
State a supply of questionnaires, Form TCFS-2. A copy of the front and of
the back of this one-sheet form is presented in figures 2(a) and 2(b).

Additional forms used in this study are shown in the
following figures:

Figure 3 Letter, signed by a State official, which should
be enclosed with the questionnaire when mailed to
the registrants.

Figure 4 Letter, signed by the same official, which should
be mailed to registrants when the data week falls
within the grace period or within the first month
of the new registration year.

Figure 5 Reminder letter, signed by the same official,
which should be mailed to reach registrants on
the data day or within one day of the data day.

Figure 6 Follow-up letter, signed by the same official,
to be sent together with another questionnaire to
respondents who have not returned their question-
naire by one week after the data day.

Figure 7 Letter, signed by the same official, which should
be enclosed with a questionnaire when mailed to
lessees.

Figure 8 Letter, signed by the same official, which should
be sent with a supplemental questionnaire to those
registrants who report multistep operation on the
data day,

Figure 9 Form TCFS-3, Shipment Record.

C. Mailing Procedures

As the names and addresses of the final sample of registrants are
determined, prepare questionnaires for mailing. On the section A side
of the form, enter in the spaces provided the name of the State, the
license number, and the data day and the date. At the top of the
section B side of the form, enter in the space provided the data day
and date.

Address envelopes and stuff each with a prepared questionnaire,
an explanatory letter, and a postpaid return envelope. Mail the
stuffed envelopes preferably four days before the data day but at any
rate not longer than seven days before the data day. The following
listing shows the recommended mailing day limits for each data day.

Preferred Earliest Acceptable
Data Day Mailing Day Mail Day

Sunday Preceding Wednesday Preceding Monday
Monday Preceding Thursday Preceding Monday
Tuesday Preceding Friday Preceding Tuesday
Wednesday Preceding Friday Preceding Wednesday
Thursday Preceding Monday Preceding Thursday
Friday Preceding Monday Preceding Friday
Saturday Preceding Tuesday Preceding Monday

One or two days before the data day, mail a reminder letter
(figure 5) to each registrant. If a response has not been received
from a registrant at the end of a one-week period, counting the data
day as the first day of the period, a follow-up questionnaire suitably
prepared should be mailed to the registrant together with the appropriate
letter (figure 6) and a postpaid return envelope. The following listing
shows the days for mailing the reminder letters and the follow-up
questionnaires for each data day.

Day for Follow-up Mailing
(Generally 1 Week
Data Day Mail Reminder After Data Day)

Sunday Preceding Friday Monday after Following Sunday
Monday Preceding Friday Following Monday
Tuesday Preceding Monday Following Tuesday
Wednesday Preceding Tuesday Following Wednesday
Thursday Preceding Wednesday Following Thursday
Friday Preceding Thursday Following Friday
Saturday Preceding Friday Monday after Following Friday

D. Special Handling of Some Returns

1. Leased Vehicles

If a questionnaire is returned by a registrant who leases his
vehicle to a firm, look for the name and address of the lessee. If
it is not provided, telephone the registrant to obtain this information.
Address a questionnaire to the lessee enclosing a Lessee's Letter of
Explanation (figure 7) and a postpaid return envelope. If the leased
vehicle is reported during the month preceding or the month after the
end of the registration year, the registrant will provide the renewal
license number. Line out the license number for the old year and enter
the new license number instead in the Registration Number Register.
From this point on, the questionnaire should be controlled by the new
license number. Write in the license number and the assigned travel
day and date in the spaces provided on the questionnaire. In the
remarks column of the Registration Number Register, write: Leased to: Name and Address of
Lessee opposite the license number. Check the "Mailed Questionnaire"
column of the Registration Number Register in red to indicate that a
questionnaire has been mailed to the lessee. This column should then
contain two check marks for that license number. Mail a follow-up
letter and questionnaire to those lessees who have not returned their
original questionnaires by one week following the assigned travel day.
Check the column of the Registration Number Register in red for each
lessee sent a follow-up questionnaire.

2. Multistop Vehicles

When reviewing the returned questionnaires for July and August,
look at the bottom of Section B for the number of stops made on the
data day. If more than 9 stops were reported, hold aside the question-
naire and mail the registrant supplemental questionnaires so that
information for the remaining stops up to 27 stops can be obtained.
For example, if the total number of stops made is 15, only one
supplemental questionnaire needs to be mailed. However, if the total
number of stops made is 42, two supplemental questionnaires should be
mailed. These, in addition to the original questionnaire, will provide
information for 27 stops--9 on the original questionnaire and 9 each on
the two supplemental questionnaires.

It is expected that very few of the respondents will indicate
more than 9 stops during the 24-hour period, Most, if not all, of the
stops should be of the urban pickup-delivery type. Telephoning
respondents, generally business firms, should be relatively simple. It
is suggested that when the supplemental questionnaires are mailed, the
respondents should be telephoned and asked to cooperate.

Mark through the Section A side of each supplemental questionnaire,
since the registrant has already supplied this information. However,
enter the license number and the data day in the spaces provided on
both sides of the supplemental questionnaire. Mail the supplemental
questionnaire with an appropriate letter, see model shown in figure 8.
Check the "Mailed Questionnaire" column of the Registration Number
Register in green when a supplemental questionnaire has been mailed to
a registrant. Check-in the returned supplemental questionnaire in green
also. Staple it to its original questionnaire and handle it throughout
the review and shipment operations as a single unit. If the supple-
mental questionnaire is not returned by two weeks after it was mailed,
telephone the registrant to remind him to supply the additional information.

3. Miscellaneous Situations

When mail returns provide the information listed below, take the
steps indicated.

Vehicle wrecked Write "wrecked" in Remarks column
of Registration Number Register.

Vehicle sold If State is one where the
original registrant keeps the
plate, enter "vehicle sold and
not replaced" in Remarks column.
If State is one where
plate goes with vehicle, use
telephone to find if new owner
resides in the study State. If
he does, get his name and address
and send him a questionnaire.
This study is based on a
sample of registrations and,
therefore, we must obtain data
for any vehicle that has been
assigned a selected license
number.

Registrant claims that Check motor vehicle department
is not his license files again. If no error is
number found, tell respondent so with
another request for data. If an
error is found in the transcription
process, send a questionnaire to the
correct registrant and change the
Registration Number Register
appropriately. Otherwise, write
"error registration" in last
column of Registration Number
Register.

Registrant is no longer This category can be reduced
at registered address if mailing envelope carries
request: "Postmaster: Please
forward if addressee has moved."

Undeliverable or Check files of motor vehicle
wrong address department. If correct address
cannot be determined, write "error
address" in Remarks column of
Registration Number Register.

Information not Accept this response. Write
available - "rental vehicle" in last column
Rental vehicle of Registration Number Register.

E. Review of Responses

When mail returns contain fully completed or partially completed
questionnaires, determine which of the questionnaires show that the
vehicles were used on travel day to carry commodities, Since this
survey is aimed primarily at obtaining information on commodities
carried and on origin and destination of vehicle movements, your
review will be aimed solely at determining that a respondent has
supplied those desired data. If he has not, an effort by telephone
must be made to correct or supplement a respondent's return. The
following indicates some of the responses requiring telephoning.
The examples should serve as a guide. They are not intended as a
listing of all possibilities. It is hoped and expected that
those assigned to review returns will have good analytical abilities
and will exercise them when reviewing returned questionnaires.

Example 1 Question 8, Section A, shows that the vehicle was
empty at 2 a.m. Section B, first stop, shows a
commodity delivered. It is possible that the
truck was parked at a loading platform. On the
other hand, it may have been driven empty to a
pickup point and that trip should be the first.

Example 2 Question 7, Section A, shows that the truck was
in a city at 2 a.m. The first stop, Section B.
does not contain any data to locate the vehicle.
(Name of city or town is needed if in an incor-
porated place. Name of county is needed if
destination was not in an incorporated place.
Name of State can be deduced. Street address is
not required but desirable for city destinations.)
Data given for the first stop show that the trip
distance is 10 miles to a truck terminal. If
the city named in question 8. Section A, is one
of

several hundred thousand population or more, a
trip of 10 miles can end up inside the city limits.
If the place in question 8 is small, say 10,000
population, then a trip of 10 miles will most
likely take the truck beyond that city.

Example 3 One of the stops shows that the truck picked up
textiles at a warehouse. There is no indication
whether the textiles were yard goods, manufactured
clothing, etc. Therefore, commodity will not be
codable in the detail desired. The desired detail
for commodity data may be found by referring to
the Appendix of the Truck Weight Study Guide issued
by the Federal Highway Administration.

Example 4 The data in Section B show three stops but no
return to home base. That information is needed
especially if the truck was loaded on its return.

Example 5 Section B shows data for 9 stops. Truck is
evidently engaged in urban multistep delivery
and pickup service. No information is given on
total number of stops. This must be obtained, if
available.

When data on origin, destination, number of stops, or commodity
require correction or additional detail, the telephone should be
employed. The respondent's telephone number should have been entered
on the questionnaire in item 11 of Section A. If it is not, the
telephone book or telephone information service should be able to supply
the telephone number in most cases.

F. Shipping Procedures

Each Friday, prepare the acceptable questionnaires for shipment to
Washington. Copy the license numbers from the acceptable questionnaires
to Form TCFS-3, Shipment Record. Locate the line in the Registration
Number Register that matches that license number. Copy the sequence
number, sampling fraction (if applicable), the registered weight,
and the registration basis from the Registration Number Register to
the Form TCFS-3. Also, copy any explanations from the remarks column
of the Registration Number Register to the Form. Check the "Shipped
to D.C." column of the Registration Number Register. Bundle the
questionnaires for mailing and enclose a copy of the Shipment Record-
form on which they are listed. Maintain a copy of the Shipment Record
form in your office files. Ship the questionnaires to:

Mr. R. T. Messer
Chief, Program Management Division
Room 3300,, Nassif Building
Federal Highway Administration
Washington, D.C. 20590

When all of the listings in the Registration Number Register for
a given month have been checked "Shipped to D.C.," or not later than
the last day of the following month, mail the Registration Number
Register, any unacceptable questionnaires for that month, and the
listings from which any subsampling was done to Mr. R. T. Messer.
The remarks column of the Registration Number Register should contain
entries of "never returned," "refused," "error address," and other
appropriate descriptions of specific causes of failure to obtain a
completed questionnaire.

Click HERE for graphic.

Figure 1

Click HERE for graphic.
Figure 2A

Click HERE for graphic.

Figure 2B

OMB No.
04-S72008
Dear Truck Owner:

As a highway user you are concerned with having available a
highway system that will satisfactorily meet your needs. To
provide and maintain such a system requires planning based on
adequate information on current usage of highways. Vehicle owners
can provide much of the needed information.

The [(State) Department of Transportation] in cooperation with
the [(State) Highway Department Federal Highway Administration of the
U.S. Department of Transportation is conducting a survey to obtain
reliable information an highway usage by truckers based on a sample
of truck registrations. Your help is needed to tell us about the
movements of a truck that was registered in this State and assigned
license plates with the registration number shown on the enclosed
questionnaire. The questionnaire is concerned with the movements of
that vehicle during the 24-hour period starting at 2:00 a.m. on the
date shown in the upper portion of the questionnaire. Please answer
the questions and return the questionnaire in the postpaid return
envelope provided.

The vehicle may possibly be away from home base for several days
and you may not have the information to answer all of the questions.
If that is the case, please complete the questionnaire as soon as
possible, and mail it back.

If the vehicle is being leased by you to another person, please
write his name and mailing address at the bottom of this sheet so
that a questionnaire can be sent to him. You need not answer your
questionnaire, but please return it to us immediately.

The information that you supply will be held in strict
confidence. It will be used for research and planning purposes
with the aim of improving the highway transportation network.

If you have any questions concerning the completion of the ques-
tionnaire please feel free to call us at___________________.

Sincerely yours,

Enclosures
MODEL LETTER OF EXPLANATION

Figure 3

OMB No.
04-S72008

Dear Truck Owner:

The [(State) Department of Transportation] in cooperation with
[(State) Highway Department ]
the Federal Highway Administration of the U.S. Department of
Transportation is conducting a survey to obtain reliable information
on highway usage by truckers based on a sample of truck registrations.

Your help is needed to tell us about the movements of a truck
that was registered in this State during the year ending
____________________and assigned license plates with the registration
number shown on the enclosed questionnaire. Even though you may have
reregistered the truck and obtained different license plates, please
provide the information requested for the truck which bore the license
number shown on the questionnaire. The questionnaire is concerned with
the movements of that vehicle during the 24-hour period starting at
2:00 a.m. on the date shown in the upper portion of the questionnaire.
Please answer the questions and return the questionnaire in the postpaid
return envelope provided.

If the vehicle is being leased by you to another person, please
write his name and mailing address and the new license number at the
bottom of this sheet so that a questionnaire can be sent to him.
You need not answer your questionnaire, but please return it to us
immediately.

The information that you supply will be held in strict confidence.
It will be used for research and planning purposes with the aim of
improving the highway transportation network.

If you have any questions concerning the completion of the ques-
tionnaire please feel free to call us at ____________________________.

Sincerely yours,

Enclosures

MODEL LETTER TO BE SENT TO REGISTRANTS DURING CHANGEOVER REGISTRATION
PERIOD

Figure 4

OMB No. 04-S72008

License
#___________________

Dear Truck Owner:

Recently we sent you a questionnaire requesting information
about a truck which you registered in this State and bearing the
license number shown above. Because of the press of other interests,
some who received the questionnaire may have forgotten to answer it.
Therefore, we are mailing this reminder to everyone to whom we sent
the questionnaire. Please disregard this reminder if you have already
answered the questionnaire and mailed it.
Sincerely yours,

MODEL REMINDER LETTER
Figure 5

OMB No.
04-S72008

License
#__________
Dear Friend:

We have not yet received your response to our request for
information concerning the use of a vehicle registered in this
State and assigned the license number shown above.

in case you have mislaid the original questionnaire, we enclose
another copy of the questionnaire and a copy of our letter of
explanation. Please complete the questionnaire and return it
in the postpaid return envelope enclosed. If you have already
mailed your reply, you may discard this material and send us
nothing more.

If you have any questions concerning the completion of the ques-
tionnaire, please feel free to call us at ___________________.

Sincerely yours,

Enclosures

MODEL FOLLOW-UP LETTER
Figure 6

OMB No.
04-S72008

Dear Friend:

As a highway user you are concerned with having available a
highway system that will satisfactorily meet your needs. To
provide and maintain such a system requires planning based on
adequate information on current usage of highways. Vehicle
operators can provide much of the needed information.

The [(State) Department of Transportation] in cooperation with
[(State) Highway Department ]
the FHWA of the U.S. Department of Transportation is conducting a
survey to obtain reliable information on highway usage based on a
sample of truck registrations.

Your help is needed to tell us about the movements of a truck or
tractor that you lease. The name of the lessor and the license number
on the vehicle are shown below.

The enclosed questionnaire is concerned with the movements of
that vehicle during the 24-hour period starting at 2:00 a.m. on
the date shown in the upper portion of the questionnaire. Please
answer the questions and return the questionnaire in the postpaid
return envelope provided.

The information that you supply will be held in strict
confidence. It will be used for research and planning purposes
with the aim of improving the highway transportation network.

if you have any questions concerning completing the
questionnaire, please feel free to call us at _______________.

Sincerely yours,

Name of lessor___________________

Address__________________________

License #________________________

Enclosures

MODEL LETTER OF EXPLANATION TO LESSEES
Figure 7

OMB No.
04-S72008

Dear Friend:

Thank you for completing our questionnaire for the Nationwide
Truck Commodity Flow Study. You have already provided information
for the first nine stops made by your truck on (day and date) We
now ask your further cooperation in supplying information on stops
after the ninth stop of your
truck during that same period.

We have assumed that the information we would get for the later
truck stops would be very similar to what we are getting for the
first nine stops. However, this assumption must be substantiated,
The value of the information you and other respondents were good
enough to supply will be reduced unless we can show to what extent
the assumption is justified.

Please fill out the Section B side of the enclosed questionnaires
for the remaining stops, but for no more than 27, if that many. Start
with the tenth stop.

As before, we assure you that the information you supply will be
held confidential.

Direct any questions you may have to Thank you again for your
cooperation.

Sincerely yours,

Enclosures

MODEL LETTER TO ACCOMPANY SUPPLEMENTAL QUESTIONNAIRE(S)
Figure 8

Click HERE for graphic.

Figure 9

Innovations in
U.S. Department Public Involvement
of Transportation for

Federal Highway Transportation
Administration Planning

�
�����������������������������������������������������������������
� January 1994

This page intentionally print blank

Innovations in Public Involvement
for Transportation Planning

� January 1994
�����������������������������������������������������������������
�

Notice
�����������������������������������������������������������������

This document is disseminated under the sponsorship of the
Department of Transportation in the interest of information
exchange. The United States Government assumes no liability for
its content or use thereof.

The contents of this report reflect the views of the contractor,
who is responsible for the accuracy of the data presented herein.
The contents do not necessarily reflect the official policy of the
Department of Transportation.

This report does not constitute a standard, specification, or
regulation.

The United States Government does not endorse products or
manufacturers. Trade or manufacturers' names appear herein only
because they are considered essential to the object of this
document.

�
���������������������������������������������������������������Ĵ
Federal Highway Administration Federal Transit Administration�

�����������������������������������������������������������������
Innovations in Public Involvement
for Transportation Planning
�����������������������������������������������������������������

TABLE OF CONTENTS
Page

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

Technique A: Charrette . . . . . . . . . . . . . . . . . . . . A-1

Technique B: Visioning . . . . . . . . . . . . . . . . . . . . B-1

Technique C: Brainstorming . . . . . . . . . . . . . . . . . . C-1

Technique D: Citizens' Advisory Committee. . . . . . . . . . . D-1

Technique E: Transportation Fair . . . . . . . . . . . . . . . E-1

Technique F: Focus Groups. . . . . . . . . . . . . . . . . . . F-1

Technique G: Collaborative Task Force. . . . . . . . . . . . . G-1

Technique H: Media Strategies. . . . . . . . . . . . . . . . . H-1

Technique I: Facilitation. . . . . . . . . . . . . . . . . . . 1-1

Technique J: Citizen Surveys . . . . . . . . . . . . . . . . . J-1

Technique K: Telephone Techniques. . . . . . . . . . . . . . . K-1

Technique L: Video Techniques. . . . . . . . . . . . . . . . . L-1

Technique M: Public Meetings/Hearings. . . . . . . . . . . . . M-1

Technique N: Americans with Disabilities . . . . . . . . . . . N-1

�
�����������������������������������������������������������������
�

INTRODUCTION

Public involvement in transportation planning has a new
emphasis since Congress passed the Federal intermodal Surface
Transportation Act of 1991 (ISTEA). Federal regulations to implement
ISTEA call for proactive public involvement processes. They must
respond not only to the requirements of ISTEA but also those of
related Federal acts, such as the Clean Air Act and the Americans with
Disabilities Act.

This set of notebook pages has been prepared to introduce agencies to
some practical techniques of public Involvement that can be used in
a variety of situations. It is geared to the needs of State agencies
and metropolitan planning organizations (MPO's), particularly some
smaller MPO's with less extensive public involvement experience. It
is intended for use both by public involvement specialists and the
others who have public involvement responsibilities. It is not the
whole answer to public involvement but a starting point to stimulate
responsiveness to ISTEA. Techniques should always be tailored to
local conditions and should be as creative and fresh as possible to
attract public interest.

In these brief leaflets, certain familiar and established techniques
are included for several reasons. First, they may be useful in areas
where they have not yet been tried. Furthermore, even though they are
familiar, some of these techniques are being used in innovative ways
to fulfill the objectives of ISTEA. The more traditional approaches
are supplemented by newer, less familiar techniques that may provide
unique approaches to involving the public. Agencies tiny want to
combine techniques to achieve the maximum impact in encouraging
involvement of the public.

Arranged in a random order, each of these leaflets outlines the
fundamentals of a technique, along with its advantages or drawbacks,
its potential applications and special uses, its utility to agencies
and citizens, and its resource requirements. There are examples of
how these techniques are being applied across the country, along with
telephone numbers for agencies where the technique is being used.

�������������������������������������������������������Ĵ Page 1
Federal Highway Administration Federal Transit Administration

Leaflet A
�����������������������������������������������������������������
CHARRETTE
�����������������������������������������������������������������
INNOVATIONS IN PUBLIC INVOLVEMENT FOR TRANSPORTATION PLANNING

What's a charrette?

A charrette is a meeting to resolve a problem or issue. Within a
specified time limit, participants work together intensely to reach
a resolution. The sponsoring agency usually sets the goals and time
limit and announces them ahead of time. The charrette leader's
responsibility is to bring out all points of view from concerned
citizens as well as agency representatives and experts.

Here are the usual components of a charrette:

definition of issues to be resolved
analysis of the problem and alternative approaches to
solutions
assignment of small groups to clarify issues
use of staff support people to find supporting data
development of proposals to respond to issues
development of alternative solutions
presentation and analysis of final proposal(s)
consensus and final resolution of approach to be taken.

Why is it useful?

A charrette is problem-oriented. The breadth of background of
participants will assure full discussion of issues,
interrelationships, and impacts. Its time limits challenge people
to rapidly, openly, and honestly examine the problem and help
potential adversaries reach consensus on an appropriate - solution.
For example, charrettes were used to formulate alternatives to a
controversial highway project in Knoxville, Tennessee, and a
downtown plan for Jacksonville, Florida, by guiding business and
civic leaders and neighborhood people to a recommended solution.

A charrette produces visible results. It is often used early in a
planning process to provide useful ideas and perspectives from
concerned interest groups. In mid-process, a charrette can help
resolve sticky issues. Late in the process, it A useful in
resolving an impasse between groups.

How does it relate to ISTEA?

A charrette can support the goals of ISTEA by expanding ISTEA's
basic concept of giving citizens a reasonable opportunity to
comment on transportation planning and programming. It provides a
special, intensive occasion dedicated to hearing comments from
citizens and working with them. It encourages public comment by
being interactive and responsive. It focuses on the generation of
fresh ideas and approaches. It is intended to bring public comment
into the planning process early, rather than at the end.

A charrette can enlarge the degree of public involvement in
transportation, reducing feelings of alienation from government.
It offers citizens interaction with public agencies and allows for
questions to be asked before decisions are made. It supplements,
but does not replace, other kinds of public involvement.

�
���������������������������������������������������������� A-1
�Federal Highway Administration Federal Transit Administration

�����������������������������������������������������������������
CHARRETTE, continued
�����������������������������������������������������������������
INNOVATIONS IN PUBLIC INVOLVEMENT FOR TRANSPORTATION PLANNING

Does a charrette
have special uses?

A charrette calls attention to an issue. It can dramatize:
the need for public attention to resolve an issue;
a deliberately participatory problem-solving process:
a public agency's openness to suggestions;
a search for all possible approaches to a question;
a democratically-derived consensus.

A charrette can generate alternative solutions to a problem. The
setting encourages openness and creativity. All suggestions from
the group-however outrageous-should be examined to encourage
thinking about better approaches. In New Hampshire's Community
Stewardship Program, for instance, volunteer experts are invited by
towns to help assess strengths and weaknesses of town planning.

Who participates?
and how?

Any citizen can participate in a charrette. A wide range of
people with differing interests should attend. Traditional
participants represent organized groups, but individuals with any
stake in the issue should be encouraged to attend.

How citizens participate depends on the charrette leader. An
experienced leader assumes that a range of views is heard. The
leader invites citizens to take a stance and present Their points
of view. All participants are assured an opportunity to speak out,
and the leader should encourage even the most reticent participant
to speak up without fear of rebuke or ridicule. The open, free-
wheeling charrette format encourages enthusiasm and responses.

How do agencies
use the output?

A charrette sharpens agency understanding of the perspectives of
interest groups. Early in project formulation, a charrette offers
a glimpse of potentially competing demands and can be a barometer
of the potential for consensus. Thus it helps generate
alternatives and identify issues. In Minnesota and Alabama, for
example, State agencies respond to the needs of individual towns by
providing experts for weekend charrettes.

Who leads a charrette?

A leader experienced in charrette techniques is a must.
To avoid chaos, a high level of discipline is required in a
charrette. The charrette leader should be familiar with group
dynamics and the substantive issues the group will face. The leader
tailors the setting, background materials, and issues to the
goal of the charrette and elicits participation from all group
members within the allotted time. One or two staff people should be
available for support to the leader and to supply data and
information.

A steering committee usually makes arrangements for the charrette.
It may be composed of representatives of Federal and State
transportation or other agencies, consultants, affected
municipalities, and citizen groups. The steering committee should
agree upon the leader of the charrette.

A-2 �������������������������������������������������������Ĵ
Federal Highway Administration Federal Transit Administration

What does a
charrette cost?

A charrette involves significant resolves. The chief items are
sufficient space and background materials and an experienced leader.
Graphics must be used so that participants can quickly comprehend the
problem and envision alternative solutions. Background materials must
be available at the start of the charrette so that no time is lost in
investigating the problem. If the preparatory work leading to a
charrette is done in-house, it can be time-consuming. If done by a
specialist, it can be expensive.

Staffing should include:

a leader experienced in the charrette technique;
staffers who understand the derivation and use of the data;
staffers who have worked on the problem;
staffers who have worked with applicable policy.

Material can include:

large maps;
overlays to allow sketching on maps;
boards to display applicable data;
large newsprint pads and markers to record ideas;
photographs of sites;
handouts of basic goals/time limits/meeting ground rules;
printed background information with background data.

How is it
organized?

Organization can be a significant task. Depending on the issue's
complexity and the intended length of the event, this work includes:
obtaining agreement on the process;
obtaining agreement on timing;
determining potential participants;
finding an experienced charrette leader;
managing special funding, if required;
seeking out resource people;
sending out invitations and background material well in

advance;
finding an appropriate space for meeting;
handling required publicity;
setting up space to encourage informal discussion;
portraying issues clearly in both verbal and graphic form.

Is it flexible?
A minimum of two hours is essential for a charrette focused on a
modest problem. However, many charrettes are day-long events.

A charrette can occur at any time in the planning process, but
preparation is crucial Advance work can take a month or more,
depending on the issue to be discussed. Charrette materials are
flexible and should be tailored to the focus of the meeting.

�
��������������������������������������������������������������� A-3
� Federal Highway Administration Federal Transit Administration

How does it relate
to other techniques?

A charrette can be combined effectively with other techniques.
When
matched with a citizen advisory committee, it can focus on solving a
specific problem. Paired with the visioning process, it is an
attractive means of eliciting ideas. A charrette can also focus on
a single issue raised during a brainstorming session. In Portland,
Maine, a two-day charrette on the long-range plan followed a
transportation fair.

What are its
drawbacks?

A charrette is a one-time event. Thus, the invitation list and timing
must be thoroughly considered and discussed to maximize through broad-
based participation. Goals must be made clear so the expectations of
the charrette do not exceed possible results. The depth of analysis
from a single short session can be disappointing. Follow-up work must
be carefully considered both before and during the charrette.

When is it most effective?

A charrette can resolve an impasse. During such a use, neutral
participants should be involved to bring fresh ideas for
consideration. When a problem is immediate, a charrette can be
effective because people are vitally interested in the outcome. For
maximum effect, a charrette should have the approval of elected
officials, agency heads, and citizens' groups.
A charrette is also useful:
early in the project;
following a brainstorming session;
when focus on a single issue is required;
when a range of potential solutions is needed.

FOR FURTHER
INFORMATION

American Institute of Architects' Regional/Urban Assistance Team
(R/UDAT),
(202) 626-7358
American Society of Landscape Architects, Community Assistance Team,
(202) 686-2752
Minnesota Design Team, Minnesota Department of Trade & Economic
Development, (612) 297-1291
New Hampshire Community Stewardship Program, (603) 271-2155
Portland, Maine, Area Comprehensive Transportation Committee,
(207) 724-9891
Urban Land Institute's Panel Advisory Service, (202) 624-7133

FOR MORE
COPIES

Federal Highway Administration
Office of Environment & Planning
(HEP-32)
400 7th Street SW
Washington, TX: 20590
(202) 366-2065

Federal Transit Administration
Office of Planning
(TGM-20)
400 7th Street SW
Washington, DC 20590
(202) 366-2360

A-4 �������������������������������������������������������Ĵ
Federal Highway Administration Federal Transit Administration

Leaflet B
�����������������������������������������������������������������
VISIONING
�����������������������������������������������������������������
INNOVATIONS IN PUBLIC INVOLVEMENT FOR TRANSPORTATION PLANNING

What's visioning?

Visioning leads to a goals statement. Typically it consists of a
series of meetings focused on long-range issues. Visioning result in
a long-range plan. With a 20- or 30-year horizon, visioning also sets
a strategy for achieving the goals. Visioning has been used to set
a long-range statewide transportation plan in Ohio, a statewide
comprehensive plan in New Jersey, and a regional land-use and
transportation plan in the Seattle region. It has been used by die
governor of Georgia, acting as "Chief Planner," to create long-range
goals for the State. Central Oklahoma 2020 is a visioning project for
a regional plan.

Priorities and performance standards can be a part of visioning.
Priorities are set to distinguish essential goals. Performance
standards allow an evaluation of progress toward goals over time. In
Jacksonville, Florida, a community report card is used to determine
priorities; each target for the future is evaluated annually. In
Minnesota a statewide report card was used to evaluate the current
status and set up goals and milestones for the future. Oregon
established benchmarks to measure progress toward its long-term goals.

Why is it useful?

Visioning offers the widest possible participation for developing a
long-range plan. It is democratic in its search for disparate
opinions from all stakeholders and directly involves a cross-section
of citizens from a State or region in setting a long-term policy
agenda. It looks for common ground among participants in exploring
and advocating strategies for the future. It can bring in often-
overlooked issues about quality of life. It helps formulate policy
direction on public investments and government programs.

Visioning is an integrated approach to policy-making. With overall
goals in view, it helps avoid piecemeal and reactionary approaches to
addressing problems. It accounts for the relationship between issues,
and how one problem's solution may generate other problems or have an
impact on another level of government It is cooperative, with multi-
agency involvement, frequently with joint inter-agency leadership.

How does it relate
to ISTEA

Visioning significantly expands ISTEA's basic concept of giving
citizens a reasonable opportunity to comment on transportation plan-
ning and programming. It goes beyond the merely reasonable by
maximizing concern for public input and by offering multiple
opportunities for such input. Its ultimate product is an integral
part of the State or regional policy guidance contemplated by ISTEA.

Visioning enlarges the degree of public involvement in transportation,
particularly for long-range plans for a State or region. It expands
the political process by soliciting citizen help in setting
generalized priorities. Similarly, it assists in establishing the
general approaches to improvement programs.

�
��������������������������������������������������������������� B-1
� Federal Highway Administration Federal Transit Administration

�����������������������������������������������������������������
VISIONING, continued
�����������������������������������������������������������������
INNOVATIONS IN PUBLIC INVOLVEMENT FOR TRANSPORTATION PLANNING

Does visioning
have special uses?

Visioning uses participation as a source of ideas in the
establishment of long-range policy. It draws upon deeply-held
feelings about overall directions of public agencies to solicit
opinions about the future. After open consideration of many options,
it generates a single, integrated vision for the future based on the
consideration of many people with diverse viewpoints. When completed,
it presents a democratically-derived consensus.

Visioning dramatizes the development of policies to get people
involved in specific topics such as transportation infrastructure.
In Ohio, the Access Ohio program was designed to establish goals
and objectives for development of transportation projects and
programs. Other States that have used visioning to establish long-
range goals include Kansas, Georgia, Texas, Florida, Iowa, Oregon,
and Minnesota.

Who participates?
and how?

Invitations to participate can be given to all citizens or to a
representative panel. A broad distribution of information is
essential. This information must be simply presented, attractive,
and tendered important and timely. It should also include clear
goals of participation and show how comments will be used in the
process.

Citizens participate through meetings and surveys. A typical
method of involving citizens is through a questionnaire format,
seeking comments on present issues and future possibilities. A
report card filled in with citizens' opinions was used in
Jacksonville, Florida. In Minnesota opinions were elicited through
small or large public meetings at locations distributed
equitably throughout the State. In the Research Triangle region of
North Carolina, participants drew pictures of their vision of the
region's future and of transit opportunities in words and pictures
on wall-sized sheets of paper.

How do agencies
use the output?

Visioning helps agencies determine policy. Through wide-spread
public participation, agencies become aware of issues and problems,
different points of view, and competing demands. Drafting responses
to comments aids in sharpening overall policy and assists in
focusing priorities among goals, plans, or programs. Visioning can
also help surface and resolve conflicts among competing priorities.
A chief governmental official can lead visioning. in several
States the governor has made visioning a cornerstone of State
policy planning for infrastructure investments and State
operational departments. The governors of Oregon, Texas, Iowa,
Minnesota, Georgia, Florida, and New Jersey have fostered visioning
for their States.

Agencies have also led visioning projects. Statewide agencies are
leading new visioning projects in Maine and Hawaii. Regional
agencies are leading visioning projects in Jacksonville,
Indianapolis, and Seattle.

B-2 �������������������������������������������������������Ĵ
Federal Highway Administration Federal Transit Administration

Visioning costs can vary. The chief items are staff time and
materials sufficient to set up and carry out the program. Staff
people should include a leader committed to the process, a community
participation specialist who is well versed in the applicable
policies, and staffers who can interpret and integrate participants'
opinions from surveys and meetings. Meeting materials are minimal but
can include large maps and newsprint pads and markers to record ideas.

If forecasts of information are developed or if alternative scenarios
are to be fleshed out, research and preparation time can be extensive.

How is it
organized?

A specific time period is scheduled to develop the vision
Statement.
The schedule incorporates sufficient time for framing issues,
eliciting comments through surveys or meetings, and recording
statements from participants and integrating them into draft and final
documents.

Visioning staff members are typically assigned from existing agencies
that are familiar with issues and essential contacts to be maintained.

In Minnesota and New Jersey, staff was assigned from the State
planning office; in Jacksonville, Florida, from the Community
Council/Chamber of Commerce; in Ohio, from the Ohio Department of
Transportation.

Is it flexible?

Visioning can be extremely flexible in terms of scheduling and staff
commitments. Scheduling can take weeks or months. Staff can be
temporarily or permanently assigned to the project.

Preparation for visioning is crucial and touches on many complex
issues. Advance work is essential to give time for staff to prepare
the overall program, agendas, mailing lists, questionnaires, and
methods of presentation and follow-up. The visioning program should
be carefully scheduled to maximize citizen input and response time
prior to selecting final policies.

How does it relate
to other
techniques?

The visioning process involves using many techniques of public
involvement. In the Seattle area, the visioning process on regional
growth and mobility futures included the most extensive regional
public involvement effort ever conducted in the area: symposiums,
workshops, newspaper tabloid inserts, public hearings, open houses,
surveys, and community meetings.

Visioning leads toward other public Involvement techniques. As a
policy umbrella, it can precede establishment of a citizen advisory
committee and guide its work in reviewing individual projects or
programs. It can lead to brainstorm sessions or charrettes to solve
individual problems. Visioning can be the basis for public evaluation
and implementation; it led to performance monitoring of State agency
activities in Oregon, Minnesota, Iowa, and Texas, followed by reports
to the public.

�
��������������������������������������������������������������� B-3
� Federal Highway Administration Federal Transit Administration

What are its
drawbacks?

Time and staff requirements are significant
to maintain contact with the numerous citizen participants and carry
the program forward. The numbers of participants varies from 100
community leaders in Jacksonville to an estimated 10,000 citizens in
Minnesota. Iistering to participants can consume several months'
time. Full-time effort is required of staff when the process is in
motion.

The staff needs patience to deal with so many diverse views and
individuals, time and schedule requirements, and complex issues and
interrelationships. Finally, visioning is a one-time event and
remains on a generalized policy level; there is a substantial risk
that the resulting document will not satisfy all interest groups.

When is it most
effective?

Visioning is of maximum use at an early point in the estab-
lishment or revision of policies or goals. Used in this way, it
demonstrates openness to new ideas or concepts that may be
suggested by the public. For maximum effect, a visioning project
should have the active support of elected officials, agency heads,
and citizen groups.

Visioning is useful

to set the stage for short-range planning activities;
to set new directions in policy;
to review existing policy;
When integration between issues is required;
when a wide variety of ideas should be heard; and
when a range of potential solutions is needed.

FOR FURTHER
INFORMATION

Iowa Department of Management (Futures Agenda), State Capitol
Building, Des Moines, IA 50319, (515) 281-3322
Jacksonville Community Council (Quality Indicators for Progress),
Jacksonville, FL, (904) 356-0800
Minnesota Planning (Minnesota Milestones), 658 Cedar Street, St.
Paul, MN 55155, (612) 296-3985
Ohio Department of Transportation (Access Ohio), 25 South Front
St., Columbus, OH 43216, (614) 466-7170
Oregon Progress Board (Oregon Shines/Oregon Benchmarks), 775 Summer
Street, NE, Salem, OR 97310, (503) 373-1220
Puget Sound Regional Council (Vision 2020), 216 First Ave. South,
Seattle, WA 98104, (206) 464-7090

B-4 �������������������������������������������������������Ĵ
Federal Highway Administration Federal Transit Administration

Leaflet C
�����������������������������������������������������������������
BRAINSTORMING

(flow76.html)