Document Sample
download.ctpp.transportation.orgFutureposters00 Powered By Docstoc
					    Origin-Destination Worker Flows at the Census Tract Level                                                                                                                                                                                                               Distance Decay
                                                                                                                                                                                                                                                                            Transportation planners have shown that the number of work trips between places is
                                                                                                                                                                                                                                                                         inversely related to distance. That is, all other things being equal, people tend to live
    Summary of Tract-to-Tract Commuter Flows by Type of Geographic Area.
                                                                                                                                                                                                                                                                         close to where they work. So, for any destination tract, we would expect more trips
    A useful way of comparing the general pattern of tract-to-tract commuter flows
                                                                                                                                                                                                                                                                         from closer tracts and fewer trips from tracts farther away. A general form of this
    across multiple data sets is to summarize them into categories. For this analysis,
                                                                                                                                                                                                                                                                         relationship is:
    tract-to-tract worker flow data from the LED and the census are compared by                                                                                                                                                                                                                                         bd-cd2
                                                                                                                                                                                                                                                                                                              Tw = T0e
    summarizing them into categories of a "flow typology" based on a simple
    classification scheme of the type of geographic area in which the tracts are
                                                                                                                                                                                                                                                                            Tw = Total work flows T0 = Internal trips          d = Distance     b,c parameters to be
    located. Tracts located in the central city of a metropolitan statistical area were
    classified as "urban." Tracts located outside metropolitan central cities but still
    within the urbanized area were classified as "suburban." Tracts located within
                                                                                                                                                                                                                                                                          When c = 0, the number of trips decreases exponentially with increasing distance.
    urbanized areas that are outside any metropolitan area were classified as “non-
                                                                                                                                                                                                                                                                         When b = 0, flows decline exponentially with the square of the distance.
    metro urbanized”. Finally, tracts located in the non-urbanized remainder of
    metropolitan areas or in non-metropolitan territory and not within an urbanized                                                                                                                                                                                      For each of the four geographic area types, the ten destination tracts that attracted the
    are were classified as “rural”. The resulting flow typology comprised intra-urban                                                                                                                                                                                    most work trips in each dataset were analyzed (or 40 total destination tracts from each
    flows, intra-suburban flows, intra-rural flows, and flows between urban, suburban                                                                                                                                                                                    data set). Great circle distances were calculated between the geometric centers of each
    and rural areas. Flows to, from, and within tracts located in non-metropolitan                                                             Average Number of Trips Per Tract-to-Tract Interchange,                                                                   origin-destination pair. Flow between tracts greater than 150 miles apart were excluded
    urbanized areas also were differentiated in the same manner.                                                                               by Type of Flow                                                                                                           from the analysis as anomalies.
                                                                                                                                                                                                Destinations                                                              Regression Results – transformed model (Ln Tw = ln T0 +bd – cd2)
Tract-to-Tract Work Trips, by Type of Flow                                                                                                                                                                 Non-Metro                                                                                           LED                                     Census
                                                                                                                                                  Origins           Urban            Suburban              Urbanized               Rural             Average
                                                           Destinations                                                                                                                                                                                                       Area Type                   Model               R2                    Model               R2
                                                                                                                                               Census                                                                                                                                      ln(Tw) = 10.6 - 0.037d             0.74   ln(Tw) = 11.3 - 0.072d             0.84
                                                                Non-Metro                                              Total                                                                                                                                              Urban
       Origins                 Urban             Suburban       Urbanized                         Rural             Productions                                                                                                                                                            ln(Tw) = 12.3 - 0.11d + 0.0005d2   0.92   ln(Tw) = 13.2 - 0.15d + 0.0005d2   0.92
                                                                                                                                               Urban                        21                     12                      8               17                    18
                                                     Percent of Trips                                                                                                                                                                                                                      ln(Tw) = 9.9 - 0.039d              0.73   ln(Tw) = 10.5 - 0.070d             0.86
                                                                                                                                                                                                                                                                                           ln(Tw) = 11.7 - 0.12d + 0.0005d2   0.94   ln(Tw) = 12.4 - 0.15d + 0.0005d2   0.94
Census                                                                                                                                         Suburban                     14                     19                      9               19                    17
                                                                                                                                                                                                                                                                          Non-Metro        ln(Tw) = 7.9 - 0.028d              0.73   ln(Tw) = 8.5 - 0.058d              0.84
Urban                                 23.4                    7.7                         --              1.8                       32.8       Non-Metro
                                                                                                                                               Urbanized                    12                       9                  109                32                    41       Urbanized        ln(Tw) = 9.2 - 0.08d + 0.0003d2    0.90   ln(Tw) = 10.4 - 0.14d + 0.0005d2   0.95
Suburban                              10.1                   28.4                         --              3.4                       41.9
                                                                                                                                                                                                                                                                                           ln(Tw) = 9.3 - 0.037d              0.75   ln(Tw) = 9.8 - 0.077d              0.83
Non-Metro                                                                                                                                                                                                                                                                 Rural
Urbanized                              0.1                      --                      1.1               0.9                       2.1        Rural                        20                     15                     38               37                    26                        ln(Tw) = 11.1 - 0.11d + 0.0005d2   0.96   ln(Tw) = 12.1 - 0.16d + 0.0006d2   0.95
Rural                                  4.9                    4.0                       1.8              12.5                      23.2
                                                                                                                                               Average                      19                     17                     48               29                    19
Total Attractions                     38.3                   40.1                       2.9              18.7                     100.0
Urban                                 21.6                    8.7                       0.1               2.0                       32.4       Urban                             5                   3                      2                 3                      4   This exploratory analysis found that the total counts of workers in the LED data for the
Suburban                              12.1                   28.2                       0.1               3.5                       43.9                                                                                                                                 State of Illinois and selected Illinois counties, aggregated up from block records on the H-
Non-Metro                                                                                                                                      Suburban                          4                   6                      1                 4                      5   B File, compare favorably with the census and the ACS. However, the LED O-D File
Urbanized                              0.2                    0.1                       1.1               1.0                       2.5        Non-Metro
Rural                                  4.8                    4.5                       1.7              10.2                      21.3        Urbanized                         3                   2                    38                  8                      9
                                                                                                                                                                                                                                                                         contains more than three times the number of tract-to-tract interchanges and nearly 1
Total Attractions                     38.7                   41.6                       3.0              16.8                     100.0                                                                                                                                  million fewer work trips than the census. This results in a much lower average number of
                                                                                                                                               Rural                             5                   4                    11                11                       7   trips per interchange in the LED data compared with the census. But when the tract-to-
Note: In the Census data, 3,471 interchanges with a destination tract code of 999999, accounting for 200,280 trips, are not included in
the table. These include 2,810 trips with urban origins, 1,876 trips with suburban origins, 23,138 trips with non-metro urbanized origins,                                                                                                                               tract interchanges and worker trips are summarized by type of origin-destination
and 174,456 trips with rural origins.                                                                                                          Average                           5                   4                    10                  7                      5
-- Less than 0.5, so rounds to zero percent.
                                                                                                                                                                                                                                                                         interchange or flow (e.g. suburban-to-urban or rural-to-suburban), the proportion of
Sources: 2001 LED O-D. Census 2000 Transportation Planning Package, Illinois, part 3.                                                                                                                                                                                    interchanges and trips by type of flow in the LED data is virtually identical to the census.
                                                                                                                                                Internal Work Trips Within the Same Tract by Area Type
                                                                                                                                                                                                                                                                         The results of this analysis, particularly the comparability of worker flow patterns
                                                                                                                                                                                                      Percent of                   Average                               between the LED and census data, demonstrate the potential utility of journey-to-work
                                                                                                                                                                                        Number of                                                    Maximum
                                                                                                                                               Area Type
                                                                                                                                                                   Number of
                                                                                                                                                                                      Internal Tract
                                                                                                                                                                                                         Trips                      Internal
                                                                                                                                                                                                                                                      Internal           data from the LED Program for use in transportation planning. Because it is national in
                                                                                                                                                                    Tracts                           Originating in                Trips Per
                                                                                                                                                                                                      Area Type                       Tract
                                                                                                                                                                                                                                                        Trips            scope, the LED Program offers many advantages over the piecemeal approach of
                                                                                                                                                                                                                                                                         individual states trying to work with their state employment security agencies to obtain
                                                                                                                                             Urban                          1,087               38,192                     2.7                35               714       and use the employment data. Many states do not allow access to ES-202 data for
                                                                                                                                             Suburban                         907               89,033                     4.7                98               657       research, so the LED Program is a means of accessing the data. The LED methodology is
                                                                                                                                             Non-Metro                                                                                                                   consistent for every state. The file that the LED Program uses to assign residence
                                                                                                                                             Urbanized                         94              16,967                    16.1               181                780       addresses to workers is consistent for every state and contains extensive confidential
                                                                                                                                             Rural                            752             129,607                    14.1               172              1,338       information not available outside the Census Bureau. Data from the LED Program are
                                                                                                                                             Total                          2,840             273,799                     6.4                96                n/a       updated quarterly. The data are cost effective, costing much less than a survey or a census
                                                                                                                                             Census                                                                                                                      special tabulation.
                                                                                                                                             Urban                          1,104             118,075                      6.8              107              2,115
                                                                                                                                             Suburban                         910             185,480                      8.4              204              1,880
                                                                                                                                                                                                                                                                         The transportation data community has embarked on a coordinated effort to study and
                                                                                                                                             Urbanized                         94              26,420                    23.4               281              1,010       adjust to the ACS. It is an opportune time to thoroughly evaluate the utility of the LED
                                                                                                                                             Rural                            754             237,659                    19.4               315              1,890       data as well. Small-area data from the ACS will be based on multi-year averages. If the
                                                                                                                                             Total                          2,862             567,634                    10.7               198                n/a       accuracy of the worker flow data from the LED Program is improved, the data could be
                                                                                                                                             Note: The LED data contained 124 tracts with no internal trips. The Census data contained 102 tracts with no internal
                                                                                                                                                                                                                                                                         used to evaluate and validate the origin-destination data from the ACS. Annual LED data
                                                                                                                                             trips. n/a = not applicable                                                                                                 could be used to measure changes in key worker flows and travel corridors, and provide
                                                                                                                                             Sources: 2001 LED O-D File. Census 2000 Transportation Planning Package, Illinois, part 3.
                                                                                                                                                                                                                                                                         evidence of new travel patterns caused by shifts in the location of employment or new
                                                                                                                                                                                                                                                                         housing developments.

Shared By: