Light Rail Now! - The Track to Better Urban Livability
Home Button Features Button News Button Events Button Facts Button Myths Button About Us Button Contact Us button Links Button Search
Philadelphia bus & rail station>
						<p><font face=
Related Links

Light Rail Now and Walkable Neighborhoods (LRN) can be contacted at:

Light Rail Now!
107 W. Eighth
Austin, Texas 78701
Phone: (512) 494-9841

Research Study: Riders Prefer Light Rail to "Bus Rapid Transit"

Introduction by Light Rail Progress

Opponents of light rail transit (LRT) – in Austin as well as in other cities – repeatedly claim that, as a supposedly cheaper alternative, "buses will do the job". in a makeover currently in vogue, so-called "Bus Rapid Transit" (BRT) (a moniker applied to almost any upgrade from regular, slow, local street-bus service) is promoted by both the Federal Transit Administration and various highway-industry interests and their retinue purportedly as a way to get the service levels and passenger attractiveness of LRT as significantly lower cost.

But, as Light Rail Progress has consistently demonstrated, in almost every case, the substance of these claims dissolves upon closer examination. in particular, for similar types and levels of service, the capital costs of BRT are quite close to (often even greater than) those of LRT. Pittsburgh's new West Busway, for example, is costing in the vicinity of $60 million per mile (for a route predominately on the surface in former railroad right-of-way) – from 1.5 to three times the typical cost of LRT.

Another major rebuttal to the BRT argument is the tremendous cost advantage of LRT: For the same basic character of service, LRT tends to perform at a significantly lower cost per passenger and/or passenger-mile than does the bus. A major factor in LRT's cost advantage – and its appeal to the public, transportation planners, and decisionmakers alike – is light rail's power to attract more passengers than buses. As we've previously noted that, since the 1970s, the steep decline in US transit ridership had been reversed:

A key factor in reversing that trend, of course, has been the expansion and re-installation of rail transit services. Rail transit ridership has climbed at a rate several times that of bus ridership, which has comparatively remained stagnant. Between 1977 and 1997, while motor bus ridership rose 5%, "heavy" rail ridership (mainly on subway/elevated transit) increased 13%, and light rail ridership skyrocketed an astounding 155%.

The main reason: buses, bogged down in increasingly gridlocked street and freeway traffic, have a hard time keeping their passengers. Also, compared to rail, bus transit is less reliable and comfortable. Bus routes are less visible and understandable. Consistently, when [pro-roadway transit operators] have converted rail transit lines into bus routes, the vast majority of the riders have just abandoned the service – the bus line is left with a small fraction of the original ridership. That should tell us something.

Many light rail supporters argue strongly that, for fundamentally equivalent types of transit service, light rail will attract more riders. in other words, the public tends to be drawn by certain specific attributes of LRT service – the permanence of the alignment, vehicle comfort, etc. – in a way and to a degree not exhibited in the case of similar bus operations. The result is substantially higher LRT ridership for a given investment in higher-quality transit (bus or rail).

Some evidence of this is provided in a 1991 research project in Philadelphia, which examined basically similar services of the Southeastern Pennsylvania Transportation Authority's (formerly Red Arrow) suburban trolley lines and the Ardmore busway – relatively fast bus service on a dedicated bus-only roadway. Curiously, the history of the Ardmore busway itself gives some corroboration to the case for the greater attractiveness of LRT, since, when the route was converted from trolleys to buses (i.e., LRT to BRT) in 1967, ridership dropped 15% – despite the replacement of older streetcar equipment by modern, airconditioned buses on a newly paved-over private right-of-way.

[Source: E.L. Tennyson, "North American Busway Experience" (paper), 1999]

To some extent, therefore, the Ardmore experience was revisited in the 1991 study – this time, contrasting BRT (busway) service with that of similarly operating LRT. Excerpts from the basic findings of that study are presented below. You may note that a somewhat curious finding was that the LRT stations – small, solidly built, enclosed waiting shelters predominantly made of stone, and apparently designed to shield occupants from winter conditions – were also attractive to vagrants, who used them as toilets! As a result, passengers tended to find the LRT shelters far less attractive than open-air bus and trolley stops! (Presumably, passengers would have had similar attitudes towards enclosed bus shelters.) Nevertheless, as a whole, passengers surveyed appeared to prefer attributes of LRT to those of BRT to a statistically significant degree.


September 1991


This is an initial report of secondary findings from an onboard survey of riders of selected bus and light rail transit (LRT, or trolley) routes operated by Southeastern Pennsylvania Transportation Authority (SEPTA) in the Philadelphia metropolitan area. The survey was conducted in June and July 1991 under the auspices of the Delaware Valley Citizens Council for Clean Air, with funding contributed by various private and public organizations


The routes selected for surveying have been SEPTA's Media and Sharon Hill LRT (trolley) lines and the Route 103 bus to Ardmore (partly using the SEPTA's private busways in Ardmore and near the 69th St. Terminal). These routes were selected for the project because of the relative homogeneity of their demographic areas and similarity of their types of service, features which present a rarely experienced opportunity to focus on mode-specific behavior and perceptions. All three lines extend from SEPTA's 69th St. Terminal into suburban communities, using a variety of alignments on public thoroughfares, reservations, and exclusive guideways. in addition to providing point-to-point connections between intermediary locations on their routes, both modes serve as feeders to SEPTA's major Market-Frankford rapid transit terminus at 69th St. as well as to other commutative rail services in Philadelphia's nearer western suburbs.

Using these routes' service areas as the focus of study, this survey project has been intended to attempt to identify hypothesized differences in passenger and public perceptions and behavior in regard to light rail vs. bus transit services, with the ultimate aim of improving the accuracy of ridership forecasting models through accounting for mode-specific characteristics. The primary focus is to analyze the actual travel behavior both of the general public in the service areas and of the actual passengers on each route, through the development of rudimentary modal split forecasting models derived from calculations of comparative trip times/costs by transit vs. automobile for both LRT and bus (then comparing the two models). The achievement of this primary study objective is a relatively lengthy, tedious process requiring reasonably accurate origin-destination analysis and lasting several weeks.

Secondary data goals are simpler and less time-consuming, including comparative assessments of sociodemographic data, use characteristics, and mode attribute perceptions for LRT vs. bus riders. Since these subordinate results have been much easier and faster to produce, they are presented in this report.


Conclusions from Preliminary Results

Based on these results, the following general conclusions and recommendations are offered.

LRT tends to attract a greater proportion of nonwork trips than bus for these generally equivalent types of transit service.

Nonmotorized access times to and from LRT stations/stops are significantly greater than those for bus for these similar types of service. This finding, which is consistent with results of some other studies, suggests that LRT patrons tend to be willing to walk (and possibly to bike) further to and from LRT services than is the case for bus, in this instance by a difference of nearly 3 minutes.

This finding also suggests that modifications to ridership forecasting models may be warranted to incorporate assumptions of approximately one-third greater walking-access distances/times for LRT than for bus.

LRT appears to project a more positive image to its passengers than in the case of similar bus service. LRT passengers evidently find LRT routes more understandable, schedule adherence more reliable, and LRT vehicles more spacious and more comfortable overall, producing less odors, fumes, and noise.

Totally enclosed passenger shelters which hide occupants from view offer convenient alterative use as urinals (and possibly other undesirable applications). The image of SEPTA's suburban trolley services – particularly passengers' sense of the attractiveness of stations – could well be enhanced by either by regular cleaning or by reconstruction to eliminate the total privacy the shelters currently afford users.

For these riders, there is evidently no significant difference between bus and LRT in terms of frequency of use, influence on residential location, car availability, and level of household car ownership.

Another surprising finding was that the LRT passengers appear to have significantly lower incomes than the Ardmore bus passengers. This result contradicts findings from some other studies. However, whether it reflects actual demographic differences in the areas served is still be evaluated.

Finally, it should be noted that, while these bus and trolley services offered an opportunity to investigate modal differences under similar service conditions and demographic surroundings, several factors unique to the situation, and quite different from most areas where new transit services such as LRT are being implemented or proposed, must also be considered.

First, the Ardmore bus and Media and Sharon Hill trolley routes operate within a context of relatively very high-level, multi-modal transit service overall. The transit-using public is undoubtedly accustomed to interchanging among a variety of modes, including many rapid-transit-like rail commuter lines of excellent quality. Thus behavior and attitudes relative to the different modes may be somewhat blurred, especially in cases where the bus or trolley is used simply as a feeder to rapid transit or commuter rail.

Secondly, these services have been in operation literally for several generations. Especially in recent times, quality has fluctuated substantially. Public attitudes and behavior can therefore be expected to be somewhat "acclimatized" and in other ways different from those toward totally new services.

Thirdly, the traditional entrenchment of the trolley lines, and to some extent of the Ardmore bus route (including its predecessor trolley service), would also suggest that these systems' influence on demographic characteristics, urban development patters (e.g. residential location), and other aspects of the urban environment would not be as fluid and dynamic as would be the case with new installations.

Rev. 00/08/28


All website material © 2000-2007 Light Rail Now Project (unless otherwise indicated)