MEMORANDUM
DATE June 6, 2013
TO David J. Mohler, Executive Director of Planning
Office of Transportation Planning, MassDOT
FROM Thomas J. Humphrey, Chief Planner, Transit Service Planning
MPO Staff
RE: Full Report on Findings from Comparisons of Results of 2008-09 MBTA Passenger Surveys with Results of Prior Surveys
During 2008 and 2009, the Central Transportation Planning Staff (CTPS) conducted surveys of passengers on all MBTA bus, rail rapid transit, light rail, commuter rail, and water transportation routes. The results were presented in 17 volumes that are available in PDF format on the Boston Region Metropolitan Planning Organization (MPO) website. Although these reports present summary tables based on analyses of a large number of variables, time and budget constraints did not allow for comparisons of the results of these surveys with the results of prior surveys of passengers on the same MBTA services. In the present study, such comparisons were conducted.
The objective of this study was to determine whether there had been significant changes in characteristics of MBTA riders and their use of the system since the 1990s. If there were not significant changes, this would suggest that the 2008-09 results could be relied on for planning purposes for many years. If there were significant changes, this would suggest that surveys should be conducted at more frequent intervals to allow the MBTA to make service changes consistent with changing needs of riders.
A summary of the comparison strategy and the major findings of the comparisons is presented in a separate memorandum. The present memorandum detailed description of the study begins with a discussion of the general methods of the survey comparison. A presentation of the survey comparison results follows. This presentation first explains details of the comparison method that are particularly important to a proper understanding of the results. The results themselves follow, presented by transit mode. The study’s conclusions are given in the final discussion in this memorandum. They are followed by tabulations of the survey comparison results.
For the 2008-09 survey project CTPS, in consultation with the MBTA, created a survey form with six variations reflecting differences in questions applicable to different transit modes. These forms were for rail rapid transit (including light rail), bus, commuter rail (excluding the Greenbush Line1), the Greenbush Line, commuter boat, and inner harbor ferry. Despite the differences in questions, similar standard reports were used to summarize the results of each of these surveys.
Prior to the 2008-09 survey project, surveys on most of the commuter rail system had been last done in 1993. The rail rapid transit and light rail lines had been surveyed in 1994, the bus system in 1995, the Old Colony commuter rail lines in 1998, and water transportation services in 2000. (The Greenbush commuter rail line, which opened in October 2007, was surveyed for the first time in 2009.)
The information needs that can best be met through conducting surveys change somewhat over time. Consequently, not all of the questions on the 2008-09 surveys were directly comparable to questions on prior surveys on the same modes. A review of the reports produced from the earlier survey efforts indicated that results could be compared with those of the following standard reports from the 2008-09 surveys:
For several of the service modes, the groupings of passengers by route segment were not the same in the published reports from the 2008-09 surveys as in the published reports from the prior surveys, and neither set of groupings was optimal for purposes of comparison. The results of the 2008-09 surveys are all stored in Access databases, from which new reports for any desired passenger groupings can easily be produced. The database programs that had been used to produce the standard reports from the earlier survey efforts were no longer available, but the records from these surveys had been exported into Excel spreadsheets. CTPS used standard database management tools to extract information as needed from these spreadsheets in forms comparable to those in the 2008-09 survey reports.
Because only one set of earlier surveys was available for the comparisons for each service mode, it was not possible to determine whether changes that occurred took place gradually over time or were more rapid responses to changes in factors such as service levels, fare structure, or economic conditions. It was also impossible to determine the extent to which the distributions of responses to the survey questions might have fluctuated in the years between the old and new surveys.
It was necessary to make comparisons at fairly high levels of data aggregation. Survey results from rapid transit and light rail lines were examined by line segment. Results from commuter rail and water transportation services were examined by route. Results from buses were examined for all routes combined at each bus garage. Comparisons at these levels could conceal changes occurring at individual rapid transit, light rail, or commuter rail stations or on individual bus routes. On the other hand, sample sizes become progressively smaller at finer levels of detail, the margins of error in the samples increases, and similarities or differences between old and new survey results become less meaningful.
Detailed results of the comparisons of 2008-09 and prior surveys are presented by mode and report in the tables at the end of this memorandum. Major findings from each table are discussed in the text. Comparisons for each of the standard reports, except those for service quality ratings, use percentages, rather than absolute numbers, of respondents in each category on each table. For purposes of discussion, all references to changes in percentages or shares refer to absolute rather than relative changes. For example, an increase from 10% to 12% in the percent of riders in a given category is referred to as an increase of 2% (12% - 10%) rather than an increase of 20% (12%/10%). Absolute changes are more meaningful when working with small numbers, where a fairly insignificant change, such as from 1% to 2%, would be reported in relative terms as a 100% gain. All changes of 3.0% or more are shown in italics.
In the summary reports from 2008-09 surveys, some origin and destination locations were broken down at finer levels of detail than they were in the corresponding previous surveys. For example, in the 2008-09 survey reports, origins and destinations in Cambridge were separated by neighborhood, but in the earlier reports they were shown only for the entire city of Cambridge. For purposes of the survey comparison project, instead of using information from the standard reports from the older surveys, the Excel database records from those surveys were examined, and origins and destinations were separated at the same level of detail as those in the 2008-09 survey reports.
In this memorandum the term “Boston Proper” conforms to the commonly accepted definition of that part of the City of Boston east of Massachusetts Avenue and enclosed by the Charles River, Boston Harbor, Fort Point Channel, the Southeast Expressway, and the Massachusetts Avenue Connector. In the standard reports generated from surveys conducted by CTPS, Boston Proper is subdivided into nine neighborhoods: Financial/Retail District, Government Center, Beacon Hill, North End, Waterfront, Back Bay, Prudential/Hancock, Park Square, and South End.
The 2008-09 surveys and the previous surveys used in the comparisons (except for the 1993 commuter rail surveys) included a question on vehicle ownership per household. In the standard reports generated from these surveys, per-capita vehicle ownership was calculated from the combined responses to the vehicles-per-household question and a separate question on number of people per household.
In the survey quality ratings, respondents rated various service attributes of each mode on a scale of 1 to 5. The comparisons use absolute changes in the average rating for each attribute for each mode.
After some preliminary comparisons of the results of the 2008-09 and 1994 rapid transit and light rail surveys, CTPS concluded that results for each heavy rail rapid transit line should be examined by segment outside Boston Proper, and that the Green Line should be examined by surface route. The reason for this approach was that during the hours that survey forms were distributed, the predominant travel flows on the system were from boarding stations outside Boston Proper to alighting stations within Boston Proper. Many of the entries to stations within Boston Proper represented reboardings by passengers residing elsewhere.
This method of examining the lines resulted in 11 sets of data comparisons for each report. The 11 line divisions were:
In both sets of surveys, home-based work trips (either home to work or work to home) accounted for by far the largest share of trips on each of the 11 route segments examined. On eight of the lines, this share changed by no more than 3.5% from 1994 to 2008-09. However, it increased by 6.6% on the Cambridge (Alewife-Kendall) end of the Red Line, by 6.3% on the surface B Line and by 7.6% on the surface D Line. Most of the increases on the Cambridge Red Line and the B Line reflected reductions in home-based school trips that were at least partly attributable to differences in the survey distribution dates relative to college schedules.
The old and new survey forms allowed passengers to check as many reasons as applied for using MBTA service. The 1994 forms asked about use of the MBTA in general, but the 2008-09 forms asked specifically about use of rapid transit. The 2008-09 forms also included more checkoff choices. “Convenience,” which was the most subjective of the check-off choices in both surveys, was the reason most often selected on each segment in 1994. It was still the most common reason on most segments in 2008-09, but on most line segments the percentage of passengers checking this reason declined somewhat. On the Braintree Branch of the Red Line “Avoid driving and traffic,” which was not a checkoff choice in 1994, was the top choice in 2008-09. The number of respondents checking “Environmentally responsible” as a reason for using the MBTA increased significantly between the old and new surveys, with gains by segment ranging from 12.3% to 21.5%.
On each segment, significantly more riders in the 2008-09 survey checked “Avoid parking at destination” than had checked the most comparable choice (“Cost of parking downtown”) provided on the 1994 form. However, this difference may indicate that parking availability is a more important consideration than parking cost, rather than indicating a change in the influence of parking on choice of mode.
The questions on alternate means of travel on the two surveys were not strictly comparable. In the 2008-09 surveys, the question was what means, if any, the respondent sometimes used instead of the line being surveyed to make the same trip. In 1994 the question was hypothetical as to the means of transportation the respondent would use for the same trip if the surveyed line were not available. Discontinuance of any of the rapid transit or light rail lines was not under consideration in either 1994 or 2008-09, and respondents to the 1994 survey were not provided with information on what other MBTA services might still be available if the surveyed line were not.
On the majority of the rapid transit and light rail lines, the percentage of riders that reported actually driving alone sometimes in 2008-09 was equal to or greater than the percentage that reported in 1994 that they would drive alone if the survey line were not available. Conversely, on the majority of lines the percentage of riders that reported actually using other MBTA services sometimes was lower in 2008-09 than the percentage that reported in 1994 that they would use other MBTA services if the usual line were not available.
Overall, the comparison results suggest that rapid transit and light rail passengers would be more inclined to use private transportation as a temporary alternative to their usual lines than as a permanent one. Passengers already using the MBTA service most convenient for them have little reason to use another MBTA service for the same trip unless forced to, but may alternate with private transportation for reasons such as needing a car to transport work materials, or being offered a ride with someone else.
In both surveys, driving alone was the most common alternative checked on most of the lines. However, on the B Line, in 2009 the most common alternative was “Other”; most of these respondents wrote in walking as the other mode. In 1994 on the B Line, “Other MBTA service” and “Other” were both slightly ahead of “Drive alone.” On the C Line, fewer actually drove in 2008-09, and more used “other” means, mostly walking, compared with the hypothetical choices checked in 1994. In both surveys “Other MBTA service” was the most common alternative checked by E Line passengers. MBTA bus Route 39 serves the same streets as the surface E Line, and some E Line stops are within walking distance of Orange Line stations.
As would be expected, on each line, towns or neighborhoods served directly accounted for the largest individual shares of origins of boarding passengers. Lines with feeder service connections or with substantial parking facilities also attracted riders from farther away, but the origins of such passengers were more dispersed. The analysis method used was first to select the origins accounting for individual shares of 1% or more of the boardings on each line in the 2008-09 survey and then to compare these shares with those from the same origins on the same lines in 1994.
In 2008-09, origins with individual shares of 1% or more of boardings accounted for over 90% of the total boardings on each of the lines examined except for the Orange Line between Oak Grove and Community College (87.5%) and the Red Line between Braintree and North Quincy (83.6%). Both of the latter lines have large numbers of park-and-ride passengers along with feeder bus service from beyond the direct service areas of the stations. In general, the towns or neighborhoods that accounted for 1% or more of rider origins on each line in 2008-09 also accounted for the vast majority of the riders on the same line in 1994, although there were some changes in relative importance of origins. On the Red Line Braintree Branch, the origin share for the city of Quincy, where four of the five stations are located, increased from 48.4% in 1994 to 58.3% in 2008-09. This was a reflection both of diversion of riders from more distant points to the Middleborough/Lakeville and Kingston/Plymouth commuter rail lines, which opened in 1997, and the Greenbush Line, which opened in 2007, and of increases in the absolute number of riders from Quincy. On the Blue Line between Wonderland and Maverick, the share of passengers with origins in East Boston increased from 32.7% to 41.0%, with absolute increases in East Boston riders and absolute decreases in riders from some more distant origins contributing to the change.
On the Orange Line between Forest Hills and Massachusetts Avenue the percentage of riders reporting Jamaica Plain origins increased from 25.5% to 32.2%, with an offsetting decrease in Roxbury origins from 22.7% to 16.3%. However, some of these changes may have resulted from differences in the way that origins close to the border of these adjoining communities were recorded in the two surveys.
On the Mattapan-Ashmont High Speed Line, the 2008-09 surveys did not get a good sample of riders boarding outbound trips at Ashmont. For passengers boarding at all stations combined on this line excluding Ashmont, the share of riders with origins in Milton increased from 27.9% to 37.9%, while the share from South Dorchester decreased from 18.7% to 8.4%.
Between 1994 and 2008-09 all of the Green Line surface branches had absolute decreases in the number of passengers making trips entirely outside the subway. This was a result of the elimination of free outbound fares in 2007. However, the losses in surface-only trips were not uniformly distributed along the routes, so percentages of total ridership showed large decreases from some origins but large increases from others. On the surface B Line, the share of Allston origins decreased from 45.5% to 34.9%, while the share from the BU area increased from 18.9% to 23.9%. On the surface C Line, the North Brookline share of origins dropped from 69.8% to 62.1%, while the share of Brighton origins increased from 7.1% to 14.0%.
The summary reports on access to the rapid transit system both for the 2008-09 survey and for the 1994 survey were based on the results of separate questions about other transit services, if any, used to get to the rapid transit boarding station, and on the means of access to the first transit service used in the trip. Responses from passengers using more than one rapid transit line in a trip were edited if necessary to show the first rapid transit station used in the trip as the boarding station. For example, a survey that originally showed the boarding station as Park Street on the Green Line but indicated access via the Red Line from Harvard Station was edited to show Harvard as the boarding station.
The 2008-09 survey forms included checkoff choices for taxi, bicycle, and shuttle van access, but in the 1994 survey passengers using any of these means of access had to write them in under “Other.” However, these accounted for relatively small shares of access trips to all of the rapid transit and light rail lines in 2008-09. In both surveys, passengers reporting bus access to rapid transit could specify either an MBTA route or a private-carrier route, but in the processing of the 2008-09 survey data, bus-access passengers not specifying a route number or name were tabulated as non-MBTA bus users. However, many of the survey respondents who reported accessing rapid transit by unspecified bus routes could only have made such trips on MBTA bus routes. Therefore, comparisons between the bus access trips in the old and new surveys were based on total bus access rather than MBTA and non-MBTA routes separately.
Several of the heavy rail rapid transit segments had increases of over 5% in the shares of passengers reporting walk-in access. These included the Blue Line between Wonderland and Maverick (35.5% to 47.8%), the Red Line between Braintree and North Quincy (29.3% to 37.8%), the Orange Line between Forest Hills and Massachusetts Avenue (48.6% to 57.0%) and the Orange Line between Oak Grove and Community College (31.6% to 37.7%). These gains were accompanied by decreases in the percentages of riders reporting park-and-ride or bus access. In general, passengers using walk-in access have origins closer to their boarding stations than passengers using park-and-ride or bus access. Therefore, the changes in distribution of access modes would be more closely related to changes in origin locations than to changes in access modes from a given origin area.
Standard reports from the 2008-09 surveys included tabulations of access times by range for walk, park-and-ride, drop-off, and other private access modes and for all of these combined. For purposes of comparison with the 1994 surveys, the combined times for all private access modes were used. On most of the line segments examined, reported average access times by private transportation were slightly longer in 2008-09 than in 1994, but on a few segments average access times decreased. In 1994, the South Shore Branch of the Red Line and the northern end of the Orange Line had the longest average access times, at 14.5 and 12.8 minutes. Both of these segments have several large park-and-ride facilities. In 2008-09 the segments with the longest access times were the same as in 1994, but these times had decreased to 13.4 minutes on the South Shore Red Line and increased to 12.9 minutes on the northern Orange Line. The surface B and C Lines had the shortest average access times, at 3.8 and 4.4 minutes. Access to these segments is predominantly by walking. In 2008-09 these times increased to 4.5 and 5.3 minutes on the B and C Lines.
Lines with significant increases in walk-in access showed decreases in the percentages of riders with access time of 5 minutes or less and increases in access times of 6 to 10 minutes or 11 to 15 minutes. The percentages of trips with private access times of over 20 minutes did not increase significantly on any of these lines, indicating that new walk-in trips were still mostly from locations within what would have been expected to be the walk-in attraction areas of stations.
The summary reports on egress from the rapid transit system both for the 2008-09 survey and the 1994 survey were based on the results of separate questions about other transit services, if any, used to get to the destination from the final rapid transit station used in the trip, and on the means of egress from the final transit service of any kind used in the trip.
On the survey forms, the checkoff choices for means of egress from the transit system were essentially the same as those for access modes, with necessary changes in wording such as “Met at station” in place of “Dropped off.” In the published reports from the 1994 and 2008-09 surveys, egress modes and times were tabulated by rapid-transit or light-rail alighting station or stop, while access modes and times were tabulated by boarding station or stop. Because the line segments selected for study in the survey comparison project were outside of Boston Proper but the majority of passengers were going to destinations served by stations in or beyond Boston Proper, the egress mode tables in the published reports from the surveys revealed little about the overall egress mode distributions of passengers boarding in each study segment. For example, in the published reports for the Blue Line, egress modes and times shown for stations from Wonderland through Maverick are the modes and times for passengers alighting at these stations, regardless of where they boarded. For purposes of the comparison study, egress modes of passengers boarding at stations from Wonderland through Maverick, regardless of where they alighted, were of greater interest. The comparison tables for this memorandum show the eventual egress modes and times from rail rapid transit for the passengers boarding at stations in each study segment.
In both the new and old surveys, as shown in the destination comparisons below, the majority of rapid transit and light rail riders had destinations in downtown Boston or Cambridge. For most of these passengers, the transit trip ended at a rail transit station. (The tabulations did not distinguish between rail transit trips that included transfers and trips that began and ended on the same route.) Most of these trips were completed by walking. In the 1994 survey between 85% and 95% of the riders boarding on all but one of the line segments examined walked to their destinations from the station or stop where they left the rail transit system. The walk-out percentage for passengers boarding on the Mattapan High Speed Line was slightly lower, at 82.8%, mostly because of the number of riders transferring to buses at Ashmont.
In contrast with the increases in the walk-in access shares on most line segments, all of the segments examined showed decreases in walk-out egress shares between the 1994 and 2008-09 surveys. Most of the decreases were between 3% and 5.5%, but among passengers boarding at stations between Braintree and North Quincy, walking egress from the final rail transit station decreased from 95.2% to 85.6% (-9.6%). Among riders boarding the Mattapan High Speed Line, walking egress trips from the final rapid transit alighting station decreased from 82.8% to 76.5% (-6.3%). For passengers boarding on most segments, decreases in walking egress from the final rail transit station were offset by increased use of MBTA or private buses or private shuttle services. Increased use of MBTA buses for egress was probably partly a result of the fare structure implemented in 2007 that made the fare for a combined rail rapid transit and bus trip the same as the fare for a trip by rapid transit only. Some of the increase in use of private shuttles resulted from increased travel to the Longwood Medical Area, which was served by shuttles during both survey efforts, and some was due to use of shuttles established after 1994. Several private shuttles that ran from South Station to the South Boston industrial area in 1994 have been replaced by Silver Line Waterfront service, which was treated as MBTA bus service in the 2008-09 survey tabulations.
Because of the predominance of downtown Boston stations as the final rapid transit exit point for trips starting on all of the line segments examined, there was much less variation in average egress times than in average access times for these lines. In 1994 average private-mode egress times on 8 of the 11 segments examined were between 5.9 and 6.2 minutes. The exceptions were the surface B Line, the northern end of the Red Line, and the surface D Line, at 5.4, 6.7, and 7.0 minutes, respectively. On each of those segments above-average percentages of boarding passengers alighted at stations before downtown Boston. Reported average egress times for passengers boarding on all 11 segments increased by 2008-09, resulting in a range from 5.7 to 8.8 minutes.
Egress times for all private egress modes combined were influenced most by walking egress times, since walking was by far the most frequently used of the private egress modes. All line segments examined showed decreases in the percentages of riders with private egress times of 5 minutes or less, and increases in percentages with times between 6 and 20 minutes. Switching from walking to MBTA buses may have occurred disproportionately in trips with shorter egress distances, as the previous extra fares for bus transfers would have given the greatest disincentive to transfer for passengers who could easily have walked. Increased average egress times may also have reflected slower walking by riders who were older, on average (as indicated in the socioeconomic characteristics).
In the published reports from the 1994 and 2008-09 surveys, trip destinations were tabulated by rapid transit or light rail alighting station or stop, while origins were tabulated by boarding station or stop. For purposes of comparison between the surveys, new reports were produced summarizing the destinations of passengers who boarded on each line segment regardless of their alighting stops. The analysis method used was first to select the destinations with individual shares of 1% or more of the boarding passengers on each line in the 2008-09 survey and then to compare these shares with those for the same destinations from the same lines in 1994.
Connections between the rail transit lines in downtown Boston allow for wider dispersal of trip destinations compared with trip origins of passengers on each of the segments examined. Nevertheless, destinations accounting for individual shares of 1% or more of total destinations accounted for 82% to 90% of the destinations reported for each of the 11 lines in 2008-09, including shares of 86% or more on all but three lines. In general, the towns or neighborhoods that accounted for 1% or more of destinations of riders boarding each line in 2008-09 also accounted for the vast majority of the destinations from riders boarding the same line in 1994, although there were some changes in relative importance of destinations.
On 8 of the 11 lines, the Financial/Retail neighborhood of downtown Boston accounted for the largest individual share of destinations in both 1994 and 2008-09. On the D and E Lines, Financial/Retail had the largest shares in 1994, but had slipped to second place behind the Longwood Medical Area on the D Line and Back Bay on the E Line in 2008-09. On the B Line the Boston University area had the largest individual destination share in both surveys, far ahead of second-place Financial/Retail. Even on the segments for which it accounted for the largest shares of destinations in both surveys, Financial/Retail had a smaller share in 2008-09 than in 1994. Decreases ranged from 0.3% to 7.8%. Surveys on the other MBTA modes also showed decreases in shares of destinations in Financial/Retail. This probably reflected reductions in financial-sector employment and closing or downsizing of several retail businesses that attracted work and shopping trips in 1994.
No individual towns or neighborhoods showed gains as destinations for riders from all 11 line segments. The share of trips to Government Center increased by 7% from the Blue Line, by 3.5% from the Mattapan Line and by 2.6% from the Ashmont Branch of the Red Line but otherwise increased by at most 2.1% or decreased. The Kendall/MIT area of Cambridge showed gains of 2.3% to 3.8% from each of the Red Line segments and the Mattapan Line. The Fenway area of Boston had gains of 1.0% to 3.0% among riders starting on the surface Green Line branches.
In the 1994 survey, the lowest checkoff age range was 17 and under, followed by 18 to 24. In 2008-09 this changed to 18 and under, followed by 19 to 24. This change was made in order to conform with current needs of databases that were expected to use the survey results. Aside from this change, the checkoff choices were the same. In 1994 the age range from 25 to 34 had the largest individual share of responses on eight of the line segments, at 27.4% to 38.6%. On the Mattapan Line, at 25.4%, it was slightly below the 45-to-64 age range, at 25.9%. On the B and E Lines, the largest shares were in the 18-to-24 age range at 43.3% and 32.3% reflecting the large number of students and recent graduates that they serve.
In 2008-09, the age range from 45 to 64 had the largest individual share of riders on the Blue Line, both ends of the Orange Line, the southern branches of the Red Line, and the Mattapan High Speed Line, at 29.0% to 40.1%. On the B and E Lines, the 19-to-24 range had the largest shares, at 44.5% and 36.4%. On the northern end of the Red Line and on the C and D Lines, the age range from 25 to 34 still had the largest individual shares, with little difference from 1994. However, both the northern Red Line and the D Line showed increased shares of riders in the 45-to-64 range and decreases in some of the younger ranges.
For gender, the 1994 survey form provided checkoff choices including only male or female. The 2008-09 form used a write-in format that allowed for transgender, but that accounted for at most 0.2% of the responses on any of the lines. Every line had more female than male respondents in both surveys. In 1994 the female shares ranged from 54.2% to 62.2%. On every line except the Blue Line, the female share was unchanged or higher in 2008-09 than in 1994, making a range from 57.2% to 62.7%. (On the Blue Line the share of female respondents decreased from 61.0% to 57.5%.) There are no other sources of information with which these percentages can be compared for accuracy. The higher percentages of female respondents could mean either that there are in fact more female than male passengers or that female passengers are more inclined to complete survey forms.
The checkoff ranges for household income differed between the 1994 and 2008-09 surveys, including a change in the top range from “over $80,000” to “$100,000 or more.” These changes were made in order to conform with current needs of databases that were expected to use the survey results. Income comparisons are of limited usefulness, since it would be expected that inflation would have raised average incomes significantly in the 14 to 15 years between these surveys. It was not possible to determine the extent to which inflation shifted respondents from one income range to another. On all but two lines, every income range up to $59,999 had a smaller share in 2008-09 than in 1994, indicating that more households were moving out of each range into a higher one than were moving in from a lower range. The exceptions were the B Line and the Mattapan Line, where the shares of incomes between $40,000 and $59,999 increased by 1.8% and 4.3%. In 1994, 7.4% to 24.4% of respondents on each line reported incomes of over $80,000. In 2008-09, household incomes of over $100,000 were reported by 22.1% to 39.4% of respondents on each line except the B Line, where the large student and recent graduate population limited the share in the top income range to 13.1%.
Changes in reported usage rates between 1994 and 2008-09 varied among the line segments, but all except the Blue Line showed decreases in the percentages of riders using the system more than five days a week. In most cases, reductions in six-day and seven-day riders were offset by increases in four-day and five-day ridership. However the surface E Line and the Mattapan High Speed Line had large increases among respondents riding less than one day a week. The 1994 survey form did not include a separate category for “Only visiting,” which was provided on the 2008-09 form. For purposes of comparison, visitors from the 2008-09 surveys were counted as less-than-one-day riders.
The 2008-09 survey forms provided a greater number of checkoff options for fare payment method than the 1994 forms did, and the fare categories on the two surveys were not all directly comparable because of changes in fare structure and fare-collection methods. Almost all of the line segments examined showed decreases in the percentages of respondents paying adult full fares. These were mostly offset by increases in use of monthly passes, or in use of seven-day passes, which were introduced in November 2000. On the surface Green Line branches, the elimination of free outbound fares was reflected in large shifts from “Other,” which would have included the free fares, to monthly and weekly passes. Full cash fares on the Green Line branches showed net decreases, but not as large as those on the heavy rail lines, implying some shifting from free to full fares. There were only small differences between the 1994 and 2008-09 surveys in the percentages of riders paying reduced fares for senior citizens, passengers with disabilities, students, or children. Comparisons with revenue reports generally show that passengers under age 18 are under-represented in surveys, but this sampling bias is not great enough to impact overall survey findings.
On each of the line segments examined, the great majority of survey respondents were licensed drivers. In 1994 this ranged from 73.6% on the Mattapan High Speed Line to 91.6% on the Braintree Branch of the Red Line. By 2008-09 the percentages of licensed drivers had increased slightly on some lines and decreased slightly on others, resulting in a range from 70.8% on the Mattapan Line to 90.5% on the C and D Lines.
The percentages of drivers with vehicles available to make the same trip on the survey day varied more widely. In 1994, only 28.9% of respondents on the B Line had vehicles available. The Braintree Branch of the Red Line had the highest vehicle availability rate, at 73.8%. On several lines, vehicle availability had decreased by 2008-09, with the largest drop being from 64.7% to 53.4% (-11.3%) on the northern end of the Orange Line. The Mattapan High Speed Line had the largest increase in vehicle availability, from 37.3% to 47.5% (10.2%). This was related more to changes in the locations of passenger origins than to changes in vehicle ownership in the same locations.
Some changes may simply have been a result in differences in wording of the questions on the survey forms. In 1994 the question was “Did you have an automobile available for this trip?” In 2008-2009, the question following the one on vehicle ownership was “Could you have used one of these vehicles instead of riding the rail line on the day you got this survey?” The 1994 question could have been interpreted by some respondents as referring to vehicles used for access to rapid transit that would not have been available to substitute for the rail segment of the trip.
Almost all line segments examined had lower reported vehicles per household and vehicles per capita in 2008-09 than in 1994. The Blue Line between Wonderland and Maverick had the largest changes, with decreases of 6.0% in two-vehicle households and 3.2% in households with three or more vehicles, while one-vehicle households increased 7.5% and no-vehicle households by 1.6%. Changes in household size contributed little to these changes. The largest exception to the general trend was on the Mattapan High Speed Line, which had a 9.3% increase in two-vehicle households and a 5.8% decrease in no-vehicle households. However, large changes in the distribution of reported passenger origins on this line were the main contributor to changes reported in vehicle ownership rates.
The 1994 and 2008-09 survey forms both included lists of several service quality measures which respondents were asked to rate on a scale of 1 to 5. In both surveys, 3 was labeled “Average.” In 1994, 1 was labeled “Very poor” and 5 was labeled “Very good,” but in 2008-09 the respective labels were “Poor” and “Excellent.” No labels were given to 2 or 4.
There were some differences between the measures listed on the two surveys. In 2008-09 the single measure “Safety and security” replaced “Personal safety at station/on train” and “Vehicle security at lot/garage” from the 1994 form. The 2008-09 form included “Station amenities,” for which the most comparable measures from 1994 were “Station condition” and “Station cleanliness.”
Overall, some measures were rated higher in 2008-09 than in 1994 and some were rated lower, but most showed increases or decreases of 0.5 points or less on any individual line segment. “Reliability (on-time performance)” showed a slight net decrease in overall rating. The largest decreases were from 3.6 to 3.3 on the south end of the Orange Line and from 3.2 to 2.9 on the surface D Line. The largest increase was from 2.7 to 3.1 on the Braintree Branch of the Red Line.
In 2008-09, “Safety and security” was rated higher on all line segments than the average of the two measures from the 1994 survey used for comparison, with an overall gain of 0.36 points. The largest improvement was from 2.8 to 3.4 on the Red Line between Ashmont and Broadway.
The rating for “Cleanliness/condition of vehicles” was either unchanged or slightly lower on all lines in 2008-09 except on the Blue Line, where it increased from 3.1 to 3.2. The Blue Line rating may have reflected the lower average fleet age on that line. The overall drop in the rating for all line segments was only 0.1 point.
“Announcement of stations” showed the most rating improvement of all the service measures, with an overall gain of 0.43 points and a higher rating on every segment in 2008-09 than in 1994. This reflects the transition from live announcements by train crews to recorded announcements and electronic displays.
“Availability of seating on trains” was rated slightly lower overall in 2008-09 than in 1994. The largest decreases were from 3.5 to 3.2 on the south end of the Orange Line and from 2.8 to 2.5 on the surface C Line. The only increases in ratings for this measure were from 2.8 to 3.0 on the Red Line between Ashmont and Broadway, and from 2.9 to 3.0 on the Blue Line. The overall decrease for all line segments was 0.07 points.
In 2008-09 on every line segment “Frequency of service” was rated either the same as or lower than it was in 1994. The largest decreases were from 3.3 to 2.7 on the surface D Line and from 3.2 to 2.7 on the surface C Line. Scheduled headways were in fact lengthened on most lines between the two surveys. However, the ratings for frequency declined even on lines that did not have frequency reductions. Operation of longer train lengths offset the capacity impacts of longer headways, but apparently passengers were more concerned with wait times. The overall drop in the rating for this measure for all line segments was 0.25 points.
In 2008-09, “Travel time/speed” was rated either the same or lower compared with 1994, except on the Braintree Branch of the Red Line, where it increased from 3.0 to 3.1. The largest decrease was from 3.4 to 3.0 on the surface D Line. The latter may reflect longer station dwell times resulting from imposition of outbound fares in 2007. The overall decrease in rating for this measure for all line segments was 0.16 points.
Between 1994 and 2008-09, the rating for “Parking availability” increased on every line segment except the Red Line Braintree Branch, where it decreased from 3.5 to 3.3. That segment had the highest reported use of park-and-ride access (37.3%) in 2008-09. The largest rating increases were from 2.1 to 2.8 on the surface E Line, and from 2.1 to 2.7 on the surface B Line. Neither of those lines has dedicated parking, and less than 2% of the respondents on them reported using park-and-ride access. The overall increase in the rating for parking availability for all line segments was 0.28 points.
On most of the segments examined, “Station amenities” received the lowest rating of all the service measures. It was also not as clearly defined as most of the other measures. The combined ratings for “Station condition” and “Station cleanliness” were higher on all segments in 1994 than the “Station amenities” ratings for the same segments in 2008-09. The overall decrease for all line segments was 0.62 points, which was by far the largest decrease for any measure, but again it should be noted that the measures in the two surveys were not entirely comparable.
For purposes of comparing the 2008-09 and 1995 bus surveys, results were examined at the bus garage level only. These garages are: Albany, Arborway, Cabot, Charlestown (combined with Fellsway in the survey reports), Lynn, Quincy, and Somerville. Adjustments were made for routes for which garage assignments had changed between the surveys, including the replacement of the former Bartlett garage with the Arborway garage. Results for the Cambridge trackless trolley routes are included in the Somerville garage reports. The Silver Line Washington Street and Waterfront routes were not included in the 2008-09 survey effort because they had been surveyed in separate projects relatively recently before that.
In the 1995 survey reports for each garage, results were presented separately for inbound and outbound trips in the origin, destination, access, and egress tables, but only for both directions combined in the other tables. At the Quincy and Lynn garages, definitions of inbound and outbound changed for several routes between 1995 and 2008-09. For the comparisons that called for breakdowns by direction, new tables were generated from the 1995 survey records, with inbound and outbound definitions revised to conform to those in 2008-09.
In addition to the comparisons for individual garages, results from five of the nine reports were compared for all garages combined. The reports for origins, destinations, access modes, and egress modes were compared only at the individual garage level, as comparisons at higher levels of aggregation did not reveal significant changes.
In both sets of surveys, home-based work trips (either home to work or work to home) accounted for by far the largest shares of trips on routes based at each of the seven bus garages. For all garages combined, the home-based work-trip share of all trips increased slightly, from 55.7% to 57.9%. The share changed by less than 3.5% at Cabot, Charlestown/Fellsway, Lynn, and Quincy, but increased by 11.5% at Albany and 7.6% at Somerville and decreased by 4.4% at Arborway.
In both surveys, home-based school trips accounted for the second-largest share of trips at all garages combined, with essentially no change overall from 1995 to 2008-09. At most of the individual garages such trips also ranked second but varied more between the surveys. Arborway had the largest increase in school trip share, from 10.6% to 14.0%. Albany had the largest decrease, from 12.3% to 7.4%, but school trips remained second in importance there. At Quincy a decrease from 10.6% to 6.3% lowered school trips from second to fifth in importance there, behind home-based work, personal business, and other trips.
The old and new survey forms allowed passengers to check as many reasons as applied for using MBTA service. The 1995 forms asked about use of the MBTA in general, but the 2008-09 forms asked specifically about use of buses. The 2008-09 forms also included more checkoff choices. “Convenience,” which was the most subjective of the checkoff choices in both surveys, was the reason most often selected at all garages combined in both surveys, being checked by 46.2% of respondents in 1995 and 55.5% in 2008-09. It was also the most often checked reason at all individual garages in 2008-09 and at most garages in 1995, but was behind “Only transportation available” at Quincy and Arborway that year. “Only transportation available” ranked second overall in both years, but decreased from 43.5% to 38.9% in the portion of riders checking it. As was the case on other modes surveyed, “Environmentally responsible” was checked by significantly more respondents in the 2008-09 survey than in the previous one. For all bus garages combined, it increased from 15.7% to 31.3%, with gains at individual garages ranging from 9.2% at Arborway to 25.8% at Albany.
The questions on alternate means of travel on the two surveys were not strictly comparable. In 1995 the question was hypothetical as to the means of transportation the respondent would use for the same trip if the bus route were not available. In the 2008-09 survey, the question was what means, if any, the respondent sometimes used instead of the survey route to make the same trip. Excluding “Other,” which allowed for a variety of written-in reasons, in both surveys, the most common alternative checked was “Other MBTA service,” increasing from 24.7% overall in 1995 to 46.0% in 2008-09. “Driving alone” was second, increasing from 17.6% to 27.7%. The same rank order applied at most individual garages. At Quincy, “Driving alone” was far ahead of “Other MBTA service” in 1995, but they were almost tied in 2008-09. At Somerville “Driving alone” was first in 1995 but was behind “Other MBTA service” in 2008-09. These findings suggest that for many riders an alternate means of transportation that is adequate for occasional use would not be the first choice as a permanent substitute for the bus route for which the survey form was filled out.
In the bus survey comparisons, trip origin locations were examined separately for inbound and outbound trips at each garage. MBTA bus routes are assigned to garages largely on the basis of the geographic areas that the routes serve, although operating constraints sometimes dictate other assignments. In designations of route direction, inbound most often refers to the direction of travel either toward a rapid transit station or toward downtown Boston, but the alignments of some routes do not lend themselves to such terminology. In the comparison tables, results for the 1995 surveys from the Quincy and Lynn garages were adjusted for reversal of inbound and outbound designations on some routes by 2008-09.
For the combined routes at each garage, there were few if any changes of more than 3.5% in the shares of inbound trips originating in any individual municipality or neighborhood. Among Arborway routes, the importance of Jamaica Plain as an inbound trip origin decreased from 20.4% to 14.2% (-6.2%), but increases in origins for routes at that garage were much less concentrated. At Cabot Garage, inbound origin shares increased by 6.5% for North Dorchester and by 4.7% for South Dorchester but decreased by 6.4% for Roxbury. Some of this may have resulted only from differences in reporting the locations of origins near the border of Roxbury and Dorchester. Some decline in Roxbury bus origins may have resulted from diversions to Silver Line Washington Street service.
At Lynn Garage, the share of inbound trip origins from Lynn decreased by 5.2% and that from Chelsea by 3.7%, while the share from East Boston increased by 4.6%. Some of these changes probably reflect increasing use of commuter rail for trips from communities served both by that mode and by Lynn Garage buses. Quincy Garage showed a 7.6% increase in inbound origins from Quincy, with the largest decrease being 3.7% in the share of origins from Randolph. The decreases in trip origins from locations outside of Quincy were related to the opening of the Middleborough/Lakeville and Kingston/ Plymouth commuter rail lines in 1997 and the Greenbush Line in 2007, as each of those lines serves locations also served by Quincy Garage feeder service to the Red Line. Randolph is one such location.
Inbound trips on most MBTA bus routes serve a combination of local riders making trips entirely on the route and trips completed by transferring to the rapid transit system to reach more distant destinations. Conversely, outbound trips serve a combination of local riders and riders whose trips begin at more distant points served by the rapid transit system. Therefore, the origins of outbound trips on the routes from each garage are more dispersed than the inbound origins.
Between the 1995 and 2008-09 surveys, there were relatively few changes of more than 3.5% in the portion of outbound trips originating in any individual municipality or neighborhood on the routes from any garage. On Albany Garage routes, the share of trips originating in the Financial/Retail neighborhood decreased 8.6%, while there were increases of 6.2% in trips from the Longwood Medical Area, 5.0% in trips from North Brookline, and 4.7% in trips from Brighton. At the Arborway garage, outbound trips originating in the Prudential/Hancock neighborhood decreased 4.6%, while there were increases of 4.4% in trips from Roxbury and 4.0% in trips from the South End. At the Cabot Garage trips from Roxbury increased by 3.8%.
At the Charlestown/Fellsway garages, outbound trips from Malden increased by 6.1%, but there were no large increases or decreases in trips from other individual communities. At the Somerville Garage outbound trips from the Spring Hill neighborhood of Somerville increased by 7.2%. In Cambridge, outbound trips from the Central Square neighborhood increased 4.4%, while those from the Harvard Square neighborhood decreased 3.8%. Some of the latter changes may have resulted from differences in reporting of origins that are located near the border of these neighborhoods.
At Lynn Garage, the share of outbound origins from Lynn fell by 8.5%, while the shares from East Boston and Revere increased by 7.1% and 4.5%, respectively. The share of outbound trips on Quincy Garage routes originating in Quincy increased by 7.6%, the same as the increase in the inbound origin share, but there were no offsetting decreases of over 2%.
In both surveys, on inbound bus trips, the bus was most often the first transit mode used in the overall trip, and walking was the most common means of access to the bus. The 1995 survey results from the Lynn and Quincy garages were adjusted for changes in inbound and outbound definitions for some routes to allow for comparisons with the 2008-09 surveys. All private access modes combined accounted for 81% to 95% of access trips to inbound bus routes at each garage in both surveys, with the only changes of over 3% between surveys being an increase from 82.1% to 85.5% among Albany Garage routes and a decrease from 95.0% to 91.0% among Lynn Garage routes. Walk-ins alone accounted for 72% to 90% of access trips to inbound buses, with little change between the surveys. Only about 4%, at most, of inbound bus riders at each garage reported using park-and-ride access in either survey, except at the Albany Garage, where about 7% did so. The latter figure reflects the inclusion of several express bus routes between western suburbs and downtown Boston in the Albany totals.
Among riders who reported access to an inbound bus by transit, the most common mode reported was another bus, but this was at most about 10% of total access trips. The 1995 tabulations showed all bus access trips as being by MBTA buses, but small percentages may have been from other buses. In 2008-09, as in 1995, surveys were distributed on all bus routes, but coverage of the entire bus system spanned several months. A passenger who received survey forms on more than one route most likely filled out only one of them. On a two-bus trip, if the survey was filled out for the first bus, the egress mode would be reported as bus and the access mode as something else. If the survey was filled out for the second bus, the access mode would be bus and the egress mode would be something else.
The majority of MBTA bus trips in the direction designated as outbound start at rapid transit stations. As would be expected, transfers from rapid transit accounted for significant shares of access trips to outbound buses. Nevertheless, the majority of outbound bus riders had access modes other than rapid transit. In 1995, rapid transit transfers accounted for 15% to 32% of the access trips to outbound buses from each garage. The shares changed little between the two surveys, except that rapid transit access trips increased from 21.0% to 26.2% among Albany Garage routes, from 16.6% to 20.9% among Quincy Garage routes, and from 15.2% to 19.0% among Lynn Garage routes.
Walking accounted for over 50% of the access trips to outbound buses from each garage in both surveys. At Albany Garage the walking access share declined from 62.7% to 53.9%, with increased rapid transit and bus access accounting for the change. At Cabot Garage, the walking access share fell from 67.4% to 61.6%, with most of the offset being in increased bus access trips. At Quincy Garage the walk-in share of outbound access trips decreased from 69.9% to 65.7%, with increased rapid transit access trips offsetting this loss.
Increases in bus and rapid transit access would be at least partly attributable to the implementation of free bus-to-bus transfers in December 2000 and free rapid transit-to-bus transfers in January 2007.
Standard reports from the 2008-09 surveys included tabulations of access times by range for walk, park-and-ride, drop-off, other private access modes, and for all of these combined. These included only trips for which the private access was directly to the bus on which the survey form was received. For purposes of comparison with the 1995 surveys, the combined times for all private access modes were used. Reported average access times to inbound and outbound buses at all garages increased or decreased slightly between 1995 and 2008-2009. Inbound Quincy Garage riders had the largest increase, with the average going from 5.5 minutes to 8.3 minutes. For outbound Quincy Garage riders, the average access time by private modes increased from 5.8 minutes to 7.3 minutes. These changes were partly a reflection of decreases in walking access trips and increases in drop-offs, as the latter had much higher reported average access times. Among outbound Arborway Garage riders, the average access time by private modes increased from 5.9 minutes to 8.0 minutes. Most of the riders on Arborway buses reported trip origins in Boston neighborhoods. Changes among these neighborhoods in the shares of trip origins were partly responsible for access time changes.
The majority of MBTA passengers on a given day make round trips, so over the course of an entire day, the patterns of inbound access modes would be expected to be close to those of outbound egress modes. Because the surveys were only distributed from about 6:00 AM to 3:30 PM, the database would include only the first halves of many of the round trips that were made in a day. Nevertheless, the majority of egress trips from outbound buses at all garages, like the majority of access trips to inbound buses, were made by private modes, and the percentages changed little between the surveys.
At most garages, imbalances between the percentages of outbound access trips and inbound egress trips made by private transportation during the survey hours changed by no more than about 5% between 1995 and 2008-09. The largest change was at Albany Garage, where the share of outbound access trips by private transportation was 7.4% lower than the private-mode share of inbound egress trips in 1995 but was 20.7% lower than the private-mode share of inbound egress trips in 2008-09. This reflected both a shift to using rapid transit instead of walking for outbound access and an opposite shift from rapid transit to walking for inbound egress.
As would be expected, given the route network, rapid transit had a much larger share of egress trips from inbound bus trips than from outbound trips in both surveys. Transfers from outbound trips to rapid transit were heaviest among Quincy Garage routes, with shares of 29.9% of all egress trips in 1995 and 33.3% in 2008-09. Nevertheless, rapid transit had even higher shares of Quincy Garage inbound egress trips, at 40.4% and 45.9%. The orientation of several of the Quincy routes makes it more convenient for passengers to access inbound rapid transit trains by starting out on outbound buses rather than on inbound ones. This is also true to an increasing extent for Charlestown/Fellsway routes, many of which have more than one rapid transit connection. The share of outbound egress trips by rapid transit from Charlestown/ Fellsway routes increased from 14.6% to 22.9% between the surveys. This may indicate that increasing numbers of riders with choices of two rapid transit connections were using the outer connection to have better chances of getting seats on the rapid transit trains.
In both sets of surveys, egress times were tabulated only for passengers reporting use of private egress modes directly from the buses on which the survey forms were received. At every garage, average egress times by private modes both from inbound and from outbound routes increased between 1995 and 2008-09. The largest reported increases were among Arborway Garage routes, where average times for private egress increased from 5.9 to 9.6 minutes inbound and from 5.5 to 8.6 minutes outbound. These changes were at least partly related to changes in the relative importance of various Boston neighborhoods as destinations for Arborway Garage riders. In areas where development is less concentrated directly on bus routes, average walking distance can be expected to increase.
In the survey comparisons, destination location results from inbound and outbound bus trips were separated for each garage. For the Quincy and Lynn garages, where inbound and outbound designations for several routes were reversed between 1995 and 2008-09, the 2008-09 designations were used in the comparisons.
For the combined routes at each garage, there were few changes of more than 3.5% in the shares of inbound trips destined for any individual municipality or neighborhood. Among Albany Garage routes, destination shares increased by 4.7% in the Fenway neighborhood and 4.1% in the Prudential/Hancock neighborhood but dropped by 4.7% in the South End. Inbound Cabot Garage trips had an increase of 4.9% in destinations in Roxbury and a decrease of 5.2% in those in the Financial/Retail area. Quincy Garage routes had an increase of 9.7% in inbound destinations in Quincy, but decreases of 4.6% and 4.1% in destinations in Weymouth and the Financial/Retail area. Somerville Garage routes had a decrease of 4.8% in the share of inbound destinations in the Harvard Square neighborhood of Cambridge, but a 3.6% increase in the share of destinations in the Kendall/MIT neighborhood.
At the time of day when the surveys were distributed, ridership was generally lower on outbound bus trips than on inbound trips. With smaller outbound sample sizes, greater variation in the results would be expected. Nevertheless, the outbound results also showed relatively few changes of more than 3.5% in the shares of destinations in individual municipalities or neighborhoods.
Among Albany Garage routes, the share of outbound destinations in Brighton decreased by 5.3%, while destinations in the South End increased by 3.6%. Among Arborway Garage routes, outbound destinations in Jamaica Plain decreased by 12.0% and destinations in Mattapan increased 6.9%. Cabot Garage routes had outbound destination increases of 7.6% to South Dorchester and 4.5% to North Dorchester, but decreases of 5.7% to Roxbury and 3.6% to Harvard Square.
Lynn Garage routes had an increase of 9.8% in outbound destinations in East Boston but decreases of 6.7% to Revere, 6.1% to Lynn, and 4.2% to Marblehead. Somerville Garage routes had a 3.6% decrease in outbound destinations in Brighton.
In the 1995 survey, the lowest checkoff age range was 17 and under, followed by 18 to 24. In 2008-09 this changed to 18 and under followed by 19 to 24. Otherwise, the checkoff choices were the same. Among routes from five of the seven garages, the largest individual share of responses in 1995 was in the age range from 25 to 34, at 23.0% to 32.9%. At the Quincy Garage, the range from 45 to 64 had the largest share, at 27.4%. At the Arborway Garage, the largest share was in ages 35 to 44, at 25.4%.
In 2008-09 for routes at five garages, the age range with the largest individual share was from 45 to 64, at 28.0% to 37.8%. At the Albany Garage, the range from 25 to 34 still had the largest share, at 30.6%, but the range from 45 to 64 was approaching this, at 28.5%. Similarly, among Somerville Garage routes, the age range from 25 to 34 had the largest share, at 30.0%, but the range from 45 to 64 was only slightly lower, at 28.0%.
For gender, the 1995 survey form provided checkoff choices including only male or female. The 2008-09 form used a write-in format that allowed for transgender, but that accounted for at most 0.2% of the combined responses from routes at any of the garages. In both 1995 and 2008-09 there were many more female than male respondents, but the shares of male respondents on routes at most of the garages increased somewhat between the two surveys. In 1995, the shares of female respondents ranged from 63.7% on Albany Garage routes to 70.6% on Cabot Garage routes. In 2008-09 the shares of female respondents ranged from 58.1% on Quincy Garage routes to 69.1% on Cabot Garage routes. Quincy Garage routes had the largest increase in male respondents, from 33.1% to 41.9%.
The checkoff ranges for household income differed between the 1995 and 2008-09 surveys, including a change in the top range from “$80,000 or more” to “$100,000 or more.” Income comparisons are of limited usefulness, since it would be expected that inflation would have raised average incomes significantly in the 13 to 14 years between these surveys. At all seven garages, the three lowest income ranges (up to $39,999) were each reported by smaller shares of riders in 2008-09 than in 1995. Four of the garages also had decreases in shares of riders with household incomes between $40,000 and $59,999. At the three other garages Arborway, Cabot, and Lynn the increases in that range were 1.0%, 2.1%, and 2.6%, respectively.
In 1995, at five of the garages the shares of riders with household incomes of $80,000 or more ranged for 4.4% at Lynn to 7.3% at Quincy. This income range was reported by 12.8% of riders at the Somerville Garage and by 13.1% at Albany. The Somerville results were pulled toward the top range by passengers using the Cambridge trackless trolleys and other routes in Cambridge and Arlington. The Albany results were pulled toward the top range by passengers on several suburban express bus routes. In 2008-09, household incomes of over $100,000 were reported by 28.7% of Albany Garage route riders and 25.9% of Somerville Garage route riders. At the other five garages, percentages at this income level ranged from 7.4% for the Lynn Garage to 17.9% for Charlestown/Fellsway.
Results from each of the garages showed decreases in the percentages of riders using MBTA buses more than five days a week, with a decrease of 15.7% for all seven garages combined. The distributions of reported frequency increases varied among garages. Five-day use had an overall gain of 7.0%, or slightly less than half of the decrease in six-day and seven-day use. The other gains were scattered among all the choices from four days to less than one day. The 1995 survey form did not include a separate category for “Only visiting,” which was provided on the 2008-09 form. For purposes of comparison, visitors from the 2008-09 surveys were counted as less-than-one-day riders. Overall, this combined category showed the second largest increase in share of riders, at 3.1%.
The 2008-09 survey forms provided a greater number of checkoff options for fare payment method than the 1995 forms did, and because of changes in fare structure and fare-collection methods, the fare categories on the two surveys were not all directly comparable. For all seven garages combined, the percent of riders paying adult full fares increased slightly, from 28.3% in 1995 to 29.4% in 2008-09. Some garages had slight increases and some had slight decreases in this category. The largest increases were 5.2% for the Quincy Garage and 4.7% for the Arborway Garage. The largest decrease was 2.2% for the Cabot Garage.
Overall, use of adult monthly passes declined 4.1%, from 54.9% to 50.8%. This was more than offset by the 5.5% of riders in 2008-09 using seven-day Link passes, which were not offered in 1995. Arborway Garage passengers had the largest reported drop in adult monthly pass use (-9.8%), with the 4.7% increase in full fares and 5.4% use of seven-day passes making up for it. Overall, the share of bus riders paying fares for senior citizens or passengers with disabilities decreased by 2.9%. Single-ride fares in these categories decreased by 8.4%. Monthly passes for senior citizens or passengers with disabilities were first introduced in November 2000 and accounted for 5.9% of all bus passenger fares in the 2008-09 survey.
In both surveys, the majority of respondents on routes from each of the bus garages were licensed drivers. In 1995, this ranged from 57.6% of Lynn Garage riders to 79.4% of Albany Garage riders. By 2008-09, the percentages of riders with licenses had increased at some garages and decreased at others. This made a range from 55.5% at the Lynn Garage to 83.5% at Albany. The overall percent of riders with licenses was essentially unchanged, at 68.4% in 1995 and 68.3% in 2008-09.
In contrast with the rate of licensed drivers, the majority of bus riders in both surveys (68.7% in 1995, 68.5% in 2008-09) did not have private vehicles available to make the same trips on the days when they were surveyed. In 1995, the percent of riders without vehicles available ranged from 56.7% at the Albany Garage to 76.1% at Lynn. Between 1995 and 2008-09 the percentages of riders without vehicles available changed only slightly at each garage.
Among all bus survey respondents, there was little change in reported vehicles per household or vehicles per capita between 1995 and 2008-09. Results from the Albany Garage showed the most change, with one-vehicle households increasing from 37.9% to 45.4%, while the percentages in all other categories decreased. Consequently, there were changes in the percentages in most of the per capita ranges at this garage, with the greatest increase being from 26.9% to 31.3% with 0.5 to 0.99 vehicles per capita, and the greatest decrease being from 23.7% to 18.1% among those with 1.0 to 1.49 vehicles per capita.
The 1995 and 2008-09 survey forms both included lists of several service quality measures which respondents were asked to rate on a scale from 1 to 5. In both surveys, 3 was labeled “Average.” In 1995, 1 was labeled “Very poor” and 5 was labeled “Very good,” but in 2008-09 the respective labels were “Poor” and “Excellent.” No labels were given to 2 or 4.
There were some differences between the measures listed on the two surveys. In 2008-09 the measure “Safety and security” replaced “Personal safety” from the 1995 survey. The measure “Travel time/directness of route” in 1995 was replaced by “Travel time/speed” in 2008-09. Three of the measures on the 2008-09 survey forms (“Stop amenities,” “Fare collection system,” and “Signage on vehicle”) did not correspond with any measures on the 1995 forms.
Most measures were rated lower by riders on routes from all garages in 2008-09 compared with 1995. The only measure with an improved rating was “Announcement of stops,” which increased from 2.5 to 3.7 for all garages combined. As on other modes surveyed, this was a reflection of the transition from live announcements by vehicle operators to recorded announcements and electronic displays.
The biggest rating drops overall were for “Reliability,” which dropped from 3.5 to 2.9, and “Travel time/speed,” which dropped from 3.8 to 3.3. The only measure for which the rating did not change was “Parking availability,” at 3.1 in both surveys. Each of the other five measures had an overall rating drop of 0.3 points.
Among respondents from routes at individual garages, Charlestown/Fellsway bus riders lowered their rating for reliability most, from 3.7 to 2.8. They tied with Albany riders in downgrading travel time/speed from 3.9 to 3.3.
The 1993 surveys on the commuter rail lines that were then in operation and the 1998 surveys on the Middleborough/Lakeville and Kingston/Plymouth Lines were conducted on all weekday inbound trains from start to end of a service day. The 2008-09 surveys were distributed on all weekday trains in both directions from start of service to about 3:30 PM. For purposes of comparison, only the inbound survey results from 2008-09 were used. The programs used to generate reports from the 1993 and 1998 surveys were no longer available, but the survey records had been exported into Excel spreadsheets. CTPS used database management tools to extract information from the 1993 and 1998 survey spreadsheets in forms comparable to those in the 2008-09 survey reports. Responses from inbound riders after 3:30 PM accounted for fairly small percentages of the total responses from each line, so in most cases the all-day results in the published reports were found to differ only slightly from the results for responses from before 3:30 only.
In both sets of surveys, home-based work trips (either home to work or work to home) accounted for by far the largest share of trips on each of the 11 lines examined as well as on the Greenbush Line, which was surveyed for the first time in 2009. On seven of the lines, this share changed by no more than 3.5% from the earlier survey. However, it increased by 4.5% on the Lowell Line, by 6.1% on the Fairmount Line, by 7.0% on the Kingston/Plymouth Line, and by 8.9% on the Middleborough/Lakeville Line. The latter two had been in operation for only about one year when they were surveyed in 1998. On the Middleborough/Lakeville Line, the increase in percentage of home-based work trips resulted mostly from a faster rate of growth in such trips than in trips for other purposes, but on the Kingston/Plymouth Line the number of home-based work trips remained fairly steady, while trips for other purposes, especially home-based school trips, declined. The Lowell and Fairmount Lines showed both absolute growth in work trips and absolute decreases in trips for most other purposes.
The old and new survey forms allowed passengers to check as many reasons as applied for using commuter rail, but the 2008-09 forms included more choices. “Convenience,” which was the most subjective of the checkoff choices in both surveys, was the reason most often selected on each segment on the lines surveyed in 1993. On the Middleborough/Lakeville and Kingston/ Plymouth Lines in 1998, “Avoid driving/traffic” which was not a checkoff choice in 1993, was slightly ahead of convenience. In the 2008-09 survey, which included “Avoid driving/traffic” as a checkoff choice, it was the most-checked reason on every commuter rail line except the Fairmount Line, where “Convenience” was ahead. “Convenience” ranked second or third on other lines, with the new checkoff choice “Avoid parking at destination” ranking second on about half of them. As on the surveys on other modes, the number of commuter rail respondents checking “Environmentally responsible” as a reason for using the MBTA increased significantly between the old and new surveys, with gains by line ranging from 6.2% to 23.6%.
The questions on alternate means of travel on the old and new surveys were not strictly comparable. In 1993, the question was hypothetical as to the means of transportation the passenger would use for the same trip if commuter rail service were not available. The 1998 Old Colony survey and the 2008-09 survey asked what alternate means, if any, respondents sometimes used to make the same trip instead of commuter rail. In both surveys, driving alone was the most frequently checked alternate means except that on the Fairmount Line in both surveys and on the Needham Line in 1993, use of other MBTA service as an alternative was more common. Driving alone was selected by more respondents in 2008-09 than in 1993 on all lines including the Fairmount Line and among Old Colony Line riders after 1998. Riders on most lines reported use of other MBTA services less in 2008-09 than in 1993. These findings suggest that commuter rail riders who sometimes make the same trips by driving would nevertheless prefer to use some form of transit for their most frequent trips if commuter rail was not an option. In 1998, many Old Colony riders had only recently switched from other MBTA transit services, and some of them were still using those services on some days.
On each line, the great majority of riders had trip origins in towns or neighborhoods with stations on that line or bordering directly on other towns or neighborhoods with stations. Outer terminal stations or stations with large park-and-ride facilities attracted riders from greater distances; this is not reflected in this memorandum since comparisons were done at the line level. The analysis method used was first to select all origins accounting for individual shares of 1% or more of the boardings on each line in the 2008-09 survey and then to compare these shares with those from the same origins on the same lines in the prior surveys. Some changes in distribution of origins resulted from extensions of lines or opening of new stations between surveys. In 1993 the Newburyport Line ended at Ipswich and the Framingham/ Worcester Line extended only as far as Framingham. Limited peak-period service was extended from Framingham to Worcester with no intermediate stations in 1994. The older survey database for that line includes surveys of Worcester Station passengers conducted in 1995, but these are not included in the published reports of the results of the 1993 survey nor in the comparisons in this memorandum.
In the 2008-09 survey, origins accounting for individual shares of at least 1% of total inbound boardings during the survey hours accounted for combined totals of over 85% of the origins on each line except the Fitchburg Line (82.7%). On 7 of the 11 lines, over 90% of riders were from origins with individual shares of at least 1%.
On each line, the combined shares of origins from the towns or neighborhoods with individual shares of at least 1% were very similar to the combined shares from the same towns or neighborhoods in the prior surveys, although there were some changes in relative importance. Increases or decreases of more than 3% in origin shares were examined in greater depth than smaller changes. On the Newburyport/Rockport Line, the share from Newburyport alone increased from 0.3% to 3.7% as the result of the extension of the line to that city in 1998. On the Haverhill/Reading Line, shares of origins from communities between Reading and Boston decreased, while shares from points north of Reading increased. This was attributable to an increase in the number of trains running through to Haverhill instead of terminating at Reading. The largest share increase was from 7.6% to 13.4% (+5.8%) at Haverhill. The largest decreases were from 18.5% to 13.2% (-5.3%) at Wakefield, and from 14.4% to 9.8% (-4.6%) at Melrose. Haverhill had both absolute and relative ridership increases, while Wakefield and Melrose had both absolute and relative decreases.
On the Lowell Line, the share of origins from Woburn decreased from 8.3% to 5.3%; no other origin showed a change as large as 3%. Total origins from Woburn changed little while overall ridership on the line increased. On the Fitchburg Line, the only origin share change of over 3% was a decrease from 13.6% to 10.5% (-3.1%) in Concord origins. This was due to faster growth in ridership from some other towns, not to a loss in Concord origins.
As would be expected, the doubling of the length of the Framingham Line resulted in substantial changes in the origins of riders. In 1993, 74.4% of the riders on the Framingham Line had trip origins in the communities that then had stations: Framingham, Natick, Wellesley and Newton. Only 6.2% had origins in communities that subsequently gained stations: Ashland, Southborough, Westborough, Grafton, and Worcester. In 2008-09, the share from the communities that also had stations in 1993 dropped to 46.8%, including a small absolute decline, while the share from cities and towns with the added stations increased to 22.7%. Other gains came from communities not directly served but contributing boardings to the new stations. These results show clearly that extending service to Worcester did not simply redistribute passengers that were previously using the line.
On the Needham Line the share of origins from Needham increased from 40.2% to 45.6% (+5.4%). The share from West Roxbury decreased from 29.2% to 24.5% (-4.7%). These changes represent both absolute growth in Needham ridership and some absolute loss in West Roxbury ridership. West Roxbury has much more frequent MBTA bus service as an alternative to commuter rail than Needham has. Fare increases implemented in 2007 may have diverted some West Roxbury riders to these buses, which connect with rapid transit service at Forest Hills.
On the Franklin Line, the share of origins from Norwood decreased 4.4%, from 25.6% to 21.2%, while the share from Dedham increased from 7.6% to 11.8% (+4.2%). These changes included both an absolute increase from Dedham and an absolute decrease from Norwood, but the reasons for these patterns are unclear.
On the Providence Line, the share of origins from Providence alone increased from 4.9% to 9.9% (+5.0%) with an absolute increase reflecting increased service on that part of the line. The share from Stoughton declined from 9.9% to 6.4% (-3.4%) as a result of an absolute decrease in origins there while overall ridership on the line was growing.
The Fairmount Line was operating at a reduced service level during the 2008-09 survey, compared with 1993, to accommodate ongoing construction work on stations and bridges, so the comparison results are not necessarily indicative of long-term trends. Ridership went down significantly. Most origins had decreases in absolute numbers of Fairmount Line riders, but some had small increases. The Hyde Park neighborhood was the largest ridership source in both surveys, but its share of origins decreased from 38.8% to 34.6% (-4.2%). The share of origins from Milton dropped from 14.9% to 5.9% (-9.0%). North Dorchester showed a gain from 6.1% to 12.7% (+6.6%), but the absolute number there increased only slightly.
The Middleborough/Lakeville and Kingston/Plymouth Lines had been in full operation for less than one year when they were surveyed in 1998, so ridership patterns were not firmly established. In both the 1998 and 2008-09 surveys, Brockton was the largest individual ridership source on the Middleborough/Lakeville Line, and its share of origins increased from 29.2% to 38.1% (+8.9%). Bridgewater, the second-largest origin source showed a drop from 17.1% to 14.1% (-3.0%). The only other change of over 3% was in trips from Randolph, with a drop from 9.4% to 5.4% (-4.0%). That town has MBTA bus connections to the Red Line at Quincy Adams, Quincy Center, and Ashmont. The 2007 fare restructuring may have encouraged some Randolph riders to switch from commuter rail to the bus and Red Line combination.
On the Kingston/Plymouth Line, the only change of more than 3% in origin shares was a decrease from 10.2% to 6.2% (-4.0%) in the share from Weymouth. This is probably a result of the 2007 opening of the Greenbush Line, which has two stations in Weymouth. The 2008-09 survey showed over three times as many boardings from Weymouth on the Greenbush Line as the number lost from the Kingston/ Plymouth Line.
The summary reports on means of access to the commuter rail system for the 1993 and 1998 surveys were based on a question that asked how the respondent got to the commuter rail boarding station, but allowed only for answers as to the means by which the respondent finally arrived at that station. The 2008-09 surveys allowed for more details as to the complete access trip from origin to boarding station if it included more than one mode of travel, but the reports prepared from them show access information in a form comparable to those of the 1993 and 1998 survey reports.
In the 2008-09 surveys, responses from passengers using more than one commuter rail line in a trip were edited if necessary to show the boarding station on the first line used as the entry station. For example if a survey form reported that the passenger was traveling outbound from North Station on the Fitchburg Line and had traveled to North Station on a train from Salem, the form was entered in the database as an inbound trip from Salem to North Station, with egress via commuter rail. Similar editing was not needed for the older surveys, because they were distributed only on inbound trains.
The 2008-09 and 1998 survey forms included checkoff choices for taxi, bicycle, and shuttle van access. These checkoff choices were not included in the 1993 survey; passengers using any of these means of access had to write them in under “Other.” However, these accounted for relatively small shares of access trips to all of the commuter rail lines in any of the surveys. In all the sets of surveys, park-and-ride accounted for the largest share of access trips on all except the Needham and Fairmount Lines, where walk-in trips were first. The latter two lines are relatively short, have relatively close station spacing, and relatively limited parking facilities.
Most of the changes in shares of access trips from 1993 or 1998 to 2008-09 consisted of shifts between walking and park-and-ride, with the direction of the shift influenced at least in part by changes in parking capacity. The Fitchburg, Middleborough/Lakeville, and Kingston/Plymouth Lines had no changes of over 3% in access mode shares.
The Framingham/Worcester Line had the largest changes, as a result of its extension to Worcester, including five new stations with large parking capacities. On that line, the park-and-ride share increased from 48.2% to 59.0% (+10.8%) and the walk-in share decreased from 36.5% to 25.2% (-11.3%). However, comparing results only for stations that were served both in 1993 and in 2008-09, the walk-in share increased slightly, to 38.2%, and park-and-ride access was unchanged.
The Needham Line had the largest increase in walk-in share, from 46.9% to 54.7% (+7.8%), and a decrease from 41.6% to 34.8% (-6.8%) in park-and-ride access. The Franklin Line had the largest decrease in park-and-ride share, from 64.9% to 55.5% (-9.4%) offset largely by increases of 6.6% in walk-ins and 1.7% in drop-offs. Constrained parking and higher parking fees likely contributed to the reduced park-and-ride use on these lines.
The Lowell Line had the largest change in the share of access trips accounted for by drop-offs, decreasing from 18.6% to 11.9% (-6.7%). The 1993 drop-off share on that line was unusually high. New or expanded parking facilities at several points along the line resulted in some diversions from drop-offs to park-and-ride, but walk-ins and other access modes also showed increased shares.
Standard reports from the 2008-09 surveys included tabulations of access times by range, for walk, park-and-ride, drop-off, and other private access modes and for all of these combined. For purposes of comparison with the 1993 and 1998 surveys, the combined times for all private access modes were used. In the old and new surveys, the vast majority of riders (88% or more on every line) reported access times of 20 minutes or less. However, all lines showed decreases in the percentages of riders with access times of 5 minutes or less, with changes ranging from 4.5% to 10%. On lines with actual diversions from park-and-ride access to walk-ins, increased access times would reflect slower walking versus driving speeds. Lines with increased park-and-ride access generally had large new parking facilities that attracted trips from greater distances than before. Nevertheless, no line showed an increase of as much as 3% in the share of trips with access times of over 20 minutes.
Average access times by private transportation increased on 9 of the 11 lines. The increases ranged from 0.1 to 0.9 minutes, except on the Middleborough/Lakeville Line, where the increase was 1.4 minutes. Average access times decreased by 0.3 minutes on the Haverhill Line and 0.1 minutes on the Providence/Stoughton Line.
The old and new commuter rail survey forms included a question on waiting time at the boarding station, but the wording was slightly different. The 1993 forms asked how long the passenger usually waited, but the 1998 and 2008-09 forms asked how long the passenger had waited on the day of the survey. The 2008-09 reported waiting times were atypically high for the Fitchburg Line because the surveys were distributed on a day when a train breakdown had resulted in long delays in AM peak service..
On all lines except the Fitchburg Line, the great majority of riders in both the old and new surveys reported waiting times of 10 minutes or less. However, all lines showed some increase in the percentage of riders reporting wait times of over 10 minutes. The Providence/Stoughton Line had the largest increase in percent of riders waiting over 10 minutes, from 5.4% in 1993 to 17.1% in 2008-09. This may have been partly a result of increased off-peak service. Off-peak riders overall ride less frequently than peak riders, and they may not time their arrival times at boarding stations as closely. Also, some riders may have begun arriving at stations earlier than before to improve their chances of finding parking spaces.
The egress mode questions on the 1993 and 1998 survey forms asked only for the means of transportation used immediately after leaving the commuter rail train. The 2008-09 forms asked for the complete egress trip from commuter rail train to final destination, and the results were summarized in a form comparable to those of the earlier surveys.
The survey comparisons were based on responses from passengers on inbound trains from the start of service to mid-afternoon. Most of the riders on these trains alighted at one of the downtown Boston stations, where there were limited options for mode of egress. Consistent with this, walking accounted for the largest share of egress trips from all lines in the old and new surveys, followed by rapid transit. The South Side lines, which offer more choices of alighting stations in Boston, had higher walk-out and lower rapid transit egress shares than the North Side lines, which have only one station in Boston. In 1993, walk-out egress from the North Side lines ranged from 50.1% to 50.5% on the Lowell, Haverhill, and Newburyport/Rockport Lines. In 2008-09 walk-outs from these lines had increased slightly, to between 51.2% and 53.3%. In 1993, The Fitchburg Line had a slightly higher walk-out rate than the other North Side lines, at 54.0%, but in 2008-09 this had dropped to 49.1%. The survey questions asked for egress mode on the survey day rather than usual egress mode. Very late trains on the Fitchburg Line on the survey day seem to have resulted in some shifting from walking egress to rapid transit to save time.
All South Side Lines serve South Station. The Framingham/Worcester, Needham, Franklin, and Providence/Stoughton Lines also serve Back Bay Station. Some trains on the latter three of these lines also stop at Ruggles Station, and some Framingham/ Worcester trains stop at Yawkey Station. Fairmount, Middleborough/Lakeville, Kingston/Plymouth, and Greenbush trains stop only at South Station within downtown Boston, but some trains on the latter three of these also stop at JFK/UMass Station. Some passengers going to the Back Bay area from trains that don’t stop there transfer to outbound commuter rail trains on other lines at South Station to complete their trips. In the 1993 surveys, shares of egress trips by walking from lines serving both South Station and Back Bay ranged from 65.4% on the Providence/Stoughton Line to 79.1% on the Needham Line. Despite not serving Back Bay, the Fairmount Line had a walk-out rate of 75.0%. This was partly because trips from the area served by that line to the Back Bay area also can be made using bus and rapid transit combinations instead of commuter rail. In 1998, walking accounted for 59.9% of egress trips from the Middleborough/Lakeville Line and 64.5% from the Kingston/ Plymouth Line. Walking egress rates from most of the South Side lines changed very little from 1993 to 2008-09. However, after 1998 walking from the Middleborough/Lakeville Line increased to 69.2% (+9.3%), and walking from the Kingston/Plymouth Line increased to 68.8% (+4.3%). Most of these gains were the result of diversions from rapid transit transfers. The Greenbush Line, which was surveyed for the first time in 2008-09, had a walking egress rate of 66.8%.
Between the old and new surveys, almost all of the commuter rail lines showed larger decreases in rapid transit egress than increases in walking egress. The differences were accounted for mostly by increased use of private shuttle van and bus services or bicycles, along with some use of MBTA buses.
Private Mode Egress Times
The 2008-09 survey reports summarized egress times from commuter rail by walking, park-and-ride, getting picked up at the station, other private egress modes, and all of these combined. For inbound trips during the survey hours, walking egress trips accounted for by far the largest share among these four egress categories. Over 92% of the private-mode egress trips from each line in 1993 or 1998 had reported times of 20 minutes or less. There were only slight changes in the percentages under 20 minutes on any of these lines by 2008-09, but most of the lines showed some decreases in the percentages with times of 5 minutes or less. Average egress times by private modes changed slightly on all lines, but all of the increases and decreases were less than 1.0 minute.
For purposes of comparisons between the 2008-09 surveys and prior surveys, reports were produced summarizing the destinations of passengers who boarded inbound trains on each commuter rail line during the span of hours covered in the 2008-09 surveys regardless of their alighting stations. The analysis method used was first to select the destinations with individual shares of 1% or more of the boarding passengers on each line in the 2008-09 survey and then to compare these shares with those for the same destinations from the same lines in 1993 or 1998.
In 2008-09, destinations accounting for individual shares of 1% or more of total destinations represented 82% to 94% of the destinations reported for each of the 11 commuter rail lines, including shares of 86% or more on all but 2 lines. About half of the lines had somewhat lower concentrations of trips to the same destinations in the earlier surveys, and about half had somewhat higher concentrations. The Fairmount Line had an unusually high percentage of trips to unspecified destinations in downtown Boston in 1993, so results from that line are less comparable.
In both 2008-09 and 1993 or 1998, the Financial/Retail neighborhood of downtown Boston accounted for the largest individual share of destinations on every commuter rail line. However, every line had a smaller percentage of its riders going to this destination in the newer surveys than in the older ones, with decreases ranging from 2.1% to 9.8%.
The relative importance of other destinations and changes in the importance of these destinations varied among lines. Other than changes in the shares of Financial/Retail destinations, there were very few increases or decreases of over 4% in the shares of trips to other destinations on any line. The Haverhill Line showed a decrease of 9.0% in trips to Government Center and an increase of 5.7% in trips to Beacon Hill, but these changes may have been partly a result of differences between the two surveys in the classification of destinations on the border of these two neighborhoods. The Lowell Line showed an increase of 5.1% in trips to Government Center. The Franklin Line showed an increase of 4.5% in trips to the Longwood Medical Area.
All lines, especially on the South Side, showed some increase in the percent of destinations in the South Boston industrial area, ranging up to 4.1% on the Franklin Line. The opening of the Silver Line Waterfront routes from South Station to the South Boston Industrial Area and beyond starting in 2004 improved the ease of access to that area from commuter rail.
In the 1993 and 1998 surveys, the lowest checkoff age range was 17 and under, followed by 18 to 24. In 2008-09 this changed to 18 and under, followed by 19 to 24. Otherwise, the checkoff choices were the same. In 1993 and 1998 most of the passengers on each line reported ages in the three ranges between ages 25 and 64, with relatively small percentages reporting ages below 25 or above 64. This was consistent with the predominance of travel to work as the trip purpose. On most of the lines the age range from 25 to 34 had the largest individual share, with the others not far behind. In 2008-2009 the majority of riders were still in the age ranges between 25 and 64, but every line showed significant shift from the two lower ranges in this group into the 45-to-64 age range. On individual lines, the increase in the share of passengers between ages 45 and 64 varied from 13.9% to 22.6%. The Greenbush Line, which was only surveyed in 2008-09, shows an age distribution very similar to those on the other commuter rail lines at that time. All lines also showed decreases in passengers under the age of 25. The survey results showed somewhat higher average household incomes and smaller household sizes for respondents between the ages of 45 and 64 than for those in other age groups. This would make commuter rail relatively more affordable for those in the 45-to-64 age range.
The 1993 survey form did not include a question on gender, so there is no basis for comparison with the 2008-09 surveys. The latter showed that overall 46.2% of commuter rail survey respondents were males and 53.8% were females. These percentages were the same for those in the 45-to-64 age range and in all other age ranges combined. On 8 of the 12 commuter rail lines no respondents reported that they were transgender, and on the other 4 lines, the highest percentage of transgender riders was 0.2%.
The checkoff ranges for household income differed among the 1993, 1998, and 2008-09 survey forms. Among the differences, the 1998 surveys had a top range of $80,000 or more, while the other two had top ranges of $100,000 or more. Income comparisons are of limited usefulness, since it would be expected that inflation would have raised average incomes significantly in the 10 to 16 years between these surveys. On every line, every income range up to $74,999 or $79,999 had a smaller share of respondents in 2008-09 than in the earlier survey, indicating that more households were moving out of each range into a higher one than were moving in from a lower range. The top range showed the largest increase. In 1993, household incomes of over $100,000 were reported by 9.3% to 21.6% of the respondents on each line. Only the Fairmount Line was below 10%, but five other lines were between 10.0% and 13.1%. The two lines surveyed in 1998 had 29.2% and 39.5% of respondents with household incomes over $80,000, but these would have included many between $80,000 and $100,000. In 2008-09 respondents with household incomes over $100,000 accounted for 53.0% to 75.5% of the riders on every line except Fairmount (37.3%), Middleborough/Lakeville (38.5%), and Newburyport/Rockport (43.9%). The Greenbush Line was near the top, at 70.6%.
The 1993 survey form asked how many days per week respondents used commuter rail. The 1998 and 2008-09 forms asked how many days per week they used the line they were riding on when they got the survey. This would be expected to make some differences in the results, because some passengers use different lines on different days. Nevertheless, between 1993 or 1998 and 2008-09 there were few changes of 3% or more in the frequencies of use reported by the passengers on each line.
On the Haverhill Line, five-day use dropped by 3.9%, with offsetting increases in three-day and four-day use. On the Lowell Line, four-day use increased by 3.8%, with apparent shifts from both more frequent and less frequent use rates. On the Fitchburg Line, combined visitor and less-than-one-day use increased by 3.3%, while six-day use showed the largest decrease, at 2.3%. On the Providence/ Stoughton Line, five-day use decreased by 5.5%, while several of the less frequent usage rates showed small increases. This was a result, not of a decrease in the absolute number of five-day riders, but of an increase in off-peak ridership which included fewer repetitive trips than peak ridership.
On the Fairmount Line, the share of riders reporting four-day use increased by 9.2%. This was mostly because cutbacks in off-peak service necessitated by bridge reconstruction resulted in a loss of off-peak riders, many of whom had used the line less than four days a week.
The 2008-09 survey forms provided more checkoff choices for fare type than the older forms. In 1993 and 1998, 12-ride adult-fare tickets and 10-ride half-fare tickets for students, senior citizens, and passengers with disabilities were combined in one category, but the 2008-09 forms had separate check-boxes for 10-ride and 12-ride tickets. Senior and disability single-ride fares were reported separately in 2008-09, but were combined in the earlier surveys. The 2008-09 forms had separate check-boxes for “Family Fare” and “Blind Access Card,” but in the earlier surveys such fares would have had to be included in the “Other” category.
Between 1993 or 1998 and 2008-09, all lines except the Fairmount and Needham Lines had decreases in the use of 10-ride or 12-ride tickets. These decreases ranged from 0.5% on the Lowell Line to 5.4% on the Middleborough/Lakeville Line. On the Framingham/Worcester Line, comparing only results from stations in operation in both 1993 and 2008-09, the decrease in use of these tickets was 7.6%. In 1993 and 1998, 12-ride tickets were priced at the cost of 10 one-way full-fare tickets, but in 2008-2009 there was no discount on 12-ride tickets. This changed the break-even point between 12-ride tickets and passes, and the survey results indicate that much of the reduction in 12-ride ticket use resulted from passengers switching to use of passes. Some lines also had apparent shifts from use of adult one-way full fares to passes. This was particularly noticeable in the Lowell Line results, where the decrease of 5.8% in adult full fares equaled the 5.8% increase in pass use. The break-even point between adult full fares and passes was slightly lower in 2008-09 than in 1993, but this should not have affected the fare mix on the Lowell Line disproportionately. When examined in terms of absolute numbers, the Lowell Line results show that most of the changes in fare-mix on that line resulted from ridership growth that was greater among pass users than among other fare categories.
The fare-mix questions on all of the survey forms allowed pass users to report the zone levels of their passes. Changes in the percentages of pass use by zone were mostly a reflection of differences in the rates of ridership changes at different stations and were consistent with the changes in reported trip origin locations.
On each of the commuter rail lines, the great majority of survey respondents were licensed drivers. In 1993 this ranged from 90.5% on the Fairmount Line to 98.1% on the Providence/ Stoughton Line, and the two lines surveyed in 1998 were both in the middle of this range. These percentages increased or decreased only slightly on each line by 2008-09, with the largest change being an increase to 94.6% on the Fairmount Line.
The majority of riders on each line in 1993 or 1998 reported having vehicles available to make the same trips. Respondents with vehicles available ranged from 78.4% to 84.4% except on the Fairmount Line (71.5%), Franklin Line (87.2%), Providence/Stoughton Line (90.3%), and Kingston/Plymouth Line (92.1%). Many of those with vehicles available drove to their boarding stations and presumably could have used the same vehicles to continue to their destinations.
All lines showed some decreases between the older and newer surveys in the percentages of riders reporting having vehicles available. These changes ranged from 0.7% on the Needham Line to 10.5% on the Newburyport/Rockport Line. Greenbush Line riders had the highest vehicle availability in 2008-09 at 90.1%. The 1993 survey did not include questions on vehicle ownership per household or household size, so it is not possible to determine the extent to which changes in vehicle availability resulted from changes in per capita vehicle ownership. Other reasons for reduced vehicle availability could include other family members having higher priority for use of shared vehicles or older vehicles not being reliable for longer commuting trips. Some differences may simply have been a result in differences in wording of the questions on the survey forms. In 1993 the question was “Did you have an automobile available for this trip?” In 1998 the wording was “Did you have a vehicle available for this trip today?” In 2008-2009, the question after the one on vehicle ownership was “Could you have used one of these vehicles instead of using commuter rail on the day you got this survey?” The 1993 and 1998 questions could have been interpreted by some respondents as referring to vehicles used for access to commuter rail that would not have been available for the entire trip.
The 1993, 1998, and 2008-09 survey forms all included lists of several service quality measures which respondents were asked to rate on a scale of 1 to 5. In all the surveys, 3 was labeled “Average.” In 1993 and 1998, 1 was labeled “Very poor” and 5 was labeled “Very good,” but in 2008-09 the respective labels were “Poor” and “Excellent.” No labels were given to 2 or 4.
There were some differences between the measures listed on the three surveys. In 2008-09 the single measure “Safety and security” replaced “Personal safety at station” and “Vehicle security at station” from the 1993 and 1998 forms. The 2008-09 form included “Station amenities,” for which the most comparable measures from 1993 were “Station condition” and “Station cleanliness.” In 1998 these were combined in the single measure “Station condition/cleanliness.”
Of the seven service quality measures used in comparisons, five rated lower on most lines in 2008-09 than in 1993 or 1998. Ratings for “Reliability (on-time performance)” decreased by 0.4 to 1.1 points on every line between the older and newer surveys. The largest decrease was from 4.0 to 2.9 on the Framingham/Worcester Line, which has had ongoing problems with delays attributed to conflicts with CSX freight operations on the line. Service is expected to improve when the MBTA takes control of the line and CSX relocates much of its freight activity from the line. On the Fitchburg Line, the reliability rating dropped from 3.9 to 3.0, but this was probably partly a reflection of passenger frustration over some very late trains on the survey day. On all other lines, the rating for reliability was between 3.2 and 3.7 despite decreases from earlier surveys. On the Greenbush Line, which was surveyed for the first time in 2009, reliability was rated at 4.5.
Ratings for availability of seating dropped between 0.1 and 0.5 points on each line except Fairmount and Middleborough/Lakeville, where they were unchanged. This is partly a reflection of ridership increasing faster than seating capacity has been added. It may also reflect increased reluctance of passengers to share seats, especially on cars with 3-and-2 seating, where few people sit in the middle seat of a 3-seat row because they feel crowded and uncomfortable.
Ratings for frequency of service declined by 0.5 to 1.3 points on every line except Kingston/ Plymouth and Middleborough/Lakeville, which each had decreases of 0.2 points. On both the Fitchburg Line and the Framingham/Worcester Line, the ratings fell from 4.1 to 2.8. The 1993 survey was conducted before the extension of service to Worcester. The decrease on the Framingham/Worcester Line was consistent with ongoing efforts by communities served by stations west of Framingham to get still more frequent service. However, even limiting the comparison to stations served both in 1993 and in 2008-09, there was a drop of 0.9 points in the rating for frequency of service.
The ratings for travel time/speed were 0.1 to 0.8 points lower on every line in the 2008-09 survey than in the prior surveys. The Lowell Line had the smallest decrease and the Fitchburg and Framingham/ Worcester Lines had the largest decreases. Improvements to the Fitchburg Line intended to reduce travel times, especially from stations on the outer end of the line, are currently in progress in response to known dissatisfaction with current scheduled times. Schedules on the Framingham/Worcester Line were lengthened a few years ago in an attempt to compensate for delays from freight service conflicts. Removal of much of the freight traffic from the line, which is underway, should allow running times to be reduced again.
As noted above, the 2008-09 ratings for stations were based on amenities, while those for previous surveys were based on condition and cleanliness. Stations were rated lower under the newer measure than under the older one on all lines except Lowell, where the rating was unchanged. The largest ratings drops were on the Kingston/ Plymouth Line, which fell from 4.5 to 2.5, and the Middleborough/Lakeville Line, which fell from 4.3 to 2.6. In 1998, the stations on these lines had been open for only one year, which contributed to their rating much higher for condition and cleanliness than stations that had been open for some time. Many of the stations on these two lines have large park-and-ride lots, which necessitated locating them away from town centers that provide the amenities associated with some stations on other lines. Some stations on other lines have enclosed waiting rooms or coffee shops, but none of the stations on the Kingston/Plymouth and Middleborough/Lakeville Lines do.
Ratings for parking availability decreased on some lines but increased on others, reflecting changes in parking capacities and utilization. The Newburyport/Rockport and Fitchburg Lines had the largest decreases, at 0.3 points each. The rating for parking on the Framingham/ Worcester Line increased by 0.9 points when the entire line is included and by 0.4 points when only stations in operation in both 1993 and 2008-09 are included. Diversions of passengers to new stations west of Framingham freed up some parking capacity at the older stations, but many of the riders at the newer stations had not used the line previously.
The only measure rated higher on almost every line in 2008-09 compared with previous surveys was safety and security. Increases ranged from 0.2 to 0.9 points, with the Fairmount Line having the largest increase. The rating for the Middleborough/Lakeville Line was unchanged, while that for the Plymouth/Kingston Line decreased by 0.2 points. The stations on the latter two lines are more isolated than those on most of the other lines, and they have had some problems with vehicle theft and vandalism.
In 2008-09, MBTA water transportation services consisted of two South Shore commuter boat routes and one Inner Harbor ferry route. The South Shore routes ran from Quincy to Long Wharf via Hull and Logan Airport, and from Hingham to Rowes Wharf. The ferry route ran from the redeveloped Charlestown Navy Yard to Long Wharf. The most recent previous comprehensive surveys on these routes had been conducted in 2000. Those surveys were conducted in April, but (at the direction of the MBTA) the new surveys were conducted in the summer of 2008. In 2000 the MBTA funded the Hull service provided by the Quincy route, but the rest of the service on that route was provided independently. Therefore, only passengers boarding at Hull were surveyed in 2000. In 2008, passengers boarding at all docks on this route were surveyed, but comparisons can be made only for the responses from passengers boarding at Hull. In 2000, Inner Harbor service also included routes from Lovejoy Wharf (near North Station) to the Charlestown Navy Yard and to the U.S. Courthouse and World Trade Center. Those routes were discontinued in 2002 due to low ridership.
Ridership on the Hull and Hingham services is heavily concentrated in the direction of Boston during the hours that the 2000 and 2008 surveys were distributed, so comparisons of the results are based only on responses from inbound riders. The Charlestown Inner Harbor route has riders in both directions at all times of day, but during the survey hours in 2008, the total number of responses from passengers boarding at Long Wharf was too low to provide reliable results for comparison with the 2000 survey. In addition, nonrepetitive tourist trips account for a much higher proportion of ridership on that route in the summer than at other seasons. For these reasons, the comparisons of the ferry surveys only use responses from passengers boarding at Charlestown. In cases where changes between 2000 and 2008 were influenced significantly by seasonal variation, additional comparisons were done using only records from passengers that reported no variation in their ferry use by season.
In both sets of surveys, home-based work trips accounted for the vast majority of boardings at Hingham and Hull. Due mostly to seasonal variation, Hingham boardings showed a decrease from 96.7% to 92.9% (-3.8%) in home-based work trips, with an almost equal increase in home-based social or recreational activities.
The Charlestown route is used by many visitors to the historic ships at the Navy Yard, especially in summer months. Among passengers boarding in Charlestown, home-based work trips dropped from 73.0% to 45.9% (-27.1%), while home-based social and recreational trips increased from 8.8% to 16.0% (+8.8%) and non-home/ non-work-based trips increased from 6.8% to 14.4% (+7.6%). These figures probably understate the actual percentage of the latter trips. Visitors who made round trips starting from Long Wharf would have had a chance to fill out surveys while riding toward Charlestown and would not have filled out surveys again on their return trips.
When including only responses from passengers who reported that their use of the Charlestown route did not vary by season, the differences between the results from 2000 and 2008 were smaller than those in the full comparison, but were still significant. Home-based work trips decreased by 15.2%, while home-based trips for social activity increased by 8.8% and home-based trips for personal business increased by 8.6%.
The old and new survey forms allowed passengers to check as many reasons as applicable for using the boats, and most of the choices were the same in both years. In both surveys, “Avoid driving/traffic” was the reason most commonly checked by Hingham passengers, with “Convenience” second. This was also the order of reasons checked by Hull passengers in 2000, but in 2008 “Convenience” and “Avoid driving/traffic” were nearly tied for first place on that line. In contrast, in both surveys among passengers boarding Inner Harbor ferries at Charlestown, “Convenience” was by far the most commonly checked reason in both surveys, with “Avoid driving/traffic” a distant second.
One of the choices on the 2000 survey forms was “Downtown parking cost/availability.” The closest choice to this in 2008 was “Avoid parking at destination.” The latter wording was checked by much larger percentages of riders at all boarding points in 2008 than the former one had been in 2000. Similarly to the results on the other modes surveyed, there were large increases in the percentage of riders checking “Environmentally responsible” as a reason for using boats.
In both surveys, “Only transportation available” was by far the least frequently checked reason for using boats among passengers boarding at Hingham, Hull, or Charlestown.
Unlike the surveys on other modes, both the old and new surveys asked for actual alternative modes used, if any. Percentages in both years were calculated only among respondents who reported using any alternatives. In 2000, driving alone was the most frequently reported alternative among passengers boarding at Hingham, Hull, and Charlestown. In 2008, other MBTA service was ahead of driving alone (63.6% to 55.3%) among Hull passengers. Among Hingham passengers, other MBTA service, at 54.5%, was catching up with driving alone, at 66.0%. These changes were mostly the result of the new alternative provided by the opening of the Greenbush commuter rail line in 2007.
On the Hingham route, the town of Hingham was the largest individual source of origins in both surveys, but its share increased from 34.5% to 45.9% (+11.4%). Between 2000 and 2008 most of the Hingham boat origin locations either got stations on the Greenbush commuter rail line or adjoined towns that did. The largest individual share reduction was from 20.0% to 15.1% (-4.9%) among passengers with trip origins in Scituate, where the Greenbush terminal is located. There is also a station at North Scituate. On the Hull route, all respondents in 2008 reported origins in Hull. In 2000 only one respondent reported an origin other than Hull.
Origins of passengers boarding ferries in Charlestown were heavily concentrated in the Charlestown neighborhood in both surveys, increasing from 96.6% to 98.7%. The sample size of passengers boarding ferries at Long Wharf was too small to allow reliable conclusions about changes in trip origins, but suggest that because of the nonrepetitive nature of many of the trips, origins vary more than those on most MBTA services.
The 2000 surveys asked for means of access to boats, but did not ask about access to initial transit modes, if any, used for boat access. The 2008 surveys included questions both about means of access to initial transit and means of access to the boat. In the survey comparisons, only the final means of transportation by which passengers arrived at their boarding docks were examined. The distributions of access means varied widely among boarding locations.
The vast majority of riders boarding at Hingham reported park-and-ride access, decreasing slightly from 94.4% in 2000 to 91.1% in 2008. This was largely a reflection of the increase in the share of riders from Hingham itself as well as seasonal variation. Walk-ins increased from 0.4% to 3.7%.
Among riders boarding at Hull, park-and-ride was the most common access mode in both surveys, but its share dropped from 76.6% in 2000 to 56.3% in 2008. In the same span, walk-ins increased from 7.4% to 18.8%, and bicycle access from 0% to 5.3%. Overall boardings at Hull more than doubled between the two surveys, as service frequency improved. Seasonal variation and constrained parking at the Hull dock could both have contributed to changes in access modes.
In both surveys, almost all of the respondents boarding in Charlestown reported walking to the dock, with the percentage increasing from 96.2% in 2000 to 98.1% in 2008.
All boarding docks showed slight increases in reported access times by private transportation. Among passengers boarding at Hingham the largest changes were an increase of 5.8% in the share of access trips of between 6 and 10 minutes and a decrease of 2.9% in the share of access trips of 5 minutes or less. At Hull, the share of private access trips of 6 to 10 minutes decreased 6.6%, and that of trips of 11 to 15 minutes increased 4.8%. At Charlestown, the share of private access trips of 6 to 10 minutes increased 11.5%, and that of trips of 5 minutes or less decreased 13.6%.
The 2000 surveys asked for means of egress from boats, but did not ask about egress from final transit modes, if any, used for boat egress. The 2008 surveys included questions both about means of egress from final transit and means of egress from the boat. In the survey comparisons, only the initial means of transportation by which passengers left their alighting docks were examined. The distributions of egress varied widely among the routes.
Walking was the most common means of egress for all boat passengers. For passengers boarding at Hingham, the alighting dock is Rowes Wharf. Walk-outs accounted for 92.7% of the egress trips in 2000 and 94.0% in 2008. Rapid transit egress had the second-largest shares, at 5.8% in 2000 and 3.5% in 2009, although Rowes Wharf is not served directly by the rapid transit system. Some passengers who reported walking as the means of egress may have walked to other transit services to complete their trips.
Passengers boarding at Hull can alight either at Long Wharf or at Logan Airport, but the surveys did not have any responses from passengers going to Logan from Hull. The walk-out egress share of trips from Hull decreased from 78.8% to 71.8% between 2000 and 2008. In 2008, bicycle egress, which had no reported use in 2000, had a 4.7% share. The rapid transit egress share increased from 17.5% to 21.5%. The Aquarium rapid transit station on the Blue Line has an entrance at the inner end of Long Wharf.
Passengers boarding at Charlestown alight at Long Wharf. The walk-out egress share there increased from 84.9% in 2000 to 88.3% in 2008. The rapid transit share decreased from 12.3% to 9.4%.
Because of the large shares of walk-out egress trips on all routes, comparisons of egress times in the old and new surveys were based only on walk-out times. Among respondents alighting from Hingham boats at Rowes Wharf, the share of walking egress times of 5 minutes or less increased from 26.9% to 32.2%, while shares in almost all higher ranges decreased. Among passengers alighting from Hull boats at Long Wharf, changes in the shares of walking egress trips by time range did not show a clear pattern: the shares of trips of 5 minutes or less and of 11 to 15 minutes decreased, while those in the 6-to-10 and 16-to-20-minute ranges increased.
Among passengers alighting from Charlestown boats at Long Wharf, the share of walking egress trips in the 11-to-15-minute range increased by 7.0%, while trips of 5 minutes or less decreased by 3.2% and trips of 6 to 10 minutes decreased by 3.6%.
Destination activities were examined for the Trip Purpose reports, so the comparisons of the Destination Locations and Activities reports looked only at destinations. The analysis method used was first to select the destinations with individual shares of 1% or more of the boarding passengers on each line in the 2008 survey and to compare these shares with those for the same destinations from the same lines in the 2000 survey. In 2008, destinations accounting for individual shares of 1% or more of total destinations accounted for 95% or more of the destinations reported for each of the three route segments examined. Destinations were only slightly more dispersed in 2000. The comparisons below include responses from all passengers surveyed on inbound trips on each route. However, the results change only slightly when only responses from passengers reporting no seasonal variation of the boat in the 2008 surveys are included.
The Financial/Retail neighborhood had the largest shares of destinations of passengers boarding at Hingham, Hull, or Charlestown in both surveys, but the share from every route decreased significantly between 2000 and 2008. Among passengers alighting from Hingham boats at Rowes Wharf, the Financial/Retail share dropped from 61.2% to 48.9% (-12.3%), However, an increase of 8.0% in trips to unspecified downtown Boston destinations probably included some trips that actually went to Financial/Retail.
Among passengers alighting from Hull boats at Long Wharf, the Financial/Retail share of destinations dropped from 40.7% to 23.4% (-17.3%), The largest gains were 5.2% in trips to Beacon Hill, 4.5% in trips to the Fenway neighborhood, and 4.7% in trips to unspecified downtown Boston destinations.
Among passengers alighting from Charlestown boats at Long Wharf, the Financial/Retail share of destinations dropped from 56.3% to 44.2% (-12.1%). The largest gains were 3.3% in trips to Government Center and 4.7% in trips to unspecified downtown destinations.
In the 2000 surveys, the lowest checkoff age range was 17 and under, followed by 18 to 24. In 2008 this changed to 18 and under followed by 19 to 24. Otherwise, the checkoff choices were the same. Similar to the findings for other modes, the latest surveys showed an increase in the average ages of respondents on all the boat routes. Among passengers going from Hingham to Rowes Wharf, the share of riders ages 45 to 64 increased by 9.8% and that of riders age 65 and older increased by 2.8%, while those in the age 25 to 34 range decreased by 9.4% and those in the age 35 to 44 range decreased by 4.5%.
Among passengers going from Hull to Long Wharf, the share of riders aged 45 to 64 increased by 19.9% and that of riders aged 65 and older increased by 5.8%, while those in the 25-to-34 range decreased by 7.3% and those in the 35-to-44 range decreased 22.1%.
Among passengers going from Charlestown to Long Wharf, the share of riders aged 45 to 64 increased by 15.5% and that of riders 65 and older increased by 15.2%, while those in the 25-to-34 range decreased by 22.1% and those in the 35-to-44 range decreased by 6.3%.
The 2000 survey forms included checkoff choices of male and female for gender. The 2008 forms used a write-in format allowing for transgender, but that accounted for only 0.5% of the responses from passengers boarding at Hingham, and none of the responses from passengers boarding at Hull or Charlestown. The Hingham boat was one of the few services surveyed that had more male than female respondents in both 2000 and 2008. Among passengers going from Hingham to Rowes Wharf, 54.7% of respondents were male in 2000 and 58.7% in 2008. Among passengers going from Hull to Long Wharf, 56.2% of respondents were female in 2000, increasing to 62.0% in 2008. Among passengers going from Charlestown to Long Wharf, 54.1% of respondents were female in 2000, increasing to 58.4% in 2008.
The checkoff ranges for household income differed between the 2000 and 2008 surveys, including a change in the top range from “over $80,000” to “$100,000 or more.” Income comparisons are of limited usefulness, since it would be expected that inflation would have raised average incomes significantly in the eight years between these surveys. The changes in patterns varied somewhat among the routes. For passengers going from Hingham to Rowes Wharf, the percentages of respondents in all but the top income ranges decreased between 2000 and 2008. In 2000, 74.7% of these riders reported household incomes of over $80,000. In 2008, 81.0% reported incomes of over $100,000; another 8.6% reported incomes between $75,000 and $99,999, which could have included some below $80,000.
Of passengers going from Hull to Long Wharf in 2000, 54.4% reported household incomes of over $80,000. In 2008, 56.0% reported incomes of over $100,000; another 13.0% reported incomes between $75,000 and $99,999, which could have included some below $80,000. Incomes in the range of $60,000 to $74,999 accounted for a slightly larger share in 2008 than that of incomes in the range of $60,000 to $79,999 in 2000. The share of incomes below $20,000 also increased slightly.
Of passengers going from Charlestown to Long Wharf in 2000, 70.6% reported household incomes of over $80,000. In 2008, 77.1% reported incomes of over $100,000; another 6.5% reported incomes between $75,000 and $99,999, which could have included some below $80,000. Incomes in the range of $30,000 to $39,999 accounted for a slightly larger share in 2008 than in 2000.
As a result of a combination of seasonal variation in use of boats by area residents and inclusion of more visitors in summer ridership, all of the routes showed shifts from five-day ridership to less frequent rates. The 2008 survey forms included a checkoff frequency choice of “I’m only visiting Boston,” but older survey forms did not. In 2000, visitors who filled out surveys most likely would have checked “Less than one day.”
Among passengers going from Hingham to Rowes Wharf, reported five-day use decreased from 76.8% to 67.3% (-9.5%). The largest increases were 2.9% each in two-day use and combined visitor and less-than-one-day use. When including only the responses from passengers who reported no seasonal variation in their use in the 2008 survey, five-day use decreased by 5.2% to 73.6%.
Among passengers going from Hull to Long Wharf, five-day use decreased from 77.8% to 72.7% (-5.1%) and four-day use decreased from 13.6% to 4.6% (-9.0%). The largest increase was from 4.9% to 11.3% (+6.4%) in three-day riders. In 2008, 4.9% of Hull riders used the boat less than one day a week, compared with none in 2000. There were no reported visitors. When including only the responses from passengers who reported no seasonal variation in their use in the 2008 survey, five-day and three-day use were almost unchanged from 2000, but four-day use dropped by 7.8%, and 5.3% were less-than-one-day riders. Some of the increases in infrequent use probably resulted from the greater choice of departure times offered in 2008, making the service a more viable option for riders who might consider using it only occasionally.
Among passengers going from Charlestown to Long Wharf, reported five-day use decreased from 56.9% to 28.6% (-28.3%). The largest increase was in combined visitor and less-than-one day use (+13.2%), followed by two-day (+8.0%) and three-day (+6.8%). When including only the responses from passengers who reported no seasonal variation in their use in the 2008 survey, five-day use decreased by 18.5%, visitor and less-than-one-day use increased by 3.9%, two-day use increased by 3.2% and three-day use increased by 9.9%.
The South Shore commuter boats and Inner Harbor ferries offer a smaller range of fare-payment options than most of the other modes that were surveyed. Changes in the relative prices of single-ride fares, multi-ride tickets, and monthly passes contributed to a number of changes in the percentages of trips using these modes.
In 2000, fare options on the Hingham route included a one-way fare of $4.00, a 10-ride ticket priced at the equivalent of 8.5 one-way rides, and a monthly pass priced at the equivalent of 34 one-way rides. Options in 2008 included a one-way fare of $6.00, a 10-ride ticket priced at the equivalent of 9 one-way rides, and a monthly pass priced at the equivalent of 33 one-way rides. Between the 2000 and 2008 surveys, use of 10-ride tickets decreased 22.7%, from 74.5% to 51.8%. In the same time span, the monthly pass-use share increased by 18.1%, from 21.8% to 39.9%, and single-ride fare use increased by 3.2%, from 1.5% to 4.7%.
In 2000, fare options on the Hull route included a one-way fare of $3.00, a 10-ride ticket priced at the equivalent of 8.3 one-way rides, and a monthly pass priced at the equivalent of 45.3 one-way rides. Options in 2008 included a one-way fare of $6.00, a 10-ride ticket priced at the equivalent of 9 one-way rides, and a monthly pass priced at the equivalent of 33 one-way rides. Between the 2000 and 2008 surveys, use of 10-ride tickets decreased by 61.9% from 92.5% to 30.6%. In the same span, the monthly pass-use share increased by 53.3%, from 3.7% to 57.0%, and single-ride fare use increased by 5.0%, from 1.2% to 6.2%.
In 2000, fare options on the Charlestown-Long Wharf route included a one-way fare of $1.00 and a 60-ride ticket priced at the equivalent of 45 one-way rides. The least expensive monthly pass valid on this route cost the same as 48 one-way rides. Options in 2008 included a one-way fare of $1.70, a 60-ride ticket priced at the equivalent of 54 one-way rides, and a monthly pass priced at the equivalent of 34.7 one-way rides. Compared with the 2000 survey, the results from the 2008 survey for passengers boarding at Charlestown show the share of adult full fares decreasing by 24.3%, from 50.4% to 26.1%, monthly pass use increasing by 13.4%, from 7.8% to 21.2%, and multi-ride ticket use decreasing by 4.8%, from 33.6% to 28.8%. Use of senior citizen and disability fares increased by 15.6%, from 7.3% to 22.5%. These differences were partly a result of seasonal variation in the mix of riders on the route. When comparing only the 2008 survey responses from passengers who reported no seasonal variation in their use of the boat with the 2000 results, the monthly pass use share increased by 18.4% rather than 13.4%, and senior or disability fare use increased only 6.9% rather than 15.6%. These adjustments still do not account for passengers who would have been included in the April 2000 survey but do not use the boat during the summer.
In both surveys, over 98% of the respondents boarding boats at Hingham, Hull, or Charlestown were licensed drivers.
Vehicle availability among passengers boarding at Hingham and Hull was very high in both surveys. Between 2000 and 2008 it decreased slightly, from 97.1% to 96.5% among Hingham riders, and from 91.4% to 88.2% among Hull riders. Vehicle availability among passengers boarding in Charlestown increased significantly, from 69.3% to 79.1% when including all responses, and to 84.8% when including only responses from riders reporting no seasonal variation in their use of the boat in the 2008 survey.
Among passengers boarding at Hingham the percentage from households with two vehicles declined slightly while the percentages in all other report categories increased slightly. The largest changes in household vehicle ownership on that route were a decrease of 6.7% in the range of 1.0 to 1.49 per capita and an increase of 4.2% in the range of 0.01 to 0.49 per capita.
Among passengers boarding at Hull, households with two vehicles declined by 8.0% and those with one vehicle declined by 2.7%, while households with three or more vehicles increased by 11.2%. These changes were partly a reflection of new ridership attracted by increased service frequency on the route. The largest changes in per-capita vehicle ownership were a decrease of 3.6% in the 0.5-to-0.99 range and increases of 3.6% in the 1.5-to-1.99 range and 4.0% in the 2.0-or-more range.
Among passengers boarding at Charlestown, households with no vehicles declined by 7.1% including all responses, or by 6.5% including only riders with no seasonal variation in their use of the boat. All other categories showed increases, with the largest being 3.5% in one-vehicle households for all riders or 4.6% including only all-year riders. Per-capita ownership in the no-vehicle and 0.01-to-0.49 ranges declined, while all other ranges up to 1.5 to 1.99 showed increases. These changes were consistent with the significant increase in reported vehicle availability.
The 2000 and 2008 survey forms both included lists of several service quality measures which respondents were asked to rate on a scale of 1 to 5. In both surveys, 3 was labeled “Average.” In 2000, 1 was labeled “Very poor” and 5 was labeled “Very good,” but in 2008 the respective labels were “Poor” and “Excellent.” No labels were given to 2 or 4.
There were some differences between the measures listed on the two surveys. In 2008 the single measure “Safety and security” replaced “Personal safety on boat,” “Personal safety while boarding/exiting,” and “Vehicle security in parking lot” from the 2000 form. The 2008 form included “Amenities at terminal,” for which the most comparable measure from 2000 was “Condition of docking facilities.”
Among passengers boarding at Hingham, most of the measures were given slightly higher ratings in 2008 than in 2000, with the largest increases being 1.0 in parking availability, 0.6 in safety and security, and 0.4 in on-time performance. Improved parking availability was partly a consequence of ridership diversions from the Hingham parking lot to the Greenbush commuter rail line, which opened in 2007. The rating for amenities at terminals in 2008 was 0.7 points lower than the rating for condition of docking facilities in 2000, and each was the lowest-rated measure in its survey.
Among passengers boarding at Hull, most of the measures were given only slightly higher or lower ratings in 2008 than in 2000, with the largest increases being 0.8 in travel time/speed and 0.5 in frequency of service. In 2000, Hull was served by only two trips a day in each direction, and boats going from Hull to Boston stopped at Quincy and Logan Airport on the way. In 2008, Hull was served by seven trips a day in each direction, including two AM peak trips that ran directly from Hull to Long Wharf. Nevertheless, frequency was the lowest-rated measure in both surveys. The rating for amenities at terminals in 2008 was 1.1 points lower than the rating for condition of docking facilities in 2000.
Among passengers boarding at Charlestown, most of the measures were given only slightly higher or lower ratings in 2008 than in 2000. However, the rating for amenities at terminals in 2008 was 1.4 points lower than the rating for condition of docking facilities in 2000. The rating for parking availability declined by 0.5 points, but since most of the respondents walked to the dock, parking availability would not have been a significant concern for them.
Despite the length of time that had elapsed between the 2008-09 surveys and the prior surveys to which they were compared, there were relatively few large changes in the percentage distributions of responses to the questions included in the comparisons. Among the questions that were directly comparable between the 2008-09 surveys and prior surveys, many did not show changes larger than could be attributed to expected variation in the survey samples. For these reasons, it does not appear to be necessary to undertake such comprehensive survey efforts at frequent intervals. Rather than spreading resources thinly and risking obtaining only marginally acceptable levels of survey responses on some services, it would be preferable to undertake more intensive surveying of more limited numbers of routes or stations on a rotating basis. These efforts should be targeted first to those services for which it appears that a better knowledge of the characteristics of passengers and their travel needs could lead to provision of better service.
TJH/tjh
1 Because this was the first comprehensive survey of Greenbush Line passengers after the line opened in October 2007, the Greenbush survey form included several questions about prior travel modes that pertained only to that line.
The tabulations of the survey comparison results will be posted (in HTML) by June 7, 2013. Should you have any questions regarding the availability of the tabulations, please contact Annette Demchur at ademchur@ctps.org, 617.973.7140 (voice), 617.973.7089 (TTY), or 617.973.8855 (fax).