Metro Rapid "After" and Limited "Before" Ratings

• The overall rating of MTA service increased by 0.09, from 3.48 to 3.57 among local riders. This change is not statistically significant.

Comparisons Between Metro Rapid and Local Service in the "After" Phase (Table 3)

• Ratings for Metro Rapid bus are higher for all elements of service compared to the "after" Local Bus ratings.
• The largest differential (0.52) between Metro Rapid and Local service is for "cleanliness."
• "Travel time on the bus" shows the next highest differential (0.45). In the "before" surveys, the differential in travel time ratings between the limited and local routes was only 0.13 (as shown in Table 4).
• "Frequency of buses" is third in terms of the greatest differentials between Metro Rapid and Local service (0.44). This finding regarding perceptions of frequency is surprising because, at least on Ventura Boulevard, local buses operated more frequently than Metro Rapid buses.
• The differences in ratings are statistically significant for all service attributes at the p=0.5 level, and for all attributes except "routes go where I need to go" and "availability of seats" at the p=.01 level.

Table 2
Local "After" and "Before" Ratings

Comparisons Between Limited and Local Service in the "Before" Phase (Table 4)

• The differences seen between ratings for Metro Rapid and for local buses are emphasized even further after an examination of the "before" ratings on limited and local service. As Table 4 shows, there were no statistically significant differences in passenger ratings of limited-stop and local service prior to the implementation of Metro Rapid.

Detailed Analysis of Service Attribute Ratings by Riders

Data collected on the before and after on-board surveys provide a wealth of information related to customer perceptions of MTA service attributes. In designing service improvements, MTA staff needs to know not only the customer ratings on individual service attributes but also the importance of each attribute in terms of overall satisfaction. The previous section focused on customer ratings; in this section, we consider the ratings together with the relative importance of each service attribute.

The simplest way to measure importance is to ask the customer to rate each element on a scale of 1 to 5, similar to the performance ratings. The drawback of this method is that it lengthens both the survey instrument and time needed to complete the survey, which in turn could diminish the response rate. An alternate technique to measure the importance of each service attribute is to derive importance by examining the relationship of each attribute to overall satisfaction.

The Bay Area Rapid Transit District in Oakland, CA has developed a practical methodology to derive the importance of individual service attributes. The methodology uses bivariate correlation analysis to estimate the importance of each service attribute. Specifically, Pearson correlation coefficients are calculated between the performance rating of each service attribute and the overall MTA service rating. While there is a degree of intercorrelation among the service attributes, the Pearson correlation coefficients can be used to measure the relative importance of each attribute. Importance is derived by calculating the ratio between the correlation coefficient for each attribute and the median correlation coefficient. An index score of 100 is assigned to the median correlation coefficient. Service attributes with a score above 100 are more correlated with overall satisfaction (as measured by the overall MTA rating), while service attributes with a score below 100 are less correlated.

Table 5 shows the Pearson correlation coefficient and the importance score for each service attribute for the before survey, the Metro Rapid after survey, and the Local after survey. Before limited stop and local services are analyzed together, based upon findings in Table 4 that there were no significant differences in passenger ratings of the two services.

The derived importance ratings are reasonably consistent across all service types. Frequency and reliability rate highly in terms of importance, while convenience ("Routes go where I need to go"), availability of seats and value for fare paid are relatively less important. Before and Metro Rapid After riders attach a high level of importance to travel time, but this attribute is less important to Local After riders, who are using a slower service. Metro Rapid After riders view cleanliness as important (and may have been attracted to Metro Rapid service by the new buses with a distinctive appearance), while Local After riders rate the ease of identifying the right bus as relatively important.

Performance and importance can be related through scatter diagrams, with derived importance on the x-axis and performance ratings on the y-axis. The scatter diagram is divided into quadrants, with an importance score of 100 and a performance rating of 3.5 (midway between "fair" and "good") serving as the dividing lines.

Items in the upper right hand quadrant represent important attributes with high performance ratings. These are things that the transit agency does well that are important to riders. The agency should take whatever actions are required to ensure continued high performance ratings on these attributes.

Items in the upper left hand quadrant receive high marks in terms of performance but are relatively unimportant to riders. Often, attributes in this quadrant receive lower importance ratings from passengers precisely because the agency does a good job in these areas. Riders, like everyone else, tend to take areas in which their needs are met for granted. This suggests that the transit agency needs to continue to monitor service delivery in these areas to ensure high performance, but that these elements of service are not top priorities for improvements.

Table 5

Importance of Service Attributes

Items in the lower left hand quadrant are relatively unimportant to riders and relatively low-scoring in terms of agency performance. While performance levels are relatively low for these attributes, these are not strong candidates for improvement due to their low levels of importance to riders.

Items in the lower right hand quadrant are key priorities for the transit agency. Riders consider these attributes important, but current performance ratings are less than desired.

Figures 1, 2, and 3 are scatter diagrams that relate importance and performance for Before, Local After, and Metro Rapid After riders and services, respectively. Figure 1 shows the results of the Before survey. No service attributes fall into the high importance/high performance quadrant (although Easy to identify the right bus is on the median for importance). Low-importance attributes are split in terms of performance ratings, with two in the upper left hand quadrant and three in the lower left hand quadrant. There are several attributes in the lower right hand quadrant, representing important service elements that need improvement: Information at bus stops, Frequency, Reliability, Travel time on the bus, and Personal safety. The Before quadrant analysis depicts the situation facing MTA and its Board when it made the decision to move forward with the Metro Rapid demonstration program.

Figure 2 presents the quadrant analysis for Local service after the implementation of the Metro Rapid program. Of the five priority items in the lower right hand quadrant on the Before chart, only two remain in the same quadrant. Frequency and Reliability are major service attributes, but Personal safety is now in the upper right hand quadrant, while Information at bus stops and Travel time on the bus are less important now to local riders (those who value Travel time highly are presumably riding Metro Rapid). Operator courtesy is now in the high importance/low performance quadrant, although just barely (its performance rating is a shade below 3.5), and Information at bus stops is on the median line for importance. Overall, however, the situation is improved for Local bus riders today compared to the Before survey.

The dramatic change in perception of MTA performance has occurred among Metro Rapid riders, as shown in Figure 3. Reliability, Frequency, Travel time, Cleanliness, and Operator Courtesy all fall into the upper right hand quadrant representing high levels of importance and performance. Only one service attribute, Availability of seats, has a performance rating below the cutoff mark of 3.5, and this attribute is judged relatively unimportant by Metro Rapid riders. In sharp contrast to the other figures, there are no service attributes in the lower right hand quadrant of Figure 3.

Taken together, the quadrant analyses clearly show that Metro Rapid riders perceive a quantum leap in service performance. Changes of this magnitude in performance ratings are rare, particularly over a relatively short time frame. MTA has essentially raised the bar in terms of service quality for its riders through the Metro Rapid demonstration program.

Demographics

Riders were asked certain questions to ascertain their age, ethnic origin, sex, income, and vehicle availability. The most interesting findings include:

• Metro Rapid has a higher percentage of male riders (54.2 percent) compared to the locals (41.4 percent) and former limited lines (42.6 percent), suggesting that the new service is drawing new, non-traditional riders (see Figure 4).
• Vehicle availability is surprisingly similar for Metro Rapid and local bus riders (Figure 5). Approximately one-quarter of riders in both groups are from households with at least two cars.
• The majority of Metro Rapid and local bus riders report Income levels below $15,000 annually (Figure 6). However, 13.1 percent of Metro Rapid riders have incomes above $50,000.

Previous Mode of Travel

The "After" survey on Metro Rapid asked riders for their previous mode of travel. Table 5 shows the results, with results broken down by Metro Rapid line.

• As expected, most Metro Rapid passengers are former transit users.
• However, 10.8 percent of Metro Rapid riders did not make this trip previously, and another 9.5 percent used a non-transit mode (most likely the automobile). Many of these new riders are new to transit.

Table 6 presents responses regarding the history of transit use

• Nearly 14 percent of Metro Rapid riders began using MTA services within the last three months (since the start of Metro Rapid and the Metro Red Line extension to the SFV). By comparison, only nine percent of local riders began using MTA services in this same time frame.

A summary of responses to all questions concerning rider demographics and usage patterns is contained in the appendix.