Volume 39, Issue 9, Pages 2068–2085

Volume 39, issue 9, pages 2068–2085 Due to the vigilance of one of our readers, an error was found in a paper recently published in this journal by Robert Cervero: “Transit-oriented development's ridership bonus†( Environment and Planning A 2007 39 2068 – 2085). Only one inclusive term and its associated θ coefficient was presented and reported for the upper-nest model in table 3 . Two inclusive terms should have been presented, for ‘living near’ and ‘living away’ transit, respectively, as notated in equation (3) on page 2077 of the paper. It is noted that a circulated working paper by the same author (R Cervero and M Duncan, “Residential self selection and rail commuting: a nested logit analysis†, 2002, working paper, University of California Transportation Center, Berkeley) presented two inclusive terms (and their associated coefficients). However, further investigation revealed these were miscalculated and were in error. This working paper has also been corrected to be consistent with the corrected table 3 below. The corrected upper-nest model has a slightly different specification than originally shown in table 3 of the paper in order to satisfy the condition that the θ values lie between 0 and 1. None of the text about the upper-nest model that appears in the first paragraph of page 2078 and on page 2080 of the paper changes nor do any of the substantive findings or conclusions of the research. The additional variables added to the upper-nest model reveal that low automobile ownership levels tended to be associated with transit-oriented living. Cervero acknowledges that automobile ownership likely both influences and is influenced by transit-oriented living; thus the coefficient on the automobile ownership variables could be subject to endogeneity bias. The revised equation also shows that, controlling for other variables in the equation, having individuals 55 years of age and above in a household reduced the likelihood of living near transit. Lastly, discussions in the last paragraph of section 6.3 regarding the θ values should now be in the plural, indicating that both θ near and θ away are within the 0–1 interval. It is also noted that the estimated coefficients in the lower-nest binomial logit models for predicting rail commuting (shown in the right-hand panel of table 3 ) are unchanged from the original table 3 . The t -statistics presented in the original lower-nest models, however, were actually Wald statistics ( t -statistics squared); in the corrected model, the square roots of the Wald statistics (ie, t -statistics) are presented. None of the significance levels changes. Because a sequential approach was used to estimate the nested logit model, the coefficient estimates of the upper-level model are consistent but not asymptotically efficient, thereby underestimating standard errors and assigning more significance than is likely the case. It is also noted that coefficients in the lower-level models were not transformed using the estimated θs from the upper-level model. Because these coefficients are unchanged, none of the discussions in sections 6.4 and 6.5 (pages 2080 – 2083) of the paper is affected by the corrections made to table 3 . The editors of Environment and Planning A are immensely grateful to the reader who pointed out this important error, and urge our readers to help maintain the quality of the journal by calling our attention to any other significant errors. Robert Cervero also acknowledges the helpful comments and suggestions of two anonymous referees and Kenneth Train in making corrections to the original model. We reproduce below the corrected table (to replace table 3 on page 2079). Table 3. Nested logit model results for upper nest (rail location choice) and lower nest (rail commute choice). Upper nest live near transit Lower nest: rail commute live near transit live away from transit coef. t -stat. coef. t -stat. coef. t -stat. Location factors    workplace within ¼ mile of 0.073 2.92 *** 1.149 10.87 ***       rail station (0–1) 0.477 1.82 * 0.670 5.85 ***    workplace ¼-½ mile of rail       station (0–1)    workplace within 1 mile of 0.345 4 41 ***       rail station (0–1)    job-accessibility index, highway 0.013 2.10 **       network, jobs (in 100 000s)    within 45-minute isochrone       of residence    job-accessibility index, transit 0.105 1.50       network, jobs (in 100 000s)       within 30-minute isochrone       of residence Transportation attribute    travel-time ratio (transit net- −1.422 10.20 *** −1.806 26.33 ***       work/highway network,       centroid to centroid) Household/neighborhood attributes    0 cars in household (0–1) 1.931 7.16 *** 3.468 6.20 *** 3.394 9.48 ***    1 car in household (0–1) 0.859 6.95 *** 1.537 4.24 *** 0.709 5.88 ***    2 cars in household (0–1) 0.302 3.10 *** 0.673 1.82 * 0.400 3.72 ***    lower-income household, 0.129 1.162       annual household       income