Attitudinal dataset for mediating the effects of public acceptance on bus reform scheme in a developing country context

This paper presents a comprehensive dataset of how public perceptions and its statistical implication on the behavioral adaptation toward the acceptability of bus reform scheme proposed by the government of Aceh, Indonesia. Studies indicate that public response to such a proposed policy is significantly related to the effectiveness of its implementation. Given the absence of studies from the developing countries context, a stated preference (SP) questionnaire is developed to investigate the public's consciousness concerning bus reform acceptance. A paper-pencil based questionnaire survey performed by direct interview was conducted in late 2017 and early 2018. The questionnaires were distributed with a total of 450 questionnaires valid. The most pertinent inquiries in our questionnaires were designated to attain (i) attitudinal indicators toward the acceptance of bus reform policy, (ii) attitudinal indicators related to perceived appropriateness of the policy, (iii) perceived awareness of problem private-mode in society, (iv) private-mode dependency, (v) inhibition of freedom of movements and (vi) correct and acceptable policy. We implemented a 4-point Likert scale such as 1 to 4 (strongly disagree to strongly agreed) in order to style the questionnaire easy to answer. A valid dataset was analyzed using Confirmatory Factor Analysis (CFA) for revealing how public perception has statistically significant explained the perceived effectiveness of the proposed bus reform policy.


a b s t r a c t
This paper presents a comprehensive dataset of how public perceptions and its statistical implication on the behavioral adaptation toward the acceptability of bus reform scheme proposed by the government of Aceh, Indonesia. Studies indicate that public response to such a proposed policy is significantly related to the effectiveness of its implementation. Given the absence of studies from the developing countries context, a stated preference (SP) questionnaire is developed to investigate the public's consciousness concerning bus reform acceptance. A paper-pencil based questionnaire survey performed by direct interview was conducted in late 2017 and early 2018. The questionnaires were distributed with a total of 450 questionnaires valid. The most pertinent inquiries in our questionnaires were designated to attain (i) attitudinal indicators toward the acceptance of bus reform policy, (ii) attitudinal indicators related to perceived appropriateness of the policy, (iii) perceived awareness of problem privatemode in society, (iv) private-mode dependency, (v) inhibition of freedom of movements and (vi) correct and acceptable policy. We implemented a 4-point Likert scale such as 1 to 4 (strongly disagree to strongly agreed) in order to style the questionnaire easy to answer. A valid dataset was analyzed using Confirmatory Factor Analysis (CFA) for revealing how public perception has statistically significant explained the perceived effectiveness of the proposed bus reform policy.
© 2019 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY license (http://creativecommons. org/licenses/by/4.0/).

Data
The dataset presents in this paper containing the socio-demographic characteristics, travel attributes, and attitudinal indicators. The dataset obtained from the questionnaire survey using Stated Preference (SP) experiment [1e5]. Target respondents were commuters, commercial visitors, employees and students within the proposed bus reform policy so-called "Trans Koetaradja" (hereafter "TK"). The TK is a new technology and bus system with a capacity of 60 passengers equipped with an air conditioner, providing more comfort to the passengers [6e8]. TK reforms required to achieve efficiency and sufficiency of the city bus system. The TK now is under trial runs, a free of charge service is applied within running corridors. The trial run corridors of TK are within the city of Banda Aceh (see Fig. 1a) as the city heavily relies on private modes such as cars and motorcycles [9,10]. The target area of this study is corridor 1 (red), 3 (dark blue), and 4 (yellow) as shown in Fig. 1b.
The distributions of socio-demographic and travel attributes are described in Table 1 and Table 2. The attitudinal indicators which are consist of the information related to (1) attitudinal indicators toward the acceptance of bus reform policy; (ii) attitudinal indicators related to perceived appropriateness of the policy; (iii) perceived awareness of problem private-mode in society; (iv) private-mode dependency; (v) inhibition of freedom of movements, and (vi) correct and acceptable policy has illustrated in Table 3.

Experimental design and material
The data for our work is consist of socio-economic characteristics, mobility/travel attributes and attitudinal indicators related to public response to the acceptance of Bus Reform policy. As for socioeconomic and travel attributes distribution can be seen in Tables 1 and 2 Concerning with attitudinal data, the SP experimental [11] is applied to deal with attitudinal data in qualitative nature. The SP with a 4-point Likert scale is employed to wide-ranging information related to the attitudinal questions. To gain more assistance by the respondents, the questionnaire form has designed as simple as possible. The vital of our questions are attitudinal intentions toward bus reform policy by the government. In this case, we adopted preceding works done by [11e13].      Table 3 shows the distribution of the answered responses to attitudinal questions using a 4-point Likert scale 1 to 4 (strongly disagree to strongly agreed) as illustrated in Fig. 2 and summarized in Table 4.

Statistical framework of analysis
A confirmatory factor analysis (CFA) approach is used to delve more deeply related respondent's psychological intents. The CFA is a kind of structural equation modeling (SEM) that deals with measurement models. That is the relationships between observed measures or indicators such as attitudinal related questions as described in Table 3. Thus, the central concern of CFA is modeling factors or latent variables that are not directly measured but manifest from psychological perceptions or indicators. Therefore, in this study, CFA is applied and used to investigate the determinant factors to accept bus reform policy based on the unobserved variable (latently). Our hypothesized in this work were (1) perceived appropriateness of the policy and awareness of problem private-mode in society would have a positive effect on their intentions to support the proposed policy; (2) determinant of private-mode dependency and inhibition of freedom of movement/less accessibility of the bus would have an adverse consequence on such intentions. By implementing CFA able to delve more deeply related respondent's behavioral intentions respect to the proposed policy. The CFA model as a part of the multiple indicators multiple causes model proposed by [14,15] systematically the CFA or known as a measurement model, given by: Where y i is a vector of observable attitudinal indicators, hi is a vector of mediators (latent variables), L are matrices of unknown parameters to be calibrated, and the terms z i are vector of measurement  errors. More application related to the latent variable modeling within a broad of travel behavioral arena can be surveyed in the previous studies done by [1e5].
Statistically speaking, the model and its parameters must be tested using several statistical diagnose tests. We used several model fit indexes to test the goodness of fit model namely the root mean square error of approximation (RMSEA ¼ 0.059), the Adjusted Goodness of Fit Index (AGFI ¼ 0.915), the comparative fit index and the Tucker-Leis index (CFI ¼ 0.939, TFI ¼ 0.972), and c2/degrees of freedom (c2/df ¼ 2.58) as described in Table 5. The result of calibrated parameters among mediators (latent 1e5) and its indicators (question Q1-Q15) using CFA can be seen in Table 5. Moreover, the estimated parameters among mediators' latent variables describe in Fig. 3. For the readers who interested in the discussion about statistical inference from the empirical modeling using CFA are invited to read [8]. Table 6. Appendix. The summary of the English translation of the questionnaire The questionnaire of attitudinal data is structured into five parts of the mediator (latent variable). The attitudinal indicators which are consist of five parts of the information related to (1) attitudinal indicators toward the acceptance of bus reform policy; (ii) attitudinal indicators related to perceived appropriateness of the policy; (iii) perceived awareness of problem private-mode in society; (iv) private-mode dependency; (v) inhibition of freedom of movements, and (vi) correct and acceptable Table 6 Calibrated parameters of CFA: coefficients among latent variables (ellipses) and indicators (rectangular); *** path coefficient significant at a 1% error level [8]. The proposed bus reform is suitable policy to deal with traffic and environmental problem 1.203*** ***,**, and *significant at 1%, 5%, and 10% level.