The Effects of On- and Before- Journey Advantages Using Ride-Sourcing in Indonesia

Round 1

Reviewer 1 Report

The study is an interesting scientific text. It has a logical and complete structure, taking into account theoretical, methodological, empirical and conceptual aspects. The presented results can be a source for comparative research. The scientific value of the article would be enhanced by a slightly broader literary studies. 

Author Response

Response to Reviewer 1 Comments

Point 1: The study is an interesting scientific text. It has a logical and complete structure, taking into account theoretical, methodological, empirical and conceptual aspects. The presented results can be a source for comparative research.

Response 1: Thank you for your kind appreciation to our paper. Furthermore, we appreciate your time to review our paper which your comments help us improve the paper.

Point 2: The scientific value of the article would be enhanced by a slightly broader literary studies.

Response 2: Thank you for your suggestion, we added more review of previous studies in the literature review in section 2 (i.e., The effects of multitasking and travel advantages of using ride-sourcing). Moreover, we also added more literature in the discussion section (section 6) to explain the findings. 

(Please see the attachment)

Author Response File: Author Response.pdf

Reviewer 2 Report

This paper explores the effects of on-journey and before-journey advantages and multitasking when using ride-sourcing on the perceived usefulness of ride-sourcing and in turn the frequency of usage. 

The paper presents a contribution. However, many points can be imporved:

1) the tables are not clear, need improve the presentation of data

2)Authors need improve the Proposed Model Structure section. There are few data about this section.

3)There are many tables, maybe figures can give a better understanding of results.

Overall, authors need improve the english grammar.

Author Response

Response to Reviewer 2 Comments

Point 1: The paper presents a contribution. However, many points can be improved:

 Response 1: Thank you for your kind appreciation to our paper. We also very thankful for your suggestion and review which improve our manuscript.

Point 2: 1) the tables are not clear, need improve the presentation of data

Response 2: Thank you for your suggestion and we revise the presentation of the data (Table 3 and 4 that describe the EFA for Attitude Towards Ride-sourcing and Negative Experiences as well as Built Environment). Moreover, We change Table 2 (the description of multitasking activity during travel) illustration to figure for improving the clarity and information about the data description. We also rearrange the paragraph before and after the table for improving the flow of the explanation.

Point 3: 2) Authors need improve the Proposed Model Structure section. There are few data about this section.

Response 3: Thank you for your suggestion. We added the reason why we used this model as well as the advantages of this model compare to others.

The revision as follow:

“Full-information Maximum-Likelihood SEM (FIML-SEM) and other types of multipath analysis usually required the endogenous variables in continuous responses. However, modified SEM can be used for various data responses (i.e., nominal and continuous data) of endogenous variables, therefore it can provide better results. The modified SEM also can solve the endogeneity issue of using instrumental variables (IV) method as similar with the two and three-stage least square method (2SLS and 3SLS) [34]. Therefore, the modified SEM has more advantages compared to the conventional SEM method [79] (detail discussion of the modified SEM can be found in [34], [80], [81]).

We also added several references regarding the method that we used.

Point 4: 3)There are many tables, maybe figures can give a better understanding of results.

Response 4: Thank you for your suggestion. We change the illustration for Table 2 (the description of multitasking activity during travel) into Figure 1 for improving the clarity and information about the data description.

Point 5: Overall, authors need improve the english grammar

Response 5: Thank you for your concern, we have proofread the paper by professional proof-reader.

(Please see the attachment)

Author Response File: Author Response.pdf

Reviewer 3 Report

This paper presents a detailed survey study about ride-sharing services and finds the correlation among various factors. The findings of the study are expected in common sense. Therefore, how such findings can be used for some (service improving/technology/user comfort, etc.) purposes should be provided, if possible, to highlight the contributions of this study.  The followings are my further comments to improve the paper.

The title is very long, vague, and difficult to understand the main focus of the paper. Make it more concise.

“effects of …advantage … on usefulness…” >> usefulness is also a kind of advantage, and hence the statement is confusing.

From the Abstract, it is difficult to understand the work well, and particularly the first few sentences are not well linked. What is meant by “on-journey and before-journey advantages”? Provide examples or a brief list of them.

“… Structural Equation Model is used”>> state the purpose of using the model in this study. Specify the information used for building the model.

“… ride-sourcing users who perceive (what?) using ride-sourcing positively” >> How are the perceived service evaluated as positive or negative?

Is the proposed model the main contribution? Provide a bit more details in the Abstract. Some observations and discussions on the results should be removed from the Abstract. Thus, considering the above, the Abstract should be revised fully.

The details of the before-journey components is provided in two place in a different way, one is in Line 80 and another is in Conclusion Line 545. Since this term is used in the title, providing the details of it in the Conclusion is very late. It should be given in the Abstract and more details in the introduction.

It is highly likely that travelers who perceive the ride-sourcing useful will use the service frequently, or those who use the service frequently may find ride-sourcing useful. Some unusual findings should be discussed, along with the possible reasons and their impacts.

What are the ranges/scales of the variables used in equation 1 (input of the model in Figure 1)? Alpha 1  is not defined after Equation 1, and the reason for keeping it in braces should be stated.

513: “Experienced users using the services and experience waiting times less than 5 minutes tend to take the service more often.“ >> What are the factors that reduce the waiting time. Are they specific to some users? I guess that experienced  & frequent users could better manage to keep the waiting time low. Therefore, the statement above should be revised.

There are some language issues throughout the paper, e.g.,

186: “the users of these services has become substantial” >>  have

320: “It seems that the current study tends to better to cover the …”

Author Response

Response to Reviewer 3 Comments

Point 1: This paper presents a detailed survey study about ride-sharing services and finds the correlation among various factors. The findings of the study are expected in common sense. Therefore, how such findings can be used for some (service improving/technology/user comfort, etc.) purposes should be provided, if possible, to highlight the contributions of this study.  The followings are my further comments to improve the paper.

 Response 1: Thank you for raising this. Our paper tries to explore the on- and before- journey advantages of the ride-sourcing services effect to the services perceived usefulness and frequency of use. We clarify the contributions of this paper below, as appears  in the introduction (section 1, line 39-46; 67-70;62-65 ) and literature review (section 2.2., line 220-228):

1. This paper tries to provide an understanding about the ride-sourcing behaviour from the countries who have two types of ride-sourcing services (i.e., car and motorcycle) which differ than western context that only offer one type of mode service (i.e., car).
2. While several scholars has investigated the ride-sourcing behaviour which using socio-demographic variables (e.g., income, age, employment, education, and races) and variables such as land use, built environment, the use of internet or social media, and instrumental variables (e.g., frequency of long-distance trips) as predictors, this paper includes non-instrumental (i.e., perception, attitude, and satisfaction) variables to explain ride-sourcing behaviour. This is in line with the suggestion by other scholars who has including these variables in the investigation of public and private transport behaviour.
3. While we use 26 questions of positive and negative experience (ranging from the services quality, productive benefits, applications reliability, and negative situations) using ride-sourcing services and classified into on- and before- journey advantages, other studies represent the instrumental variables as a variable of attitudinal questions only in their ride-sourcing research.
4. Lastly, this paper contributes for the development of policy in managing ride-sourcing in Indonesia as well as other developing countries, who has experienced substantial increase of ride-sourcing usage and has initiated a strong development of urban public transport services (i.e., mass rail transport, light rail train, and bus rapid transit) for the coming decades.

Furthermore, we clarify the recommendations from the findings of this study below and in the discussion and conclusion section (section 6, line 571-576; 595-603):

1. Navigation technology: Perceiving to have better travel time still correlates with a high magnitude of usage. With the high-level congestions in urban areas, it is the needs for more efficient travel time trip has influenced people to use ride-sourcing [21], [64]. Therefore, more advanced navigation systems in route selections and improving the familiarity of the operation areas of the drivers can be suggested as ways to reduce travel time and, in turn, increase usages.
2. Pooling ride-sourcing driver in more potential locations: While it is proposed that ride-sourcing providers to pool the drivers around dense activity location to reduce waiting time, this pooling management might not need in good neighbourhoods located in some suburban areas.
3. Potential incentives for users: Since beginner users are those who perceive negative experiences and take the mode infrequently, free rides or discounts can be provided to those who are new users or who have used the mode less than 2 months for improving the perceived usefulness of the mode and increasing the usage. Incentives such as discounts should be given to users who have a waiting time of more than 15 minutes; that can improve the perceived usefulness and the usage. However, given the issue of ride-sourcing competition with public transport, this incentive should be considered carefully, therefore the ride-sourcing not substituted the public transport demand [9], [61]. Integrating the financial incentives for ride-sourcing trips to public transport stations can be alternatives to support the public transport services. For an example, the provision of incentive to use ride-sourcing as a feeder to public transport, where the incentive will be covered by government as a part of subsidy to support integration between ride-sourcing and public transport.

.

Point 2: The title is very long, vague, and difficult to understand the main focus of the paper. Make it more concise.

Response 2: Thank you for your suggestion, we change the title of the paper to “The effects of on- and before- journey advantages using ride-sourcing in Indonesia”

Point 3: “effects of …advantage … on usefulness…” >> usefulness is also a kind of advantage, and hence the statement is confusing.

Response 3: Thank you for your concern. We have reviewed several previous studies that investigate the benefits or perception of ride-sourcing services to the usefulness/loyalty/ frequency of used in the section 2. We found various findings which indicated that not all of the ride-sourcing benefits and perceptions (i.e., travel cost, reliability of the apps, travel time) has not influenced positively to the ride-sourcing use or its’ perceived benefit or loyalty. Looking at this fact, we believe that we have to investigate the various types of advantages of ride-sourcing to the perceived usefulness and use. For instance, this study found the positive effect of vehicle quality to ride-sourcing use, while we found no effect of the services coverage to ride-sourcing use.

Thus, we argue that we need break down in more detail the idea of advantages and differentiate the idea of advantages with usefulness. Something perceived as an advantage does not necessarily always useful for the users. Usefulness have an effect in the long run.

Point 4: From the Abstract, it is difficult to understand the work well, and particularly the first few sentences are not well linked. What is meant by “on-journey and before-journey advantages”? Provide examples or a brief list of them.

Response 4: Thank you for your questions, we added the brief list of the on-journey and before-journey advantages in the abstract. We define the on-journey in the services characteristics of the services during the travel time such as travel and waiting time as well as driver quality and before journey is the characteristics of non-travel time quality (considerations before decide to use ride-sourcing) such as reliability of application or coverage of the services in the introduction section.

It is similar to public transport, travelling with ride-sourcing includes journey (e.g., travel and waiting time) and before-journey (e.g., reliability of the applications and the service coverage area) components, whereas before-journey components are not found in taking private vehicles.

Point 5: “… Structural Equation Model is used”>> state the purpose of using the model in this study. Specify the information used for building the model.

Response 5: The revision as appears in section 4 line 377-389 is as follow:

“Full-information Maximum-Likelihood SEM (FIML-SEM) and other types of multipath analysis usually required the endogenous variables in continuous responses. However, modified SEM can be used for various data responses (i.e., nominal and continuous data) of endogenous variables, therefore it can provide better results. The modified SEM also can solve the endogeneity issue of using instrumental variables (IV) method as similar with the two and three-stage least square method (2SLS and 3SLS) [34]. Therefore, the modified SEM has more advantages compared to conventional SEM method [79] (detail discussion of the modified SEM can be found in [34], [80], [81]).

Point 6: “… ride-sourcing users who perceive (what?) using ride-sourcing positively” >> How are the perceived service evaluated as positive or negative?

Response 6: Thank you for your correction, we revised the sentences. We found that the users that appreciate the ride-sourcing services tend to use the services more frequent.

The revision as follow:

“As expected, ride-sourcing users who appreciate the usefulness of ride-sourcing services correlate with a higher frequency of using the mode.”

Point 7: Is the proposed model the main contribution? Provide a bit more details in the Abstract. Some observations and discussions on the results should be removed from the Abstract. Thus, considering the above, the Abstract should be revised fully.

Response 7: Thank you for raising this issue. The main contribution of this study is the understanding regarding the on-journey and before-journey advantages and its effects on the perceived usefulness and frequency of use. This contribution is filling the gap of knowledge in the field of travel behaviour. Regarding the proposed model, we argue that the model is an additional contribution in studying travel behaviour where the model has been applied in several previous studies.

Based on your questions, we agree to revise the abstract as follows:

This paper explores the effects of on-journey (i.e., waiting and travel time reliability, driver quality) and before-journey (i.e., services coverage, application quality, fare, etc.) advantages and multitasking when using ride-sourcing on the perceived usefulness of ride-sourcing and in turn the frequency of usage. Assuming a structural form without any reciprocal effects, the modified Structural Equation Model is used. This study collected data from 497 ride-sourcing users in Bandung city in 2018. As expected, ride-sourcing users who appreciate the usefulness of ride-sourcing services correlate with a higher frequency of using the mode. This study found that the situational variables or variables related with travel and built environment conditions (e.g., on-journey advantages, built environment, and travel characteristics) play a role in keeping the travellers using the services. This study confirms that multitasking is not a reason for ride-sourcing users to take the services more often. Moreover, those who previously used motorcycles and CBRS users are more loyal travellers than car and public transport users and MBRS, respectively.

Point 8: The details of the before-journey components is provided in two place in a different way, one is in Line 80 and another is in Conclusion Line 545. Since this term is used in the title, providing the details of it in the Conclusion is very late. It should be given in the Abstract and more details in the introduction.

Response 8: Thank you for your concern, we added the detail of on- and before journey components in the abstract.

The revision as follow:

“This paper explores the effects of on-journey (i.e., waiting and travel time reliability, driver quality) and before-journey (i.e., services coverage, application quality, fare, etc.) advantages and multitasking when using ride-sourcing on the perceived usefulness of ride-sourcing and in turn the frequency of usage.”

Point 9: It is highly likely that travelers who perceive the ride-sourcing useful will use the service frequently, or those who use the service frequently may find ride-sourcing useful. Some unusual findings should be discussed, along with the possible reasons and their impacts.

Response 9: Thank you for raising this issue. Similar with previous studies in the context of public transport in developing and developed countries, we also found that perceived usefulness also shaped the frequency of use in ride-sourcing modes. We believe it makes sense, given the rationale that if the people feel more benefit of the services, therefore, they take the services more often.

Moreover, we argue that there are no unusual findings. However, we found the different result from several studies in developed countries, which are mainly because the different background and socio-economy-culture characteristics.

For an example, as appears in line 584-594 as follows:

Different from research in developed countries [11], [35], [87], [88], educated generation X and older people are those who perceive this service to be useful and take this mode more often. On the other hand, educated millennials may not feel this service to be useful but keep using this mode. This is presumably because generation X feels that this mode improves their mobility compared to what it was before the existence of this mode. Males are those who take this mode more often, as in the UK [87]. In developing countries, particularly Indonesia, males are those who are workers [33], thus this mode seems to help male workers to improve their mobility. Moreover, motorcyclists and CBRS users are more loyal travellers than car and public transport users and MBRS, respectively. Therefore, CBRS users and motorcyclists who are part of educated gener-ation X and older can be offered a loyal user programme (e.g., as premium or titanium users with more benefits in using this service).

Point 10: What are the ranges/scales of the variables used in equation 1 (input of the model in Figure 1)? Alpha 1 is not defined after Equation 1, and the reason for keeping it in braces should be stated.

Response 10: Thank you for your concern, we added clarification of the ranges of the scales of the variables that used in the equation 1.

The revision as follow:

“Moreover, there are various scale of the predictors that used in the models. Since most of personal and travel characteristics, including multitasking activities during travel, are nominal variables, for incorporating into regression analysis, we converted all these into dummy variables. Dummy variables are used because it give us opportunities to use a single regression equation to represent multiple categories [80]. Latent variables such as Bi, Oi, Di, and Ei are treated as continuous variables that ranging from minus ∞ to positive ∞ as they are generated using factor analysis.”

Point 11: 513: “Experienced users using the services and experience waiting times less than 5 minutes tend to take the service more often.“ >> What are the factors that reduce the waiting time. Are they specific to some users? I guess that experienced  & frequent users could better manage to keep the waiting time low. Therefore, the statement above should be revised.

Response 11: Thank you for your concerns. The characteristics of ride-sourcing services is users order the services in their smartphone and driver that accepted the order come to pick the users. Therefore, the waiting time is influenced by the location of the driver which influence to the time needed to pickup users.

As we found that less waiting time and latent variables of reliable travel time and perceived productive benefit influenced the trip frequency, we suggested that ride-sourcing providers might need to pool the drivers around the activity locations which will decrease users’ waiting time.

The description of the recommendation is written in discussion and conclusion section as follow:

“Even though travellers know that Bandung is a congested city, perceiving to have better travel time still correlates with a high magnitude of usage. With the high-level congestions in urban areas, it is the needs for more efficient travel time trip has influenced people to use ride-sourcing [21], [64]. Therefore, more advanced navigation systems in route selections and improving the familiarity of the operation areas of the drivers can be suggested as ways to reduce travel time and, in turn, increase usages. While it is proposed that ride-sourcing providers to pool the drivers around dense activity location to reduce waiting time, this pooling management might not need in good neighbourhoods located in some suburban areas as defined by Tarigan et al. [89] and Dharmowijoyo et al. [68] but focus the services near the city centre to reach more loyal users. There is a tendency for ride-sourcing to be used for short and medium distances in developing countries, as also found in other studies [32], [63], [65], for reaching public amenities or shopping centres or public transport networks around the city centre.

This study confirms that multitasking is not a reason for ride-sourcing users to take the services more often. Those who are enjoying the view or reading/sending emails during the journey are those who perceive negative experiences during the services, while those who are enjoying social interactions with the driver tend to be infrequent users.

Different from research in developed countries [11], [35], [87], [88], educated gen-eration X and older people are those who perceive this service to be useful and take this mode more often. On the other hand, educated millennials may not feel this service to be useful but keep using this mode. This is presumably because generation X feels that this mode improves their mobility compared to what it was before the existence of this mode. Males are those who take this mode more often, as in the UK [87]. In developing countries, particularly Indonesia, males are those who are workers [33], thus this mode seems to help male workers to improve their mobility. Moreover, motorcyclists and CBRS users are more loyal travellers than car and public transport users and MBRS, respectively. Therefore, CBRS users and motorcyclists who are part of educated gener-ation X and older can be offered a loyal user programme (e.g., as premium or titanium users with more benefits in using this service).

Since beginner users are those who perceive negative experiences and take the mode infrequently, free rides or discounts can be provided to those who are new users or who have used the mode less than 2 months for improving the perceived usefulness of the mode and increasing the usage. Incentives such as discounts should be given to users who have a waiting time of more than 15 minutes; that can improve the perceived usefulness and the usage.”

Point 12: 513: “There are some language issues throughout the paper, e.g.: 186: “the users of these services has become substantial” >>  have 320: “It seems that the current study tends to better to cover the …”

Response 12: Thank you for your correction, we revised the sentences. Moreover, the manuscript also been professionally proofread before we submitted.

(Please see the attachment)

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

The authors have addressed my comments and I have no further comments.