Examining the Drivers of E-Commerce Adoption by Moroccan Firms: A Multi-Model Analysis

Comments

Answers

- The title of paper needs to be re-phrased to reflect precisely what has been done in the paper. there are two aspects in the paper, one is studying the key factors affecting E-commerce adoption. the second is suggesting the growth strategies which haven’t been studied. So remove it from the Title. Please note that qualitative research is also missed from the title. Further, Authors can add the Logistic regression or CMP framework, etc. to the Title. Moreover there is a need to determine that your study from Employees’ perspective.

Thank you for your valuable feedback on the title of our paper. We have revised it to more accurately reflect the content of the research. We hope this new title accurately encapsulates our study’s essence.

- The Abstract Sections needs also to be re-written to consider the aforementioned comments. It should include methodology, population, sample, data instrument, study constructs, used software…etc.

- Wrong use of ‘We’, ‘Our’ in Abstract Section.

- In line 8, Wrong use of ‘is’ two times.

- In line 12, bad use of present tense "we use……..".

Thank you for your meticulous feedback. We acknowledge the points raised regarding the abstract and have accordingly conducted a comprehensive revision. The revised abstract now includes the necessary details about the methodology, population, sample, data instrument, study constructs, and used software. Furthermore, we have corrected the language errors, including the incorrect use of pronouns and verb tenses. We trust that these modifications have improved the quality and clarity of the abstract.

- Add qualitative, Facilitating Conditions, Technology Adoption to keywords if applicable. While remove unrelated words such as integration, innovation.

Thank you for your comment. We have revised the keywords in the abstract as per your suggestion. We have added “qualitative”, “Facilitating Conditions”, and “Technology Adoption” to the keywords and removed unrelated words such as “integration” and “innovation”. We believe that these revised keywords accurately reflect the focus and content of our study.

- Wrong claim of paucity of literature and research in context of E-commerce adoption in morocco. Many studies have been conducted in to investigate E-commerce in Many developing countries in Morocco, Asia, Africa and South America.

Thank you for your feedback. We agree that numerous studies have indeed examined e-commerce adoption in various developing countries across Asia, Africa, and South America. However, our claim of paucity was specifically regarding the context of Morocco, which has not been as extensively researched in comparison to some other developing countries.

We apologize if this was unclear in the initial manuscript. In our revised paper, we will clarify that our claim of a research gap pertains specifically to the Moroccan context, rather than the general topic of e-commerce adoption in developing countries. We believe that our study contributes valuable insights to this underexplored area and will ensure that this is clearly conveyed in our revised manuscript.

- In lines 63, 94  ITU, UNCTAD ---------à stands for

Thank you for noting this. We’ve added the full form of the acronyms ITU (International Telecommunication Union) and UNCTAD (United Nations Conference on Trade and Development) in the revised manuscript.

- GDP in line 65.---------à stands for

We have clarified that GDP stands for Gross Domestic Product in the revised manuscript.

- Page-2, Don’t use Footer, rather move the GPT definition to an appropriate place in the text.

We appreciate your feedback. We have relocated the GPT definition to an appropriate location within the main text as per your suggestion in the revised manuscript.

- No previous works section.

Thank you for your feedback. I would like to clarify that our paper indeed acknowledges and integrates previous research in the field of e-commerce adoption, particularly focusing on the Moroccan context, within our Literature Review section. We have also enriched our study with recent relevant research to provide a more comprehensive and up-to-date perspective. In the revision, we will further highlight these references and ensure their contribution to the current study is explicitly stated, to make their incorporation more apparent to readers.

- Capitalize the Word "internet" in the whole text of paper.

We have made the correction throughout the paper and the term “Internet” is now capitalized in the revised manuscript.

- In line 174, I can contend that there are several rather than little works of EC adoption in Morocco and other similar countries (Tunisia, Algeria, Egypt….etc).

Thank you for your insightful comment. We agree with your point that there is a growing body of work on e-commerce adoption in Morocco and other similar countries, such as Tunisia, Algeria, and Egypt. Our statement was not intended to diminish the significance of those studies. Instead, we sought to highlight the need for further research given the unique challenges and opportunities in these regions. We have revised the text to more accurately reflect the current state of research in these areas, acknowledging the important contributions made by these studies. We appreciate your input and believe it has strengthened our work.

- In Line 249-250, Linear is more commonly used than probit and logit. Clarify?

Thank you for your insightful comment. In this study, we opted for the probit and logit models over the linear model mainly due to their distributional characteristics, which are more suitable for the nature of our data. The probit and logit models, unlike the linear probability model, do not produce probabilities outside the [0,1] interval, making them better equipped to handle binary dependent variables. Furthermore, the error distribution functions of the probit and logit models fit our data better than the linear model. However, we acknowledge the widespread use of linear models and understand that the choice among these models depends largely on the specifics of the research question and the nature of the data.

- How did you select the Xi factors. No solid theoretical foundation has been investigated for that choice.

- How did you select the Zi factors (computers, web, IT staff). No solid theoretical foundation has been investigated for that choice.

Thank you for your insightful comments regarding the selection of the Xi and Zi factors in our models. Your feedback provided an opportunity for us to further clarify the basis of our variable selection.

The selection of Xi variables, such as firm size, age, location, economic sector, digital skills of the workforce, and product innovation, were derived from a thorough review of existing literature on the determinants of e-commerce adoption, much of which is presented in the literature review section of our paper. Some of these studies also specifically discuss the context of Morocco, thereby contributing to the relevance of our choice of variables.

Regarding the Zi factors (i.e., the availability of computers, websites, and IT staff), they were chosen as instruments based on their expected influence on facilitating conditions for e-commerce adoption (FCi) and their anticipated lack of direct correlation with e-commerce adoption itself (EAi).

It is important to mention that our selection of variables was also constrained by those included in the questionnaire that forms the basis of our study. As this questionnaire was not designed by us, we selected variables that best fit the research questions and theoretical frameworks of our study, while acknowledging the limitations this imposes.

We have revised the manuscript to clearly state the rationale behind our choice of variables, and we appreciate your comment as it has helped us improve the clarity of our study. We hope this explanation sufficiently addresses your queries, and we are open to further discussions on this matter.

- In Fig-1, what is the difference among the two columns (firms having online sales, firms adopting e-commerce).

Thank you for your inquiry regarding Figure 1. We understand the potential confusion and apologize for any lack of clarity. In our study, we make a distinction between firms that only sell online (28.24%) and those that both buy and sell online (31.23%). We refer to the latter as firms adopting e-commerce, aligning with our adopted definition of e-commerce as a business model that enables both buying and selling of goods and services over the internet. We hope this response clarifies the distinction made in Figure 1.

- Table-1 different count of observations for the variables?

The different count of observations for the variables is due to the presence of missing responses. Not all participants answered every question, resulting in a varying number of observations across variables.

- Table-1 doesn’t include firm website, Internet access, using computers for business process.

We have taken your feedback into account and decided to remove Table 1 from the manuscript to avoid redundancy and potential confusion. We believe that Table 2 provides a comprehensive overview of the descriptive statistics for each variable, making Table 1 unnecessary. We appreciate your insight and guidance in making this decision.

- How did you analyze the data? What is the Software? How is it used?

Thank you for your question regarding our data analysis methods. We used Stata version 17 for data analysis in our study.

Specifically, for the estimation, we used the CMP (conditional mixed process) module proposed by Roodman (Roodman, D. Fitting Fully Observed Recursive Mixed-process Models with CMP. Stata J. Promot. Commun. Stat. Stata 2011, 11, 159-206). This module is specifically designed to estimate multi-equation, mixed process models, potentially with hierarchical random effects. The term "mixed process" implies that different equations can have various kinds of dependent variables, such as continuous (like OLS), tobit (left-, right-, or bi-censored), probit, ordered probit, or fractional probit.

The CMP module offers robust and flexible capabilities to simultaneously estimate multiple equations, with each potentially having different types of dependent variables, making it an ideal tool for our analysis.

- Line 341, "we find……". A wrong tense.

Thank you for your note about tense consistency. We have adjusted the sentence in line 341 to maintain a consistent tense throughout. I appreciate your keen eye for detail.

-Line 349, what about other factors such as sector…….etc.

Thank you for your insightful suggestion. We expanded our analysis to account for other sectoral factors significant at the 5% and 10% levels, thus offering a more comprehensive view of sectoral influences on e-commerce adoption, as detailed in the revised text.

- Table-3, Four items for firm size but three rows of outputs. A missing outputs. The same for firm age…… Please check other factors.

Thank you for pointing out the potential confusion in Table-3. For nominal or ordinal variables with more than two levels, one category is often selected as a reference level (in our case, for example, “Firm age” of ‘5 years and under’). The outputs in the table represent comparisons against this reference level, which is a common practice in logistic regression analysis.

To clarify, for the “Firm age” variable, the youngest firms (aged 5 years and under) were used as the reference category. Firms aged between 11 to 15 years and those over 15 years are respectively 0.46 and 0.26 times less likely to adopt e-commerce than firms aged 5 years and under, as inferred from the significant and negative coefficients.

The same principle applies to the “Firm size” and other variables, and the reference categories were chosen accordingly. I hope this clarifies your question. Please let me know if further explanation is needed.

- Table-3, No explanation of  Constant values, and Facilitating conditions outputs.

Thank you for your insightful comments. In the revised version of our paper, we have provided clear explanations for the constant and facilitating conditions.

The constant or Intercept in our probit model is -1.2083, significant at the 1% level. It denotes the predicted value of the dependent variable when all other explanatory variables are zero.

The facilitating conditions variable represents the presence of the necessary organizational and technical infrastructure that supports the use of e-commerce. Our two-stage methodology initially estimates the instrumental variables (Computers use, Firm’s website, and Internet access) associated with facilitating conditions, then we assess its effect on the e-commerce adoption probability. As our logit model illustrates, facilitating conditions increase the odds of e-commerce adoption by 1.68 times.

- Table-3 , The #of observations is 462. How is this? You said 807 early.

Thank you for your inquiry about the difference in the number of observations. In our study, we initially had 807 responses. However, due to the presence of missing data in some of the variables, the logit and probit models were estimated for a reduced sample where all variables were fully reported. As a result, the number of observations in Table-3, specifically for the estimation of logit or probit models, is 462. This inconsistency has been clarified in the revised version of our paper.

- A need for interpretations of log pseudo, prob> chi2, pseudo R2, rho_12 in Table-3.

We greatly appreciate the reviewer’s suggestion to provide interpretations for the log pseudo, prob> chi2, pseudo R2, and rho_12 statistics in Table-3. The statistic atanhrho_12, which measures the covariance of error terms in our two equations, was found to be negatively significant. This suggests the presence of unobserved factors that negatively influence both e-commerce adoption (the outcome variable) and the endogenous variables (facilitating conditions). The rho_12, however, was found to be insignificant (p-value>0.1), which implies that unobserved characteristics do not significantly influence firms’ e-commerce adoption decisions. This further validates our model and attests to the absence of omitted variable bias. For improved clarity, these interpretations have been added to the revised manuscript. We hope this addresses the reviewer’s concerns and we’re grateful for the opportunity to clarify this aspect of our analysis.

- In Table-3, no explanation of values for probit and CMP values for all variables. Please make interpretations of the numbers resulted.

We appreciate your comment on the need for clarification on the interpretation of the values generated from the probit and CMP models for each variable. In our revised manuscript, we have provided an explanation for the interpretation of these values.

- in Section-6, How did you come with recommendations for business strategy and retention strategy. Try to link them with the results and don’t state general recommendations.

Thank you for your valuable comments. In response to your feedback, we have revised our Conclusions and Recommendations section to better reflect the results of our study and to offer more specific recommendations.

- In comparison with others, state what others revealed then make the comparison. Do not state only "consistent with findings of [   ]. So how the readers can know and guess the results and findings of previous researchers.

Thank you for your constructive feedback. We understand the importance of providing context from previous research and realize that our initial approach may have lacked sufficient detail. We have revised the text to include explicit findings from the cited literature, thereby making the comparison more transparent to the reader. This allows us to show more clearly how our research aligns with, expands upon, or diverges from previous studies.

- No interpretation of all results in Table-3.

We appreciate your comment and agree that more detailed interpretation of the results in Table-3 would enhance the understanding of the readers. We have now added a comprehensive interpretation of all results.

- The discussion Section is general and came isolated from the study results and findings. Re-consider this vital section.

Thank you for your valuable feedback. We have revised this section substantially to more closely align it with our study’s results and findings. We have made concerted efforts to draw direct connections between our research outcomes and the discussions of these outcomes, ensuring that each point in our discussion section is rooted in our empirical findings.

Moreover, we’ve expanded our discussion to engage more deeply with existing literature, allowing us to better situate our findings within broader academic conversations. By discussing our results in the context of prior studies, we’ve sought to highlight the unique contributions of our research to the understanding of e-commerce adoption in SMEs.

- An excellent References Section but there is a need for more relevant references to the current study.

Thank you for your constructive feedback and recognition of our effort in assembling the References section. We acknowledge your suggestion to incorporate more directly relevant references in our study. To that end, we will thoroughly revisit our literature review, identify potential gaps, and ensure inclusion of additional sources particularly centered around the challenges to e-commerce adoption in Morocco and similar contexts. We believe this would add to the comprehensiveness and pertinence of our study.

Comments

Answers

In a literature review, you cannot start a sentence by quoting, for example, "[8] argue..." (line 125, 127, 130, 133, etc.). You should use name of the author and citation, eg. "Smith [8] argue..."

Thank you for your constructive feedback. We have revised the manuscript to properly attribute the referenced work by including the author names along with the citations. The modifications have been applied throughout the manuscript to ensure a consistent and proper citation style.

The literature review is inadequate, above all, there is a lack of reference of the obtained results to other research results, not only in underdeveloped countries.

Thank you for your feedback regarding the literature review. We acknowledge the importance of relating our findings to those of other research results for a broader perspective. We have now revised our literature review to better incorporate and reference relevant studies.

In addition, there is a lack of reference in the literature to the methodology used, whether works on similar topics have used methods like probit, logit or CMP.

Thank you for your constructive feedback. Based on your suggestions, we have revised the manuscript to include references to studies that have employed similar methodologies. These references further support the use of logit, probit, and CMP models in our analysis of e-commerce adoption.

Explain in more detail what it is atanhrho_12 (line 286)

Thank you for pointing out the need for clarification. The atanhrho_12 is a statistic derived from our model, measuring the correlation between error terms of our two equations. Its significant negative value in our study suggests that there are unobserved factors with opposing effects on e-commerce adoption and the facilitating conditions. This absence of omitted variable bias has been clarified in the revised manuscript. We appreciate your attention to this detail.

What is the purpose of the statistics in Table 1, some of them are really strange and difficult to interpret, such as the mean of ‘Gender of the company’s owner’ 0.8079 - what does the mean tell us if it is a binary value. Similarly for ‘Digital platform use. There is no point in giving these descriptive statistics in Table 1, as it is well described and duplicated in Table 2.

The purpose of the statistics in Table 1 is to provide a summary of each variable’s basic descriptive statistics, including the mean, standard deviation, minimum, and maximum values. For binary variables such as ‘Gender of the company’s owner’ and ‘Digital platform use,’ the mean value effectively indicates the proportion of responses coded as ‘1’. For example, a mean of 0.8079 for ‘Gender of the company’s owner’ suggests that approximately 80.79% of firms are owned by the gender coded as ‘1’. Similarly, a mean of 0.0781 for ‘Digital platform use’ suggests that around 7.81% of firms are using digital platforms.

We have taken your feedback into account and decided to remove Table 1 from the manuscript to avoid redundancy and potential confusion. We believe that Table 2 provides a comprehensive overview of the descriptive statistics for each variable, making Table 1 unnecessary. We appreciate your insight and guidance in making this decision.

In the conclusions, please address the limitations, such as the fact that the research is only for a certain region and thus the conclusions of the research are quite limited.

We appreciate your comment on highlighting the limitations of our study in the conclusion. Indeed, as our research focuses on a specific region, the findings may not be universally applicable. We will certainly add a section in our conclusion to address this limitation, emphasizing that while our study offers valuable insights into e-commerce adoption in Morocco, the extrapolation of these findings to other regions should be done cautiously.

Comments

Answers

Theoretical framework for adoption of e-commerce used as basis for the article should be described.

Thank you for your feedback. We have revised the manuscript to provide a more detailed description of the theoretical framework for the adoption of e-commerce. We have incorporated relevant literature and clearly explained the theoretical underpinnings of our study.

Data source is mentioned but more detail explanation should be included.

Thank you for your comment. We have expanded on our data source in the revised manuscript, providing more details about the survey, the data it contains, and how it was conducted.

Researched variables should be explained more clearly.

Thank you for your comment. We have enhanced the clarity of the explanation regarding the researched variables.

Article mentions "enabling conditions" and "facilitating conditions" in one paragraph - findings regarding this should be added and described.

Thank you for your comment. We’ve clarified the terms “facilitating conditions” and “enabling conditions” in the revised manuscript, and we have streamlined the usage of these terms and provide a more detailed analysis of the findings related to facilitating conditions.