Social-Sciences- and Humanities-Based Profiling of Energy Consumers Towards Increasing Demand Response Engagement

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The willingness and barriers to participation in demand response (DR) programs among 284 residents across six European countries is examined in this paper. A questionnaire survey is used to segment participants based on two key dimensions: DR willingness and DR familiarity. Promotion strategies tailored to these segments were then proposed. The following questions for the authors to clarify.

1. Although the two dimensions of DR willingness and familiarity are effective in practice, more theoretical support is needed as to why these two dimensions are more suitable as the basis for segmentation than other potential dimensions.

2. In Table 2, the sample sizes varied significantly across countries, and the sample sources were unique, which may limit the generalizability of the findings. A discussion of the sample limitations and their impact on the results is recommended.

3. The study found significant differences in DR participation across countries, but lacked analysis of the underlying reasons behind these differences. A more in-depth discussion is recommended, taking into account various factors such as energy policies, electricity pricing structures, and cultural factors.

4. Table 2 is information-dense and could be broken down or simplified to highlight key comparison points. The Sankey diagram in Figure 1 could be annotated to clarify the meaning of the flow directions.

5. The paper does not report the reliability and validity of the questionnaire. It is recommended that the pre-testing and validation steps in the questionnaire development process be supplemented to enhance the rigor of the method.

Author Response

The willingness and barriers to participation in demand response (DR) programs among 284 residents across six European countries are examined in this paper. A questionnaire survey is used to segment participants based on two key dimensions: DR willingness and DR familiarity. Promotion strategies tailored to these segments were then proposed. The following questions are for the authors to clarify.

1. Although the two dimensions of DR willingness and familiarity are effective in practice, more theoretical support is needed as to why these two dimensions are more suitable as the basis for segmentation than other potential dimensions.

The variance, along with the corresponding references, has been added to support the dimensions selection.

2. In Table 2, the sample sizes varied significantly across countries, and the sample sources were unique, which may limit the generalizability of the findings. A discussion of the sample limitations and their impact on the results is recommended.

In the Introduction section, we have added a new subsection: “Contribution and positioning”, which clearly clarifies our contribution and the significance of our findings. We have added an explicit paragraph on sampling limitations and generalizability in the Discussion section.

3. The study found significant differences in DR participation across countries, but lacked analysis of the underlying reasons behind these differences. A more in-depth discussion is recommended, taking into account various factors such as energy policies, electricity pricing structures, and cultural factors.

The differences span far further than those generic factors, because the group profile is very specific per country. Besides the added analysis in the “Contribution and positioning” and the “Discussion” parts, the scoring across each pilot and its specific segments is now available in the Appendix section.

4. Table 2 is information-dense and could be broken down or simplified to highlight key comparison points. The Sankey diagram in Figure 1 could be annotated to clarify the meaning of the flow directions.

Table 2 has been enhanced with highlights that help locate the key distinguishing features of each participant group.

5. The paper does not report the reliability and validity of the questionnaire. It is recommended that the pre-testing and validation steps in the questionnaire development process be supplemented to enhance the rigour of the method.

The reliability of the questionnaire is covered by the Cronbach’s alpha calculations, while the validity is covered by a factor analysis.

Reviewer 2 Report

Comments and Suggestions for Authors

The paper investigates residential demand response (DR) participation using a questionnaire-based segmentation across six European countries. The topic is relevant and the approach that links social sciences with energy research is interesting. However, there are several points that need to be addressed as follows:
-    The study highlights consumer segmentation for DR engagement, but similar survey-based approaches have already been reported in the literature. The paper has applied the method to new country samples, but it is not clear to me if the work provide significant findings that are different from conclusions of other samples.
-    The segmentation process depends on DR Willingness and DR Familiarity, but there is no statistical evidence that these are the most suitable or sufficient axes for clustering. It seems that the segmentation method is more descriptive rather than analytically strong. It is important to explain and justify this choice.
-    Several pilots are based on very small numbers (Italy and Spain with only 17 responses). There is a large gap in number of responses, compared to Romania for example with 112. The results from these small groups may not be sufficient to draw reliable conclusions.
-    The results section presents averages and segment profiles, but more statistical analysis such as correlations or regression models may help find which factors, such as energy literacy, income, or others, are affecting DR willingness.  
-    It is recommended to add discussion about how utilities could enhance communication or design incentives based on the findings.
-    The current organization of the manuscript may need to be enhanced. The results section should be after the methods since it is important to explain the data collection and analytical choices first then report the results.

Author Response

The paper investigates residential demand response (DR) participation using a questionnaire-based segmentation across six European countries. The topic is relevant and the approach that links social sciences with energy research is interesting. However, there are several points that need to be addressed as follows:

-    The study highlights consumer segmentation for DR engagement, but similar survey-based approaches have already been reported in the literature. The paper has applied the method to new country samples, but it is not clear to me if the work provides significant findings that are different from the conclusions of other samples.

In the Introduction section, we have added a new subsection: “Contribution and positioning”, which clearly clarifies our contribution and the significance of our findings.

-    The segmentation process depends on DR Willingness and DR Familiarity, but there is no statistical evidence that these are the most suitable or sufficient axes for clustering. It seems that the segmentation method is more descriptive rather than analytically strong. It is important to explain and justify this choice.

The justification has been provided through question variance both in the main body and the appendix.

-    Several pilots are based on very small numbers (Italy and Spain, with only 17 responses). There is a large gap in the number of responses, compared to Romania, for example, with 112. The results from these small groups may not be sufficient to draw reliable conclusions.

We agree and have added an explicit paragraph on sampling limitations and generalizability in the Discussion section.

-    The results section presents averages and segment profiles, but more statistical analysis, such as correlations or regression models, may help find which factors, such as energy literacy, income, or others, are affecting DR willingness.

The justification has been provided through Cronbach’s alpha and factor analysis.

-    It is recommended to add a discussion about how utilities could enhance communication or design incentives based on the findings.

We agree and have added a dedicated discussion on utility-facing implications, outlining segment-specific communication strategies. These additions appear in the Discussion section.

-    The current organisation of the manuscript may need to be enhanced. The results section should be after the methods, since it is important to explain the data collection and analytical choices first, then report the results.

We have changed the organisation of the manuscript as suggested by the reviewer.

Reviewer 3 Report

Comments and Suggestions for Authors

The comments of this reviewer can be found below:

1. The hierarchical segmentation approach has been used but it is not well justified why it has been used. Its advantages over other well-known clustering algorithms such as latent class analysis need to be discussed.
2. It is required to elaborate more on recruiting the participants and state the critical limitations of data representativeness and generalizability.
3. The literature review has poorly been written and it does not support the core idea and novelty of the paper.
4. The validation process of the questionnaire such as the pilot testing should be reported and the reliability metrics should be provided for Likert process.
5. The Sankey diagram illustrated in Fig. 1 is not consistent with the segmentation approach employed for the Spanish and Austrian pilot size.
6. The analyses require statistical tests like ANOVA and chi-square to validate the importance of the differences observed between the derived consumer segments.
7. The concept persona needs a more structured definition while its derivation from the segment data should be clearly explained.
8. A Conclusion section needs to be added.

Author Response

The comments of this reviewer can be found below:

1.The hierarchical segmentation approach has been used, but it is not well justified why it has been used. Its advantages over other well-known clustering algorithms, such as latent class analysis, need to be discussed.

Due to the limited data size, no elaborate approaches are applicable; notwithstanding, the variance of questions has been presented both in the main body and the appendix as a quantitative justification.

2.It is required to elaborate more on recruiting the participants and state the critical limitations of data representativeness and generalizability.

We have improved the “Survey design and implementation” subsection by elaborating more on the recruitment of the participants. We have also stated the limitations of data representativeness and generalizability in both the Discussion (limitations paragraph) and Conclusions and future work.

3.The literature review has been poorly written, and it does not support the core idea and novelty of the paper.

We have improved the literature review. The revised review now supports the core idea and novelty by clearly situating our work within consumer-behaviour and DR literatures. We believe these changes fully address the reviewer’s concern.

4.The validation process of the questionnaire, such as the pilot testing, should be reported, and the reliability metrics should be provided for the Likert process.

The reliability of the questionnaire has been covered through Cronbach’s alpha, while its validity has been covered through factor analysis.

5.The Sankey diagram illustrated in Fig. 1 is not consistent with the segmentation approach employed for the Spanish and Austrian pilot sizes.

No such discrepancy has been detected.

6.The analyses require statistical tests like ANOVA and chi-square to validate the importance of the differences observed between the derived consumer segments.

The questions variance shall cover this part, considering the limited data size.

7.The concept persona needs a more structured definition, while its derivation from the segment data should be clearly explained.

All tables with the analysis of questions per pilot and segment have been added in the Appendix for further exploration purposes.

8.A Conclusion section needs to be added.

A Conclusion part has been added.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

I have no further comments.

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have addressed all my concerns. I have no further comments.

Reviewer 3 Report

Comments and Suggestions for Authors

All comments have been addressed and the paper could be accepted