How do Consumers Perceive Electric Vehicles? A Comparison of German Consumer Groups

Abstract

Electric vehicles (EVs) are currently being discussed as a promising means to increase the energy efficiency and sustainability of today's transport systems. While technological progress and cost reduction are certainly crucial topics for their successful diffusion, consumer acceptance is another issue that warrants further analysis. Based on a large online survey (N = 969), we compared four consumer groups which differ in their likelihood to purchase an EV with regard to their socio-demographic characteristics, their willingness to pay (WTP) and their perceptions of EVs. The findings indicate that early users in Germany are most likely to be middle-aged men living with their families in a multi-vehicle household who have a higher WTP for an EV. Perceived compatibility of an EV with personal needs seems to be the most influential factor on the stated willingness to purchase an EV. With regard to the promotion of EVs, strengthening their environmental advantages and providing financial incentives for purchase are rated as important measures by a majority of the sample, while performance characteristics which are comparable to conventional vehicles seem to be less important for most participants. Based on the data analyses, we provide recommendations for measures regarding the further development and promotion of EVs.

Key Words:

• Acceptance
• diffusion of innovation
• early adopters
• survey
• preferences

Acknowledgements

For valuable comments on the survey conducted in this study, the authors thank Jana Hoffmann and Bettina Ryf. They specially thank Raphael Agosti and Mareike Popp who had a crucial role in realizing this survey and who supported data collection and analysis. The authors would also like to thank three anonymous reviewers for their valuable comments which helped to improve this paper.

Funding

The authors gratefully acknowledge funding by the German Federal Ministry of Education and Research (BMBF, Grant Number: 13N10597-13N10600).

Notes

1. Until August 2013, the overall number of EVs in Germany has now risen to slightly above 10,000; however, the overall market share is still low (<0.02 %) (KBA, 2013).

2. For the variable relative advantage, the meaning of the scale values differed: 1 = very low/not sufficient and 7 = very high/absolutely sufficient.

3. In the analyses, we use—analogous to the other model variables—the positive term ease of use instead of complexity and code items so that higher values for this variable indicate an easier use of EVs and are assumed to correlate with the target variable.

4. In total, the survey was accessed 3497 times during the data collection period. 2301 people dropped out before completing the survey. As the majority dropped out on the first page, i.e. before the actual start of the survey, the completion rate of 34.2% is regarded as satisfactory.

5. A χ²-test shows that the frequency distribution significantly differs from a random distribution (χ² = 56.0; df = 3; p < .01).

6. If age is analysed as a continuous variable and included in a one-way ANOVA, the non-interested group and the affine group are significantly younger than the group with purchase intention and the group of actual users (F = 7.0; p < .01; df = 3/965; post hoc Scheffé-test, p < .05).

7. A one-way ANOVA shows that these two groups differ significantly (F = 3.0; p < .05; df = 3/965; post hoc Scheffé-test, p < .05).

8. A χ²-test shows that the frequency distribution significantly differs from a random distribution (χ²= 13.0; df = 6; p < .05).

9. F = 23.9; p < .01; df = 3/733.

10. Significant differences among all the groups concerning the intention to use and purchase an EV could be observed in a post hoc analysis, except for the groups of users and consumers with concrete purchase intention.

11. The scale measuring compatibility shows multicollinearity with other variables in the model for the overall sample as well as within the group of consumers not interested in EVs. Compatibility is especially correlated to social norm (r = .674) and relative advantages (r = .561), i.e. estimations may be less stable then it would be desirable.