Hypothesis Testing

HI (Null): p (customers preferring high mileage) > p (customers preferring high yearly fuel cost savings)

  • • Null hypothesis rejected as P < °° 0.05
  • • Proposing alternative hypothesis: customers prefer high fuel cost savings over high mileage as a cost incentive
  • • Thus, high yearly fuel cost savings has a positive effect on PHEV adoption likelihood

H2 (Null): p (customers preferring charging infrastructure) = p (customers preferring lower charging time) = p (customers preferring longer driving range)

  • • Null hypothesis rejected as P < °° 0.05
  • • Proposing alternative hypothesis: customers preference regarding charging station infrastructure, charging time and driving range differs
  • • Driving range has the most influencing effect
  • • Thus, driving range is the most influencing factor on PHEV adoption likelihood (high p value)

H3 (Null): p (customers preferring higher charging station infra) > p (customers preferring lower charging time)

  • • Null hypothesis rejected as P < °° 0.05
  • • Proposing alternative hypothesis: customers prefer lower charging time over high charging station infrastructure
  • • Thus, high charging time has a negative effect on PHEV adoption likelihood

H4 (Null): p (customers preferring tax subsidies) > p (customers preferring lower vehicle price)

  • • Null hypothesis rejected as P < °° 0.05
  • • Proposing alternative hypothesis: customers prefer lower vehicle price over tax subsidies as a financial incentive
  • • Thus, high vehicle price has a negative effect on PHEV adoption likelihood (Figure 10.3)

164 Amit Dutta et al.

Sampling Framework

Figure 10.3 Sampling Framework.

PHEV Attribute Rankings - Dubai versus Sydney

  • • Similar preferences were shown amongst the sample respondents across the studied geographies of Dubai and Sydney
  • • Driving range is the most important factor affecting PHEV adoption likelihood
  • • Tax subsidy was the least important factor affecting PHEV adoption likelihood
  • • Charging Infrastructure had a higher attribute ranking amongst the respondents in Sydney than in Dubai
  • • Vehicle price ranked higher in Dubai as compared to Sydney as per the sample survey
  • • Annual cost savings was the second most important factor across both the cities

Factor Analysis

Factor analysis was used to identify the latent variables and for the purpose of grouping similar variables into dimensions and reducing the factors into a fewer number of dimensions. The same helped in simplifying the data and reduce the number of variables. For the data analysis of the collected data, the statistical software statistical program of social science (SPSS) version 22 for Windows was used in performing the necessary calculations so as to obtain accurate data and thus minimise any data processing errors.

  • • A KMO score of 0.6 suggested suitability of the collected data for factor analysis
  • • Further, a significance level of 0.000 (Bartlett’s test of sphericity) suggested enough correlation between the variables for conducting the factor analysis
  • • There are four factors with eigenvalues > 1 and thus were retained for the purpose of the analysis
  • • 70.73% of variances were accounted for by the four extracted factors

Cluster Analysis

Cluster analysis refers to the technique of classifying objects or cases into comparable homogeneous groups known as ‘clusters’. The most important attribute of this analysis is the lack of prior information about the group membership. The primary usage of this aggregation technique for data analysis is to segment consumers based on the benefits sought from the purchase of the Prius hybrid vehicle and used to identify homogeneous groups of PHEV buyers.

The same involved formulating the problem, selecting the measure of the distance and the clustering procedure, determining the number of clusters, interpretating the cluster profiles and finally concluding with the assessment of the validity of the clustering process based on the data collected through the sample survey indicating adoption intention for the Prius hybrid. The distance between the cluster centres indicated the degree of separation of the individual pair of clusters of Prius hybrid buyers. The analysis helped in recommending the segments of customers favourable for PHEV adoption and indicated the distinct clusters which shall be desirable to the marketing department of Toyota for the formulation and implementation of specific segment-oriented marketing strategies so as to target Prius hybrid customers more effectively and increase penetration levels.

К-means cluster procedure (SPSS) was employed to group customers with similar preferential behaviours.

  • • Primary consumer clusters for PHEVs
  • • Cluster 1 has the highest number of customer groupings
  • • Cluster 3 has the least number of customer groupings
  • • Focus on relative advantage, compatibility, trialability, observability so as to increase awareness levels of PHEVs in order to increase groupings in the ideal cluster (Cluster 2)
< Prev   CONTENTS   Source   Next >