Public Perceptions of Electric Vehicle Adoption in Kuwait: The Role of Low Electricity Tariffs, Charging Constraints, and Fire-Safety Concerns

Abstract

This study examines public perceptions of electric vehicle (EV) adoption in Kuwait, a high-income petroleum-dependent country characterized by highly subsidized electricity, low fuel prices, limited charging infrastructure, and extreme climatic conditions. Using a structured survey of 1753 licensed drivers, the study evaluates how economic incentives, practical constraints, environmental perceptions, technological confidence, and safety concerns shape expectations of future EV diffusion. Descriptive statistics, principal component analysis, and ordinal logistic regression were used to examine the factors associated with respondents’ expectation of widespread EV adoption in Kuwait over the next ten years. The regression results show that low-tariff/delayed-bill perception was the strongest positive predictor of expected EV adoption, indicating that Kuwait’s low-cost electricity environment may strengthen expectations of EV diffusion. However, the findings also demonstrate that electricity tariffs alone do not explain public expectations. EV performance perception, environmental benefit perception, workplace charging, battery warranty, prior passenger experience in an EV, and higher weekly fuel expenditure were also positively associated with stronger expectations of EV adoption. In contrast, perceived complexity was negatively associated with expected adoption, highlighting the importance of consumer familiarity and ease of use. Safety-related perceptions, particularly concerns regarding EV fire-extinguishing difficulty and lower perceived safety compared with conventional vehicles, were also significant, suggesting that fire safety remains a salient issue in the Kuwaiti context. The findings contribute to the literature on sustainable transportation adoption in petroleum-based economies and extreme climates by showing that EV diffusion depends on a combination of economic, infrastructural, technological, environmental, and safety-related factors. Policy efforts in Kuwait should therefore combine charging-infrastructure development, workplace charging expansion, consumer education, battery-warranty assurance, and EV-specific safety and emergency-response measures.

1. Introduction

In 2022, Kuwait’s greenhouse gas (GHG) emissions were estimated at 136 MtCO₂e, equivalent to approximately 32 tCO₂e per capita [1]. This per capita level ranked Kuwait as the seventh-highest emitter globally and was nearly five times higher than the European Union average [1]. According to the Kuwait Environment Public Authority [2], the transport sector was the third highest GHG-emitting sector, with 16% of total GHG emissions being attributed to road transportation [3]. Furthermore, from 2005 to 2014, the number of vehicles surged by 162%, accompanied by a 61% rise in the number of private cars [3].

The increased usage of electric vehicles (EVs) as cleaner and greener technologies has been identified as an effective strategy for becoming more energy efficient and abating GHGs. It may therefore contribute significantly to the desired shift toward sustainable mobility [4,5,6,7]. Regionally, the State of Qatar has set a target for EVs to represent 10% of all new vehicle sales by 2030, aligning with the Qatar National Vision 2030, which emphasizes economic growth, social prosperity, and environmental management as key pillars of sustainable development [8].

The widespread use of EVs is mainly driven by consumers’ attitudes, as they decide whether to purchase newly introduced technology such as EVs [6,9,10]. Several difficulties may negatively influence consumers’ attitudes toward EVs and consequently reduce their widespread use. These include vehicle price, battery replacement cost, recharging time, driving range, availability of infrastructure, and EV safety—particularly concerns over battery fires.

In Kuwait, a country traditionally reliant on fossil fuels, the relatively low electricity tariff may present a unique opportunity to accelerate the adoption of EVs. With electricity prices significantly subsidized by the government, the cost of operating an EV could be substantially lower than that of operating conventional internal combustion engine vehicles (ICEVs). This economic advantage can serve as a powerful incentive for consumers to transition to electric mobility, thereby contributing to reduced GHG emissions and improved urban air quality. However, the risk of battery fires in EVs due to commonly high temperatures may cause concern and hinder their widespread adoption.

The move toward EVs in Kuwait is shaped by conditions that differ from many other countries. Kuwait is a high-income, oil-exporting country with one of the lowest household electricity tariffs in the world. On the other hand, the dominant hot climate might trigger battery fire hazards [11]. These are the two main factors likely to influence how people think about EVs ownership. For this reason, the present study examines attitudes toward EVs among drivers in Kuwait. The survey was employed to explore views on cost, convenience, environmental impact, performance, and day-to-day use. It also considers whether low electricity tariffs and delays in bill collection may encourage drivers to choose EVs instead of ICEVs. At the same time, the study looks at concerns about EV battery fires, particularly the difficulty of extinguishing them in hot weather, and how such concerns may discourage adoption.

The analysis uses responses from 1753 participants, providing a broad picture of public opinion across different groups in society. By combining issues of cost, safety and public perception, the study offers evidence that may be useful for transport and energy policy in Kuwait. More generally, the findings help explain how pricing systems and climate-related risks shape EV acceptance in oil-dependent economies.

2. Background and Literature Review

Most studies on EVs, such as those focusing on attitudes toward EVs environmental aspects and best practices for the implementation of governmental policies to increase EV adoption, focus on industrialized nations (often referred to as the Global North), with some attention given to rapidly developing countries like those in the Gulf Cooperation Council. However, limited research has examined policy implementation in these areas. For instance, Al-Buenain et al. [8] conducted a study Qatar and found that although the country generated all its electricity from natural gas, substantial carbon reduction was still achieved through electromobility. Their research highlights the importance of practical incentives and subsidies from the government to shift away from gasoline dependence and toward EV adoption. In Bahrain, Shareeda et al. [12] concluded that government financial incentives, such as tax breaks or price reductions, were needed to make EVs competitively priced in comparison with ICEVs. Additionally, they emphasized the importance of infrastructure, particularly a fast-charging network, to enable rapid EV adoption. In the United Arab Emirates, Kiani [13] recommended that EVs manufacturers should be encouraged by creating a supportive investment environment that facilitated product launches, marketing, public awareness. As population growth accelerates in developing countries, the concept of sustainable mobility is gaining increasing attention from academics, leaders, managers, and governments worldwide. The rise of combustion engine use in transportation led to a notable increase in GHG emissions from traffic [7,13,14,15,16,17,18]. Consequently, measures to address the current environmental crisis should focus on sustainable mobility and environmentally friendly technologies, such as EVs with low or zero GHG emissions. To encourage the public to adopt eco-friendly products like EVs, it is crucial for governments to implement policies, directives, and support mechanisms that promote greener modes of transport, fostering environmentally responsible behaviors and advancing sustainable mobility [19,20,21,22,23,24,25].

Previous research on consumer demand for EVs has consistently shown that evaluations of their functional performance significantly influence the likelihood of adoption. Factors such as fuel cost have also been identified as key influences on preferences for EVs [26]. Although the significance of EVs studies lies in climate change mitigation and addressing energy scarcity, there are still research gaps in evaluating EV adoption in different environments and regions, such as Kuwait, where the relatively low energy tariff may accelerate the rate of EV adoption.

The State of Kuwait is one of the hottest countries in the world; temperatures of 54 °C have been officially recorded [27,28,29,30]. This increases the risk of EVs fires, as high-temperature conditions are a known contributing factor, particularly when the vehicle is stationary [31]. Extinguishing a fire in an EV is more challenging than in an ICEV [32], as standard fire suppression techniques may not be effective. Successfully extinguishing EV fires often requires significantly larger quantities of water compared to fires involving petrol or diesel vehicles. Moreover, there is a risk that the batteries may reignite even after the fire has been put out [31,32]. This could be a main barrier to the widespread use of EVs in hot countries.

Another main barrier to the adoption of EV is their initial price, which is often twice the price of their equivalent ICEVs. The Genesis G80 was selected as the base model of comparisons in this study as both electric and conventional versions of this model are available on the Kuwaiti market. In Kuwait, the initial cost of the EV G80 Genesis is US$94,000, which is nearly double the price of the equivalent ICEVs, based on 2023 prices. A study on the Kuwaiti market attributed this to the cost of large EV batteries [33]. However, the maintenance costs of EVs are significantly lower than those of ICEVs, especially for the first 150,000 km. Some parts of ICEVs such as engine and transmission fluids need to be regularly replaced, while these parts do not exist in EVs [34]. In addition, even though the cost of EV battery replacement is exorbitant, the battery warranty that extends to 7 years or 150,000 km may cover this cost, and the running costs of EVs, which are the main continuous cost of vehicle operation, are much lower than the running costs of ICEVs.

Some studies conducted in Kuwait and neighboring regional countries are presented in

Table 1. A comparative review of EV-related studies across Kuwait and surrounding regional countries.

Categories of EV Features Examined
Cost of EV ownership	Al-Buenain et al. [8] (Qatar), Shetty et al. [22] (India and Sri Lanka), Shareeda et al. [12] (Bahrain), Eneizan [19] (Jordan), Haider et al. [20] (India), Abu-Alkeir et al. [21] (Jordan), Ottesen et al. [33] (Kuwait), Khandakar et al. [34] (Qatar), Ottesen et al. [35] (Kuwait)
EV Infrastructure	Shetty et al. [22] (India and Sri Lanka), Shareeda et al. [12] (Bahrain), Kiani [13] (UAE), Haider et al. [20] (India), Ottesen et al. [33] (Kuwait), Khandakar et al. [34] (Qatar), Ottesen et al. [32] (Kuwait)
Governmental policies and incentive mechanisms	Al-Buenain et al. [8] (Qatar), Shetty et al. [22] (India and Sri Lanka), Shareeda et al. [12] (Bahrain), Kiani [13] (UAE), Haider et al. [20] (India), Ottesen et al. [33] (Kuwait), Khandakar et al. [34] (Qatar), Ottesen et al. [35] (Kuwait)
EV driving range	Shetty et al. [22] (India and Sri Lanka), Shareeda et al. [12] (Bahrain), Haider et al. [20] (India), Ottesen et al. [33] (Kuwait), Khandakar et al. [31] (Qatar), Ottesen et al. [35] (Kuwait)
Environmental considerations	Al-Buenain et al. [8] (Qatar), [22] (India and Sri Lanka), [13] (UAE), Haider et al. [20] (India), Ottesen et al. [33] (Kuwait), Ottesen et al. [32] (Kuwait)
Battery life and performance	Shetty et al. [22] (India and Sri Lanka), Shareeda et al. [12] (Bahrain), Kiani [13] (UAE), Haider et al. [20] (India), Ottesen et al. [33] (Kuwait), Khandakar et al. [34] (Qatar), Ottesen et al. [35] (Kuwait)
Levels of public awareness	Shareeda et al. [12] (Bahrain), Eneizan [19] (Jordan), Haider et al. [20] (India), Khandakar et al. [34] (Qatar), Ottesen et al. [35] (Kuwait)
Behavior of consumers	Eneizan [19] (Jordan), Abu-Alkeir et al. [21] (Jordan), Ottesen et al. [35] (Kuwait)

, which examines several key dimensions, including the total cost of EV ownership compared with that of ICEVs; the availability of EV infrastructure and public charging networks; governmental policies and incentive frameworks; driving range per charge and issues related to range anxiety; sustainability and environmental considerations; battery lifespan and performance; and consumer awareness, education, and behavior associated with EV adoption. There is a scarcity of literature in these countries exploring public concerns regarding local electricity tariffs, EV fire risks, and fire suppression methods.

EVs consume electricity, and electric tariff varies from country to country. In Kuwait, the electricity tariff is only US$0.0065 per kWh for household and US$0.01625 per kWh for businesses. This includes all the components of the electricity bill such as the cost of power and distribution. These tariffs are very low compared to the average global price of electricity, which is US$0.169 per kWh for households and US$0.190 per kWh for businesses. The battery capacity of an EV G80 Genesis is 87.2 kWh, giving a range of 520 km. Based on the electricity tariffs in Kuwait, the fuel costs of an EV G80 Genesis would be US$0.273/100 km with business power rates. On the other hand, the ICEV G80 Genesis has a consumption of 13.3 km/L. With the current Kuwait petrol price of US$0.28/L for Premium 91, the running costs are US$2.11/100 km Premium 91. It is obvious that EVs are much cheaper to run than ICEVs, basing the comparison on a business electricity tariff and Super 95 petrol prices.

Table 2. Estimation of fuel cost for G80 Genesis (EV and ICEV).

Fuel costs for EV G80 Genesis

Based on electricity business tariff $\mathrm{T}\mathrm{a}\mathrm{r}\mathrm{i}\mathrm{f}\mathrm{f} \left({\displaystyle \frac{\$}{\mathrm{k}\mathrm{W}\mathrm{h}}}\right)\times {\displaystyle \frac{\mathrm{B}\mathrm{a}\mathrm{t}\mathrm{t}\mathrm{e}\mathrm{r}\mathrm{y}\mathrm{s}\mathrm{i}\mathrm{z}\mathrm{e} \left(\mathrm{k}\mathrm{W}\mathrm{h}\right)}{\mathrm{R}\mathrm{a}\mathrm{n}\mathrm{g}\mathrm{e} \left(\mathrm{k}\mathrm{m}\right)}}$ $0.01625{\displaystyle \frac{\$}{\mathrm{k}\mathrm{W}\mathrm{h}}}\times {\displaystyle \frac{87.2\mathrm{k}\mathrm{W}\mathrm{h}}{520\mathrm{k}\mathrm{m}}}=0.00272{\displaystyle \frac{\$}{\mathrm{k}\mathrm{m}}}=0.2725{\displaystyle \frac{\$}{100\mathrm{k}\mathrm{m}}}$

Fuel costs for ICEV G80 Genesis

Based on Super 95 petrol price ${\displaystyle \frac{\mathrm{F}\mathrm{u}\mathrm{e}\mathrm{l}\mathrm{p}\mathrm{r}\mathrm{i}\mathrm{c}\mathrm{e} \left(\frac{\$}{\mathrm{L}}\right)}{\mathrm{F}\mathrm{u}\mathrm{e}\mathrm{l}\mathrm{c}\mathrm{o}\mathrm{n}\mathrm{s}\mathrm{u}\mathrm{m}\mathrm{p}\mathrm{t}\mathrm{i}\mathrm{o}\mathrm{n} \left(\frac{\mathrm{k}\mathrm{m}}{\mathrm{L}}\right)}}$ ${\displaystyle \frac{0.34\frac{\$}{\mathrm{L}}}{13.3\frac{\mathrm{k}\mathrm{m}}{\mathrm{L}}}}=0.0256{\displaystyle \frac{\$}{\mathrm{k}\mathrm{m}}}=2.56{\displaystyle \frac{\$}{100\mathrm{k}\mathrm{m}}}$

illustrates estimations that compare the fuel cost of an EV and ICEV G80 Genesis.

The collection of payments for energy bills in Kuwait is entirely entrusted to government agencies, represented by the Ministry of Electricity and Water. Although electricity and water prices are relatively low, the government finds it very difficult to collect these payments. In 2005, this led the government, in agreement with Parliament, to pledge to pay US$6400 toward the electricity bills of each household, with the condition that the remainder could be settled in comfortable installments [36]. Since then, most citizens have lagged behind with their energy bill payments, resulting in large arrears. The question here is whether the government’s inefficiency in enforcing the payment of energy bills might encourage Kuwaiti citizens to purchase EVs, regarding electricity as effectively free or half price, and people may also hope for a repeat of the government’s 2005 initiative and a renewed pledge to pay another US$6400 per household toward electricity bills.

This article investigates the underexplored relationship between electricity pricing and public attitudes toward EV adoption in Kuwait, with a particular focus on how relatively low energy tariffs may serve as a catalyst for increased EV penetration. By also addressing the often-overlooked challenge of managing EV fire risks, the study provides a holistic view of both economic and safety-related barriers to EV acceptance. This research fills a gap in the literature by providing empirical insights specific to the Gulf region, where energy subsidies and extreme climate conditions create a unique context influencing the potential for widespread EV adoption. This work contributes to the broader field of sustainable transportation by highlighting how tailored policy instruments, such as tariff restructuring, can influence consumer behavior and accelerate the transition to cleaner mobility in oil-rich economies.

3. Materials and Methods

The purpose of this study was to conduct a survey of road users in Kuwait to explore their attitudes toward EVs. Questionnaires written in Arabic were distributed electronically, seeking participation from drivers of personal vehicles in Kuwait. An online survey was selected as the principal data collection method. The use of internet-based surveys has seen rapid growth over the past decades and is now a widely applied method of data collection in academic research [37]. The swift adoption of online surveys is attributed to their comparative advantage over the traditional methods of postal, telephone and in-person sampling. Sills and Song [38] highlighted several benefits, including reduced time for delivery, response, data entry, and data cleaning, as well as lower transaction costs. The electronic version of the questionnaire was created through Microsoft Forms and was made available online. Participation in the survey was sought through several social media platforms such as WhatsApp (Version 24.20.78).

The questionnaires were distributed electronically between October and December 2024. They were designed to avoid redundant questions, and extra care was taken to maintain participant anonymity and the confidentiality of their responses by not collecting any identifying information. Additionally, the data were analyzed and presented in aggregate form to ensure that individual participants could not be identified. The aim of the study was to examine whether different respondent characteristics influenced their perspectives on the adoption of EVs. Each question was rated on a scale of 1 to 5, where 1 represented ‘strongly disagree’ and 5 represented ‘strongly agree’. After the survey was completed, response quality was ensured by removing invalid or incomplete questionnaires.

A questionnaire was considered invalid if completed in less than two minutes, and participation was restricted to individuals holding a valid driving license. A hypothesis was proposed that the electricity tariff in Kuwait would positively influence EV adoption, while the region’s extreme heat and concerns about EV battery fires would negatively impact the widespread use of EVs in Kuwait.

The survey questions were developed and divided into five parts: (1) demographic information such as gender, age, level of education, etc.; (2) household economics and housing characteristics; (3) electricity bill payment history and participation in government initiatives; (4) perceptions and attitudes toward EVs; and (5) perceptions of EV fire suppression.

Table 3. Perceptions and attitudes toward EVs and EV fires.

Question Statements Related to Perceptions and Attitudes Toward EVs

I know a lot about the modern types of EVs recently launched in the automotive market Harmful emissions from ICEVs can be reduced by using EVs EVs are relatively more expensive to purchase, but their costs can be offset through cheaper electricity usage compared to petrol/diesel fuel expenses An EV with a 200 km daily range meets all my commuting needs I would value the feature of charging my EV at home I would value the feature of charging my EV at work EVs offer sufficiently high performance I could seriously consider purchasing an EV if its battery warranty is 10 years or more It is easy for me to find (public or private) locations to charge an EV EVs are less reliable than ICEVs Using EVs would be complicated Due to low electricity tariffs and the government’s delay in collecting electricity bills, the adoption of EVs could increase

Question Statement Related to Perceptions of EV Fire Suppression

It is very difficult to extinguish an EV fire The difficulty of extinguishing an EV fire, especially in hot weather, would deter me from using an EV I would feel relatively less safe using an EV compared to an ICEV

I expect widespread adoption of EVs in Kuwait over the next ten years

shows the two sets of statements for the 4th and 5th sections of the questionnaire.

3.1. Sampling Procedures and Size

An electronic survey was distributed among 2000 road users in Kuwait. A total of 1753 valid responses were collected; the remaining responses were either incomplete or submitted after the survey period had ended. Hence, the final sample size comprised 1753 responses.

The study was initially designed with a stratified random sampling method, as suggested by Valliant et al. [39], to account for Kuwait’s demographic diversity and ensure robust, representative insights. However, in practice, data collection was carried out through an online survey. Therefore, it is more accurate to state that the sample used was a stratified convenience sample, rather than a fully probabilistic stratified random sample. To improve the representativeness of the sample, participants’ characteristics, such as age, gender, and nationality, were tracked during data collection. The design integrated proportional allocation across strata and adjusted for potential intra-stratum variability using a Design Effect Multiplier. The target population of interest consisted of nearly two million individuals aged 18 and older who hold a driving license in Kuwait, segmented by age, gender, and nationality (Kuwaiti vs. non-Kuwaiti). Specifically, the population was divided into 24 strata (six age groups, two genders, and two nationalities) to capture subgroup-specific dynamics.

For an appropriate sample size calculation, the base sample size was derived for a 95% confidence level and a 5% margin of error using the following Yamane [40] formula:

$$\begin{gathered} n_{base}={\displaystyle \frac{N}{1+N\left(e^2\right)}} \\ ={\displaystyle \frac{\mathrm{2,000,000}}{1+\mathrm{2,000,000}\left({0.05}^2\right)}} \\ =400 \end{gathered}$$

(1)

Given that:

• n_base = desired sample size
• e = margin of errors (0.05) at 95% level of confidence
• N = total population

To enhance reliability, the sample size was adjusted for design effect (Deff), a conservative multiplier accounting for potential intra-stratum variability [38]. While stratification typically improves precision, a Deff of 2.0 was applied to safeguard against unmodeled heterogeneity using the following formula taken from Valliant et al. [38]:

$$\begin{gathered} n_{adjusted}=n_{base}\times Deff \\ =400\times 2.0=800 \end{gathered}$$

(2)

Given that:

• Deff = design effect

The achieved sample size of 1753 significantly surpassed the target, reducing the effective margin of error to 2.3% (calculated retroactively using Cochran’s formula), thereby increasing result precision [41]. This large sample ensures robust subgroup analyses and mitigates risks of Type II errors.

The Statistical Package for the Social Sciences (SPSS, Version 28) was used for descriptive statistics, reliability analysis, group-comparison tests, and principal component analysis, while the R programming environment (RStudio, Version 2024.04.2+764) was used to conduct the ordinal logistic regression analysis. The demography characteristics and attitude toward EV of the respondents were presented in descriptive statistics including frequency, percentage, mean and standard deviation. Internal consistency of the measurement scale was checked with Cronbach’s alpha coefficient. Several inferential statistical tests were used to examine differences in demographic characteristics. An independent samples t-test compared mean perception scores between groups within dichotomous categories (gender and nationality). A one-way ANOVA assessed differences in mean perception across age groups, and where significant differences were found, a Scheffé post hoc test was applied to identify which specific groups differed. The alpha level for statistical significance was set at p < 0.05.

Method of Principal Component Analysis (PCA) was used to reduce the dimensionality of the measures and examine the fundamental structure of consumer attitude toward EVs. To check whether the data were appropriate for factor analysis, the Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy and Bartlett’s test of sphericity were applied. All extracted components needed to have an eigenvalue greater than 1 (Kaiser criterion) and the rotated solution was obtained through an orthogonal rotation procedure (Varimax rotation). Loadings above 0.50 were used to interpret the factor. PCA was applied because the research is exploratory research in nature, exploring the underlying dimensions that influence consumer attitude, rather than confirming the hypothesized factor model in measurement scales. The flowchart in Figure 1 illustrates the methods adopted in the study.

3.2. Ordinal Logistic Regression Analysis

To extend the analysis beyond descriptive statistics and exploratory techniques, an ordinal logistic regression analysis was conducted to identify the factors independently associated with respondents’ expectations of future EV adoption in Kuwait. While descriptive statistics, group-comparison tests, and principal component analysis provide important insights into general response patterns and underlying perception dimensions, they do not estimate the adjusted relationship between individual predictors and the outcome while controlling for other relevant factors. Therefore, ordinal logistic regression was used to examine how sociodemographic characteristics, household and vehicle-related factors, economic perceptions, practical considerations, technological confidence, environmental awareness, and safety concerns were associated with expectations of widespread EV adoption [42].

The dependent variable was respondents’ level of agreement with the statement: “I expect widespread adoption of EVs in Kuwait over the next ten years.” This item was selected as the outcome variable because it most directly represents the main focus of the study, namely public expectations regarding future EV diffusion in Kuwait. Unlike items measuring specific attitudes, barriers, or incentives, this statement captures respondents’ overall perception of whether EVs are likely to become widely adopted in the national context. It also avoids treating conditional or mechanism-specific items, such as low electricity tariffs or battery warranty, as outcomes. Instead, these variables were included as explanatory factors to assess whether they were independently associated with expectations of future EV adoption. Accordingly, the selected dependent variable provided a conceptually appropriate outcome for examining how economic, practical, technological, environmental, and safety-related perceptions contribute to expectations of EV market diffusion.

Ordinal logistic regression was selected because the dependent variable was measured on an ordered five-point Likert scale [42]. Responses were coded from 1 to 5, where 1 represented “strongly disagree” and 5 represented “strongly agree.” Higher values therefore indicated stronger expectations of widespread EV adoption in Kuwait over the next ten years. The model was estimated under the proportional odds framework, which is appropriate when the response categories are ordinal and the objective is to model the cumulative odds of being at or below a given response category [42].

The general form of the ordinal logistic regression model can be expressed as

$$\mathbf{l}\mathbf{o}\mathbf{g}\left({\displaystyle \frac{\mathbf{P}\left({\mathbf{Y}}_{\mathbf{i}}\le \mathbf{j}\right)}{\mathbf{P}\left({\mathbf{Y}}_{\mathbf{i}}>\mathit{j}\right)}}\right)={\mathbf{\alpha }}_{\mathbf{j}}-{\mathbf{\beta }}^{\mathbf{T}}{\mathbf{X}}_{\mathbf{i}},\mathbf{j}=\mathbf{1},\mathbf{2},\mathbf{3},\mathbf{4}$$

where (${\mathrm{Y}}_{\mathrm{i}}$) represents the ordinal response category for respondent ($\mathrm{i}$), ($\mathrm{j}$) denotes the cumulative threshold between response categories, (${\mathsf{\alpha }}_{\mathrm{j}}$) represents the threshold-specific intercept, (${\mathrm{X}}_{\mathrm{i}}$) is the vector of explanatory variables, and (${\mathsf{\beta }}^{\mathrm{T}}$) is the vector of estimated coefficients. The model estimates the adjusted association between each explanatory variable and the odds of reporting a higher level of agreement with expected EV adoption, while controlling for all other variables included in the model.

Three nested model specifications were developed. Model 1 included only sociodemographic, household, and vehicle-related control variables, including gender, age group, nationality, education level, employment status, monthly income, housing situation, number of cars owned by the household, and weekly fuel expenditure. Model 2 extended Model 1 by adding economic and practical EV-related perception variables, including perceived influence of low electricity tariffs and delayed bill collection, perceived operating-cost offset, sufficiency of a 200 km daily driving range, value of home charging, value of workplace charging, perceived charging-location availability, EV performance perception, and willingness to consider an EV if the battery warranty was 10 years or more. Model 3 represented the full model and further incorporated EV knowledge, previous EV driving experience, previous passenger experience in an EV, perceived environmental benefits, reliability concerns, complexity concerns, perceived difficulty of extinguishing EV fires, concern regarding EV fires in hot weather, and perceived safety compared with conventional vehicles.

Categorical variables were included using indicator variables, with reference categories selected to provide meaningful comparisons. The reference groups were male, age 18–25 years, Kuwaiti nationality, high school or less education, public-sector employment, monthly income below $1600, living with a family member, one household car, and weekly fuel cost below $17. Likert-scale explanatory variables were treated as ordered numeric predictors ranging from 1 to 5, where higher values indicated stronger agreement with each statement. Binary EV exposure variables were coded as 0 for “No” and 1 for “Yes.”

The final model was estimated using complete-case analysis. After excluding observations with missing values in the model variables, 1753 responses were included in Model 3. Results were reported as odds ratios (ORs) with 95% confidence intervals. An OR greater than 1 indicates higher odds of reporting stronger agreement with the expectation of widespread EV adoption, whereas an OR below 1 indicates lower odds of stronger agreement. Because the data were cross-sectional and based on a stratified convenience sample, the estimated associations should be interpreted as adjusted relationships rather than causal effects.

4. Results and Discussions

The following sections present the key findings derived from the survey analysis and discuss their implications in relation to the study objectives. The interpretation focuses on respondents’ perceptions and expectations of future EV diffusion in Kuwait. By critically examining the observed patterns, associations, and differences across survey responses, this section provides insight into the economic, infrastructural, environmental, technological, and safety-related factors that may shape public views of future EV adoption.

4.1. Demographic Factors, Household Economics, and Housing Characteristics

Demographic factors, household economics, and housing characteristics significantly influence attitudes toward EVs. For example, age, income, and household size often dictate a person’s willingness and ability to adopt new technologies like EVs. Younger generations, typically more environmentally conscious, may be more inclined to embrace EVs, while older individuals may face more barriers due to concerns about the technology or higher initial costs [43].

4.1.1. Demographic Characteristics of the Respondents

The demographic characteristics of the respondents provide essential insights into the composition of the surveyed population. Examining the variables of gender, age, marital status, nationality, education, employment, and professional level offers a deeper understanding of the potential factors influencing the adoption of EVs in Kuwait. Figure 2 shows the demographic characteristics of the respondents.

In terms of gender distribution, the sample is predominantly male, with 69.0% of respondents identifying as male, while females account for 31.0%. This gender imbalance may affect the study’s findings, as previous research suggests that men are more inclined to accept technological advancements and make automobile-related decisions [44]. A more balanced gender representation in future studies would provide a broader perspective on EV adoption.

Age distribution reveals that younger individuals dominate the sample, with 40.0% falling within the 18–25 age group, followed by 30.7% in the 26–39 age category; the proportion decreases with increasing age. These results indicate that a significant majority of respondents (70.7%) are under 40 years old, a key factor, as younger generations tend to be more receptive to sustainable technologies [45]. However, the lower representation of older respondents may limit the study’s ability to assess financial and infrastructural concerns that influence EV adoption among more established professionals and retirees.

The data on educational qualifications show that 49.1% of respondents hold higher education degrees, which is consistent with the studies by Ottesen et al. [35] in Kuwait and Khandakar et al. [34] in Qatar, where more than half of the participants held bachelor’s degrees. This is followed by 24.4% with diplomas and 18.9% with a high school education. A smaller portion, 5.5%, have qualifications below high school, while 2.2% report having no formal education. With nearly three-quarters of the respondents (73.4%) holding at least a diploma or higher degree, the study captures a highly educated sample, which could influence perceptions of EVs in terms of technological awareness and environmental responsibility.

The employment status of the respondents varies, with the largest segment working in the public sector (45.2%). The oil sector, a crucial component of Kuwait’s economy, accounts for 6.0%, while the private sector is represented by 6.1%. Additionally, retirees constitute 10.0%, students make up 25.6%, and 7.1% fall into other employment categories. The dominance of public sector employees and retirees (55.2%) suggests that government-driven initiatives may significantly influence EV adoption. Furthermore, the strong presence of students may indicate a growing interest in sustainable transportation, although affordability and infrastructure remain key factors in their decision making.

4.1.2. Household Economics and Housing Characteristics

The economic and housing characteristics of the respondents provide valuable insights into their financial stability, living arrangements, and vehicle ownership, which are critical factors influencing the adoption of EVs in Kuwait. These details are summarized in Figure 3.

In terms of total monthly income, the distribution highlights significant variation among respondents. The largest group, 25.6%, earns less than US$1600 per month; this is likely because students represent almost a quarter of the sample, and they receive a monthly allowance of US$640. Approximately 43% of participants in the current study reported a monthly income below US$3300, whereas Ottesen et al. [35] observed a slightly higher proportion, with over half of respondents falling into this lower income bracket in their sample in Kuwait. A smaller proportion of the respondents earn between US$6401 and US$8000 (7.5%), and 9.3% have a monthly income exceeding US$8001. These findings suggest that a substantial portion of the sample (67.7%) earns US$4800 or less, indicating that affordability and financial incentives may play a crucial role in EV adoption.

Regarding the housing situation, 56.8% of respondents live with family, making this the most common housing arrangement. Homeownership accounts for 27.6%, while 15.6% are tenants. The high percentage of individuals living with family suggests that household decision making regarding vehicle purchases may be more collective than individual, potentially impacting the willingness to switch to EVs.

4.2. Electricity Bill Payment Habits, History, and Government Initiative Participation

The analysis of electricity bill payment behaviors and participation in the 2005 government initiative provides significant insights into consumer habits and preferences in Kuwait. The analysis findings are shown in Figure 4, and they offer vital insights into electricity bill payment habits among Kuwaiti citizens, both on a personal level and in terms of the general population’s perception.

4.2.1. Electricity Bill Payment Habits

Regarding personal responsibility for paying electricity bills, the participants were asked to what extent they agreed with the statement ‘I regularly pay the electricity bills for my private residence’. A significant majority of respondents affirmed their active role in this practice. Specifically, 41.0% strongly agreed, and 33.3% agreed, indicating that 74.3% of participants take responsibility for paying their private residence electricity bills. Conversely, only 8.2% either disagreed or strongly disagreed, suggesting that a minority do not handle their electricity bills directly. The chi-squared test result (p = 0.000) indicates a statistically significant association, reinforcing the reliability of this trend. The high agreement over electricity bill payments may be attributed to the 2017 ministerial decree that authorized service disconnection for nonpayment [46].

When considering the general perception of whether Kuwaiti citizens regularly pay their home electricity bills, responses show a similar trend but with a slightly lower level of strong agreement. Here, 28.6% strongly agree, and 33.1% agree, totaling 61.7%. However, neutrality is more pronounced in this category, with 30.5% selecting this option. The disagreement rate remains low, with only 7.8% expressing skepticism about regular bill payments among Kuwaiti citizens. The chi-squared test result (p = 0.000) again confirms statistical significance, indicating that perceptions about bill payment habits are non-random and meaningful.

4.2.2. Participation in the 2005 Government Initiative

Regarding the 2005 government initiative that provided a partial exemption of US$6400 in exchange for structured repayments, 72.4% of respondents reported that a member of their family had benefited from this policy. This high participation rate suggests that the initiative was widely utilized, reflecting either financial relief needs or strategic consumer behavior to take advantage of government support. Conversely, 27.6% stated that they had not benefited from the program, which may indicate financial independence or ineligibility at the time of implementation. The significant uptake of this initiative underscores the impact of governmental and parliamentary interventions on electricity bill payment behaviors.

4.3. Understanding of Modern EV Technologies

Understanding modern EV technologies is crucial for increasing the adoption of EVs worldwide. As advancements in battery technology, charging infrastructure, and energy efficiency continue to evolve, informed consumers and policymakers will be better able to make educated decisions. The availability of private and public charging solutions helps address range anxiety and improves user confidence in EVs. Additionally, awareness of the lower running cost of EVs due to a relatively cheap electricity tariff and slow energy bill collections could encourage more people to transition from fossil fuel-powered vehicles.

4.3.1. Experience with EVs: Driving and Passenger Perspectives

The data in Figure 5 provide insights into the respondents’ prior exposure to EVs, whether through direct driving experience or as passengers. The findings suggest that familiarity with EVs remains relatively limited among the surveyed population, with statistically significant differences (p < 0.01) between those who have had such experiences and those who have not.

Regarding driving experience, only 29.8% of respondents had previously driven an EV, while the majority, 70.2%, had never had this opportunity. This considerable gap suggests that a lack of firsthand driving experience may be a barrier to EV adoption in Kuwait, as consumers who have not tested the technology may be hesitant to transition from ICEVs. In terms of riding experience, 39.8% have been passengers in an EV, whereas 60.2% have never had this experience. Although this percentage is higher than the proportion of those who have driven an EV, it still indicates that direct exposure to EVs remains limited. Passenger experience may contribute to shaping consumer perceptions of comfort, efficiency, and performance, potentially influencing willingness to adopt EVs in the future.

4.3.2. Perceptions and Attitudes Toward EVs

The analysis of perceptions and attitudes toward EVs provides valuable insights into public awareness, perceived benefits, concerns, and potential adoption factors. The statistically significant results (p < 0.01) suggest that these perceptions are not random but might reflect meaningful trends within the surveyed population. Details related to the perceptions and attitudes toward EVs and descriptive statistics of 12 statements are shown in

Table 4. Perceptions and Attitudes Toward EVs and Descriptive Statistics.

Statement		Strongly Disagree	Disagree	Neutral	Agree	Strongly Agree	p Value	Mean	Std. Dev.	Rank
I know a lot about the modern types of electric cars recently launched in the automotive market	n	110	188	474	526	455	p < 0.01	3.59	1.16	7
	%	6.3	10.7	27.0	30.0	26.0
Harmful emissions from traditional cars can be reduced by using electric vehicles	n	66	71	367	638	611	p < 0.01	3.95	1.03	1
	%	3.7	4.1	20.9	36.4	34.9
Electric cars are relatively more expensive to purchase, but their costs can be offset through cheaper electricity usage compared to petrol/diesel fuel expenses	n	81	79	497	640	456	p < 0.01	3.75	1.04	3
	%	4.6	4.5	28.4	36.5	26.0
All my personal commuting needs can be met using an electric car with a daily battery range of up to 200 km	n	124	169	472	588	400	p < 0.01	3.55	1.15	10
	%	7.1	9.7	26.9	33.5	22.8
I would value the feature of charging my electric car at home	n	74	118	391	645	525	p < 0.01	3.82	1.07	2
	%	4.2	6.8	22.3	36.8	29.9
I would value the feature of charging my electric car at work	n	102	145	463	576	467	p < 0.01	3.66	1.13	5
	%	5.8	8.3	26.4	32.9	26.6
Electric cars offer sufficiently high performance	n	87	171	538	574	383	p < 0.01	3.57	1.08	9
	%	5.0	9.8	30.7	32.7	21.8
I could seriously consider purchasing an electric car if its battery warranty is 10 years or more	n	129	164	380	580	500	p < 0.01	3.66	1.19	6
	%	7.4	9.4	21.7	33.1	28.4
It is easy for me to find (public or private) locations to charge an electric car	n	245	294	513	408	293	p < 0.01	3.12	1.27	12
	%	14.0	16.7	29.3	23.3	16.7
Electric cars are less reliable than traditional cars	n	72	111	537	609	424	p < 0.01	3.69	1.04	4
	%	4.1	6.3	30.6	34.7	24.3
Using electric cars would be complicated	n	104	225	588	537	299	p < 0.01	3.40	1.09	11
	%	5.9	12.8	33.5	30.6	17.2
Due to low electricity tariffs and the government’s delay in collecting electricity bills, the adoption of electric cars could increase	n	89	91	622	614	337	p < 0.01	3.58	1.02	8
	%	5.1	5.2	35.5	35	19.2

.

4.3.3. Awareness and Knowledge of EV

A moderate level of awareness about modern EVs was observed, with 56.0% (n = 981) agreeing or strongly agreeing that they were familiar with the latest EV models. However, 27.0% (n = 474) remained neutral, and 17.0% (n = 298) disagreed, indicating that a considerable portion of the population may still lack sufficient knowledge about EVs. This is further supported by a mean score of 3.59 (SD = ±1.16), ranking the statement ‘I know a lot about the modern types of EVs recently launched in the automotive market’ seventh among all statements, as shown in Table 4. The findings suggest that although awareness is growing, more educational initiatives may be needed to enhance public understanding of the latest EV technologies. Meanwhile, Khandakar et al. [34] reported a lack of awareness regarding EVs among the general public in Qatar, but noted high awareness among technical respondents who studied or worked at universities and research centers.

4.3.4. Environmental and Economic Benefits

The statement ‘Harmful emissions from traditional cars can be reduced by using EVs’ has the strongest consensus regarding the environmental benefits of EVs, with 71.3% agreeing or strongly agreeing with this statement. Only 7.8% disagreed, while 20.9% were neutral, indicating broad recognition of the ecological advantages of EVs. This statement has the highest mean score (3.95, SD = ±1.03), suggesting that environmental awareness is a key driver of EV interest. This finding is consistent with Ottesen et al. [35], who observed that nearly half of the participants in their study were willing to pay a premium of 6–20% for environmentally friendly EVs compared to ICEVs in Kuwait.

Regarding the economic aspect, 62.5% agreed or strongly agreed that although EVs are expensive to purchase, their costs can be offset through lower electricity expenses compared to petrol or diesel. Only 9.1% disagreed, while 28.4% remained neutral. The mean score for this perception was 3.75 (SD = ±1.04), ranking it third, reflecting a general but not overwhelming belief in the long-term cost efficiency of EVs. This aligns with estimate by Liu et al. [47] that EVs with a range of less than 200 km require five years to achieve cost parity with comparable ICEVs.

4.3.5. Practicality and Charging Infrastructure

Concerns about the practicality of EVs and the availability of infrastructure were evident. Regarding whether a 200 km daily range is sufficient for personal commuting, 56.3% agreed or strongly agreed, while 16.8% disagreed, and 26.9% remained neutral. The mean score of 3.55 (SD = ±1.15), ranked tenth, suggesting that while EVs may meet the daily travel needs of many, uncertainty remains for a notable portion of respondents. Home charging was valued by 66.7%, with only 11.0% disagreeing, indicating strong demand for private charging solutions. This perception ranked second (Mean = 3.82, SD = 1.07). Workplace charging was slightly less valued, with 59.5% in agreement, ranking fifth (Mean = 3.66, SD = ±1.13). Infrastructure concerns were evident in the responses about finding public or private charging stations, with only 40.0% agreeing or strongly agreeing, while 30.7% were neutral, and 30.7% disagreed. With a mean score of 3.12 (SD = ±1.27), the lowest among all items, this finding underscores a significant barrier to EV adoption, as many respondents perceive charging accessibility as inadequate. Several studies have found that these factors raise concerns among EV drivers during long trips [6,20,33,34,35,48].

4.3.6. Performance, Reliability, and Complexity

Perceptions about EV performance were generally positive, with 54.5% agreeing or strongly agreeing that EVs offer high performance. However, 14.8% disagreed, and 30.7% were neutral, indicating a mixed level of confidence in EV capabilities. The mean score of 3.57 (SD = ±1.08), ranking ninth, suggests moderate acceptance but room for improvement in public perceptions of EV performance.

Reliability concerns were evident, as 59.0% agreed or strongly agreed that EVs are less reliable than ICEVs. Only 10.4% (n = 183) disagreed, while 30.6% were neutral. The high mean score of 3.69 (SD = ±1.04), ranking fourth, indicates that addressing reliability perceptions is crucial for boosting consumer confidence.

Complexity was another concern, with 47.8% agreeing or strongly agreeing that using EVs would be complicated, while 18.7% disagreed, and 33.5% were neutral. The mean score of 3.40 (SD = ±1.09), ranking eleventh, highlights a moderate perception of difficulty, suggesting the need for user-friendly EV interfaces and better consumer education.

4.3.7. Battery Warranty and Government Policies

Battery warranty was identified as a critical factor in EV adoption, with 61.5% agreeing or strongly agreeing that they would seriously consider purchasing an EV if its battery warranty was 10 years or more. Only 16.8% disagreed, while 21.7% remained neutral. With a mean score of 3.66 (SD = ±1.19), ranking sixth, this finding highlights the importance of long-term battery guarantees in consumer decision making. The impact of low electricity tariffs and delayed bill collection on EV adoption was another noteworthy insight. The majority (54.2%) agreed or strongly agreed that such factors could encourage EV adoption, while 10.3% disagreed, and 35.5% remained neutral. This perception ranked eighth (Mean = 3.58, SD = ±1.02), indicating that while electricity costs are a consideration, they may not be the primary motivator for EV adoption. The results highlight key facilitators and barriers to EV adoption in Kuwait. While environmental benefits and economic efficiency are well recognized, concerns about charging infrastructure, reliability, and complexity remain significant deterrents. The strong interest in home and workplace charging solutions indicates that improving private charging infrastructure could drive EV adoption. Moreover, the findings suggest that long battery warranties and financial incentives, such as government subsidies or low electricity costs, could significantly influence consumer willingness to transition to EVs; this aligns closely with previous findings [49]. Given the statistically significant results (p < 0.01), these insights provide a basis for policymakers, automakers, and energy providers to design targeted strategies for promoting EV adoption in Kuwait.

The results of the PCA indicate that two key components explain a substantial proportion of variance in the dataset. The first component accounts for 44.52% of the variance, while the second component contributes an additional 12.39%, bringing the total explained variance to 56.91% after rotation. The eigenvalues suggest that only these two components hold significant explanatory power, as the remaining components have eigenvalues below 1 and contribute minimally to variance explanation. The first factor includes strong loadings from variables related to performance, cost efficiency, range, home/work charging convenience, and environmental benefits. These items collectively represent “Perceived Benefits and Feasibility of EVs”, reflecting consumer views on the advantages and practicality of EV adoption. The second factor is characterized by high loadings on items related to complexity and reliability concerns, such as the belief that EVs are less reliable than ICEVs and may be complicated to use. This factor can be labeled as “Perceived Barriers and Challenges of EVs”, capturing the skepticism and potential obstacles that consumers associate with EVs.

4.3.8. Overall Perceptions and Attitudes Toward EV

The overall mean score across the 12 items is 3.61, indicating a generally positive perception of EVs among respondents. While there is moderate agreement on the benefits and feasibility of EVs, some concerns remain, particularly regarding infrastructure and reliability. The overall standard deviation is (±1.11), suggesting a moderate level of variability in responses. This indicates that while many respondents hold favorable attitudes toward EV adoption, there is some level of disagreement or uncertainty, which may stem from differing experiences, knowledge levels, or access to EV infrastructure.

The results highlight that while most respondents acknowledge the benefits of EVs, factors such as charging convenience, perceived complexity, and reliability concerns continue to shape consumer attitudes. Addressing these barriers through policy incentives, public awareness campaigns, and infrastructure improvements could further strengthen the overall perception and encourage greater adoption of EVs in Kuwait.

4.4. EV Fire Suppression

The analysis of perceptions and descriptive statistics regarding EV fire suppression are shown in

Table 5. Perceptions Toward EV Fire Suppression and Descriptive Statistics.

Statement		Strongly Disagree	Disagree	Neutral	Agree	Strongly Agree	p Value	Mean	Std. Dev.	Rank
It is very difficult to extinguish an EV fire.	n	70	111	557	570	445	p < 0.01	3.69	1.04	2
	%	4.0	6.3	31.8	32.5	25.4
The difficulty of extinguishing an EV fire, especially in hot weather, would deter me from using an EV	n	60	129	588	612	364	p < 0.01	3.62	1.00	3
	%	3.4	7.4	33.5	34.9	20.8
I would feel relatively less safe using an EV compared to an ICEV	n	106	190	512	577	368	p < 0.01	3.52	1.12	4
	%	6.0	10.8	29.2	32.9	21.0
I expect widespread adoption of EVs in Kuwait over the next ten years	n	76	106	457	662	452	p < 0.01	3.75	1.04	1
	%	4.3	6.0	26.1	37.8	25.8

, which reveal concerns about safety and usability, particularly in Kuwait’s hot climate. The results show that while respondents recognize challenges in extinguishing EV fires, their concerns do not overwhelmingly deter them from expecting widespread EV adoption. The statistically significant results (p < 0.01) suggest that these perceptions are meaningful and not random.

4.4.1. Concerns About Fire Suppression Difficulty

A considerable portion of respondents perceive that extinguishing an EV fire is challenging, with 57.9% agreeing or strongly agreeing with this statement ‘It is very difficult to extinguish an EV fire’, while 10.3% disagreed. A significant 31.8% remained neutral, reflecting a mix of certainty and uncertainty about this issue. With a mean score of 3.69 (SD = ±1.04), ranked second, this finding indicates that fire suppression difficulty is a moderate but notable concern in EV adoption. This is in agreement with the study by Kim et al. [50]. Following a major EV fire in South Korea, they documented changing attitudes among drivers. Half of those who lacked EV driving experience reported negative perceptions of the technology. Additionally, 28% of experienced EV drivers voiced reluctance toward future adoption.

4.4.2. Fire Risk as a Deterrent to EV Adoption

When considering whether fire suppression difficulty, especially in hot weather, would deter respondents from using EVs, 55.7% agreed or strongly agreed, while 10.8% disagreed. The neutral category (33.5%) suggests that a significant number of respondents are uncertain or have mixed views. This statement had a mean score of 3.62 (SD = ±1.00), ranking third, confirming that while safety concerns exist, they may not be a decisive barrier for all respondents.

4.4.3. Perceived Safety of EVs Compared to ICEVs

Concerns about EV safety relative to ICEVs are present but less pronounced. The majority, 53.9%, agreed or strongly agreed that they would feel less safe using an EV, while 16.8% disagreed. A large neutral segment (29.2%) indicates that many respondents remain undecided on this issue. With a mean score of 3.52 (SD = ±1.12), ranked fourth, safety concerns are notable but not the dominant factor in shaping EV perceptions. In comparison, a study conducted in Kuwait found that 40% of participants believed EVs were safe, a figure that took the risk of car crashes into consideration [35].

4.4.4. Expectations of EV Adoption in Kuwait

Despite fire suppression and safety concerns, 63.6% agreed or strongly agreed that EVs will see widespread adoption in Kuwait over the next ten years, while 10.3% disagreed. A neutral response was given by 26.1%, indicating some uncertainty about the pace of EV adoption. This statement ‘I expect widespread adoption of EVs in Kuwait over the next ten years’ had the highest mean score of 3.75 (SD = ±1.04), ranking first, suggesting strong optimism about the future of EVs in Kuwait, despite existing concerns.

4.4.5. Overall Perceptions and Attitudes Toward EV Fire Suppression

The overall mean score across the four items is 3.65, indicating that respondents generally perceive fire suppression and safety risks as moderate concerns regarding EVs. While there is some agreement that extinguishing EV fires is difficult and that safety risks exist, these concerns do not appear overwhelmingly negative. The overall standard deviation is (±1.05), reflecting a moderate level of variability in responses. This suggests that while many respondents share concerns about EV fire suppression, there is still a diversity of opinions, likely influenced by varying levels of awareness, personal experiences, and trust in EV technology.

Overall, the findings indicate that fire suppression and safety concerns exist but do not strongly deter optimism about EV adoption. To enhance public confidence, efforts should focus on educating consumers about fire safety measures, improving emergency response capabilities, and highlighting advancements in EV battery technology that reduce fire risks.

4.5. Comparison of Ordinal Logistic Regression Models

To strengthen the interpretation of the survey findings beyond descriptive statistics, three nested ordinal logistic regression models were estimated and compared. The dependent variable was respondents’ expectation of widespread EV adoption in Kuwait over the next ten years, measured on a five-point ordinal scale. Model 1 included only sociodemographic, household, and vehicle-related control variables. Model 2 extended Model 1 by adding economic and practical EV-related perception variables. Model 3 further incorporated EV knowledge, prior EV exposure, environmental perception, reliability and complexity concerns, and EV fire-safety perceptions. The model comparison results are presented in

Table 6. Comparison of ordinal logistic regression models predicting expected EV adoption in Kuwait.

Model	Variables Included	LL(β)	LR χ2 vs. Null	Degree of Freedom	p-Value	McFadden Pseudo R2	AIC	BIC	Nested LR Test
Model 1	Sociodemographic, household, and vehicle-related controls	−2347.22	93.65	29	<0.001	0.020	4760.43	4940.69	—
Model 2	Model 1 + economic and practical EV perception variables	−1921.76	944.56	37	<0.001	0.197	3925.53	4149.48	χ2(8) = 850.91, p < 0.001
Model 3	Model 2 + EV knowledge, prior EV exposure, environmental perception, reliability and complexity concerns, and EV fire-safety perceptions	−1875.72	1036.64	46	<0.001	0.217	3851.44	4124.55	χ2(9) = 92.08, p < 0.001

.

Model 1 was statistically significant overall, with a likelihood-ratio chi-square value of 93.65 (p < 0.001). However, its explanatory power was limited, as indicated by a McFadden pseudo R² of 0.020. This suggests that sociodemographic, household, and vehicle-related characteristics alone provide only a modest improvement over the baseline model, indicating the need to incorporate perception-based variables to better account for respondents’ expectations of future EV adoption.

Model 2 provided a substantial improvement over Model 1 after adding economic and practical EV-related perception variables. The McFadden pseudo R² increased from 0.020 to 0.197, while the AIC decreased from 4760.43 to 3925.53 and the BIC decreased from 4940.69 to 4149.48. The likelihood-ratio improvement test comparing Model 2 with Model 1 was statistically significant (LR χ² = 850.91, df = 8, p < 0.001). This indicates that perceptions related to electricity tariffs, operating-cost advantages, charging convenience, driving-range sufficiency, EV performance, and battery warranty provide substantially greater explanatory power than sociodemographic characteristics alone.

Model 3 produced the best overall fit among the three specifications, as shown in Figure 6. After adding EV knowledge, prior EV exposure, environmental perception, reliability and complexity concerns, and EV fire-safety perceptions, the McFadden pseudo R² increased further to 0.217. Model 3 also produced the lowest AIC and BIC values, at 3851.44 and 4124.55, respectively. The likelihood-ratio improvement test comparing Model 3 with Model 2 was statistically significant (LR χ² = 92.08, df = 9, p < 0.001), confirming that the additional perception, exposure, and safety-related variables contributed meaningful explanatory value.

4.6. Interpretation of the Full Ordinal Logistic Regression Model

Model 3 was selected as the final specification because it provided the best overall fit among the three ordinal logistic regression models. The dependent variable was respondents’ expectation of widespread EV adoption in Kuwait over the next ten years. As shown in

Table 7. Ordered logistic regression results for Model 3 predicting expectations of widespread EV adoption in Kuwait over the next ten years.

Variable	Category	Estimate	Z-Stat	p-Value	Odds Ratio	95% CI (Odds)
						Lower	Upper
Gender	Female	−0.415	−3.51	<0.001	0.66	0.52	0.83
	Male *
Age group	≥50 years	0.084	0.37	1.09	1.09	0.70	1.69
	40–49 years	0.299	1.58	0.113	1.35	0.93	1.95
	26–39 years	0.141	0.97	0.332	1.15	0.87	1.53
	18–25 years *
Nationality	Non-Kuwaiti	0.165	0.90	0.369	1.18	0.82	1.69
	Kuwaiti *
Education	Postgraduate degree	−0.004	−0.02	0.987	1.00	0.66	1.51
	University degree	0.115	0.90	0.369	1.12	0.87	1.44
	Diploma	−0.072	−0.53	0.594	0.93	0.71	1.21
	High school or less *
Employment status	Unemployed	−0.401	−1.70	0.090	0.67	0.42	1.06
	Retired	−0.405	−2.11	0.035	0.67	0.46	0.97
	Student	0.006	0.03	0.975	1.01	0.70	1.44
	Private/oil/family business	−0.298	−1.84	0.065	0.74	0.54	1.02
	Public sector *
Monthly income	> $11,201	0.209	0.60	0.546	1.23	0.63	2.42
	$9601–$11,200	−0.074	−0.19	0.848	0.93	0.44	1.98
	$8001–$9600	−0.305	−1.01	0.312	0.74	0.41	1.33
	$6401–$8000	0.017	0.07	0.947	1.02	0.62	1.66
	$4801–$6400	0.044	0.21	0.832	1.05	0.70	1.57
	$3201–$4800	−0.029	−0.16	0.875	0.97	0.68	1.39
	$1601–$3200	0.028	0.15	0.878	1.03	0.72	1.46
	<$1600 *
Housing situation	Tenant	−0.246	−1.71	0.087	0.78	0.59	1.04
	Homeowner	0.058	0.45	0.653	1.06	0.82	1.36
	Living with a family member *
Number of household cars	Five cars or more	−0.216	−1.31	0.189	0.81	0.58	1.11
	Four cars	−0.230	−1.28	0.201	0.79	0.56	1.13
	Three cars	0.087	0.49	0.623	1.09	0.77	1.54
	Two cars	0.184	1.09	0.277	1.20	0.86	1.68
	One car *
Weekly fuel cost	>$65	0.751	3.65	<0.001	2.12	1.42	3.17
	$49–$64	0.728	3.41	<0.001	2.07	1.36	3.15
	$33–$48	0.438	2.12	0.034	1.55	1.03	2.33
	$17–$32	0.289	1.51	0.132	1.33	0.92	1.94
	< $17 *
Low-tariff/delayed-bill perception	One-point increase **	0.527	8.12	<0.001	1.69	1.49	1.92
Lower operating-cost offset perception	One-point increase **	0.102	1.52	0.128	1.11	0.97	1.26
200 km range sufficiency	One-point increase **	0.109	1.85	0.065	1.12	0.99	1.25
Value home charging	One-point increase **	−0.036	−0.50	0.619	0.96	0.84	1.11
Value workplace charging	One-point increase **	0.197	3.19	0.001	1.22	1.08	1.38
Charging-location availability	One-point increase **	0.007	0.13	0.898	1.01	0.91	1.11
EV performance perception	One-point increase **	0.444	6.10	<0.001	1.56	1.35	1.80
10-year battery warranty consideration	One-point increase **	0.203	3.24	0.001	1.22	1.08	1.38
Knowledge of modern EVs	One-point increase **	0.061	1.22	0.222	1.06	0.96	1.17
Previously drove an EV	Yes	−0.018	−0.12	0.903	0.98	0.74	1.31
	No *
Previously rode in an EV	Yes	0.285	2.16	0.031	1.33	1.03	1.72
	No *
Environmental benefit perception	One-point increase **	0.280	4.47	<0.001	1.32	1.17	1.50
Reliability concern	One-point increase **	−0.027	−0.47	0.637	0.97	0.87	1.09
Complexity concern	One-point increase **	−0.221	−3.83	<0.001	0.80	0.72	0.90
EV fire-extinguishing difficulty	One-point increase **	0.179	2.81	0.005	1.20	1.06	1.36
Fire concern in hot weather prevents use	One-point increase **	0.070	1.06	0.291	1.07	0.94	1.22
Perceived lower EV safety	One-point increase **	0.254	4.69	<0.001	1.29	1.16	1.43

* Reference category. ** Likert-scale predictors were coded from 1 = strongly disagree to 5 = strongly agree; therefore, their odds ratios represent the effect of a one-point increase in agreement.

, the full model identified several statistically significant predictors, with low-tariff/delayed-bill perception showing the strongest positive association with expected EV adoption (OR = 1.69, (p < 0.001)), while complexity concern was negatively associated with expected adoption (OR = 0.80, (p < 0.001)). Therefore, the results should be interpreted as factors associated with higher or lower odds of reporting stronger agreement with expected EV adoption, rather than as determinants of actual EV ownership or observed EV penetration.

Among the sociodemographic variables, gender was significantly associated with expectations of future EV adoption. As shown in Figure 7, female respondents had lower odds of reporting stronger agreement with expected widespread EV adoption compared with male respondents (OR = 0.66, p < 0.001), holding all other variables constant. This indicates that female respondents were less likely than male respondents to expect widespread EV diffusion in Kuwait. Age group, nationality, and education level were not statistically significant in the full model, suggesting that these characteristics did not independently explain expectations of future EV adoption after accounting for economic, practical, technological, and safety-related perceptions.

Employment status showed a limited but relevant association. Retired respondents had significantly lower odds of expecting widespread EV adoption compared with public-sector employees (OR = 0.67, p = 0.035). The private/oil/family business category and unemployed category were marginal but not statistically significant at the 0.05 level. Monthly income was also not statistically significant across all income categories. This suggests that, after controlling for fuel expenditure and EV-related perceptions, income level alone was not a strong independent predictor of expected EV adoption.

Household and vehicle characteristics produced mixed results. Housing situation and number of household vehicles were not statistically significant in the full model. However, weekly fuel expenditure was positively associated with expectations of EV adoption. Compared with respondents spending less than $17 per week on fuel, those spending $17–$32 had higher odds of expecting widespread EV adoption (OR = 1.55, p = 0.034). The association was stronger among respondents spending $49–$64 per week (OR = 2.07, p < 0.001) and those spending more than $65 per week (OR = 2.12, p < 0.001). These findings suggest that respondents facing higher fuel-cost burdens may be more likely to view EVs as a plausible future alternative, particularly in a context where electricity prices are relatively low.

The low-tariff/delayed-bill perception variable was the strongest positive predictor in the model, as shown in Figure 8. Each one-point increase in agreement with the statement that low electricity tariffs and delayed electricity-bill collection could increase EV use was associated with 69% higher odds of reporting stronger agreement with expected widespread EV adoption (OR = 1.69, p < 0.001). This finding supports the argument that Kuwait’s low electricity tariff environment may contribute to favorable expectations regarding EV diffusion. However, because several other perception variables were also significant, the result should not be interpreted as evidence that low electricity tariffs alone are sufficient to drive EV adoption.

Several practical and technology-related perception variables were significantly associated with expected EV adoption. Respondents who perceived EVs as offering sufficiently high performance had higher odds of expecting widespread adoption (OR = 1.56, p < 0.001). Similarly, valuing workplace charging was positively associated with expected adoption (OR = 1.22, p = 0.001), as was willingness to consider an EV if the battery warranty were 10 years or more (OR = 1.22, p = 0.001). These results suggest that expectations of EV diffusion are influenced not only by operating-cost considerations but also by confidence in vehicle performance, battery durability, and charging opportunities outside the home.

In contrast, some practical variables were not statistically significant. Perceived operating-cost offset, 200 km range sufficiency, home charging, and charging-location availability did not independently predict expectations of widespread EV adoption in the full model. Their non-significance does not necessarily imply that these factors are unimportant. Rather, it suggests that once other related factors—such as low-tariff perception, performance perception, workplace charging, battery warranty, and safety-related perceptions—were included, these variables did not provide additional independent explanatory value.

Prior EV exposure showed a meaningful association. Respondents who had previously ridden as passengers in an EV had higher odds of expecting widespread EV adoption (OR = 1.33, p = 0.031). This suggests that even indirect exposure to EV technology may improve familiarity and strengthen expectations of future market diffusion. However, previous EV driving experience was not statistically significant (OR = 0.98, p = 0.903). This may reflect the relatively limited availability of direct EV driving experience among the surveyed population, while passenger exposure may be more common and sufficient to shape general perceptions.

Environmental perception was also significant. Each one-point increase in agreement that EVs can reduce harmful emissions from ICEVs was associated with higher odds of expecting widespread EV adoption (OR = 1.32, p < 0.001). This indicates that respondents who recognized the environmental benefits of EVs were more likely to expect their future diffusion in Kuwait. This finding highlights the importance of environmental awareness as part of public perceptions of EV adoption, even in a petroleum-dependent economy.

Perceived complexity was negatively associated with expectations of future EV adoption. Each one-point increase in agreement that using EVs would be complicated was associated with lower odds of expecting widespread adoption (OR = 0.80, p < 0.001). This result indicates that complexity concerns may reduce confidence in EV diffusion. Therefore, perceived ease of use, clarity of charging procedures, and consumer familiarity with EV operation are likely important factors in shaping public expectations.

Safety-related variables showed a nuanced pattern. The perceived difficulty of extinguishing an EV fire was positively associated with expected widespread adoption (OR = 1.20, p = 0.005), and perceived lower safety of EVs compared with conventional vehicles was also positively associated with expected adoption (OR = 1.29, p < 0.001). These findings should be interpreted carefully. They do not imply that safety concerns encourage personal EV use. Rather, they may indicate that some respondents expect EVs to become more widespread despite recognizing safety-related concerns. This interpretation is consistent with the nature of the dependent variable, which measures expectations of future market diffusion rather than individual purchase intention. In contrast, the statement that difficulty extinguishing EV fires in hot weather would prevent EV use was not statistically significant (OR = 1.07, p = 0.291), suggesting that general fire-safety perceptions and personal deterrence may represent different dimensions of public concern.

Overall, the full ordinal logistic regression model indicates that expectations of widespread EV adoption in Kuwait are shaped by a combination of economic, practical, technological, environmental, experiential, and safety-related factors. Low electricity tariffs and delayed electricity-bill collection were strongly associated with expectations of EV diffusion, but they were not the only relevant factors. Higher weekly fuel costs, EV performance perception, workplace charging, battery warranty, previous passenger exposure to EVs, environmental benefit perception, perceived complexity, and safety-related perceptions also contributed to respondents’ expectations. These findings support a balanced interpretation: low electricity tariffs may create a favorable economic context for EV adoption, but successful EV diffusion in Kuwait will also depend on infrastructure readiness, battery-confidence measures, public education, safety preparedness, and broader consumer familiarity with EV technology.

4.7. Policy Implications for EV Adoption in Kuwait

The findings have several implications for EV policy development in Kuwait. First, the significant positive association between low-tariff/delayed-bill perception and expected EV adoption suggests that Kuwait’s low electricity tariffs may create a favorable economic environment for EV diffusion. However, the results do not support a policy strategy that relies solely on subsidized electricity as the main mechanism for encouraging EV adoption. Although low operating costs may increase the perceived attractiveness of EVs, the full model demonstrates that adoption expectations are also influenced by performance, workplace charging, battery warranty, environmental awareness, complexity concerns, safety perceptions, and fuel-cost burden. Therefore, low electricity tariffs should be viewed as a supportive condition rather than a sufficient policy instrument.

Second, the positive association between weekly fuel expenditure and expected EV adoption suggests that households with higher fuel costs may be more receptive to EVs. This has practical relevance for targeting awareness campaigns and incentive programs. Policy communication could emphasize the potential operating-cost advantages of EVs for high-mileage users and households with multiple vehicles. However, such communication should be evidence-based and transparent, accounting not only for electricity costs but also for vehicle purchase price, charging availability, battery warranty, maintenance, and long-term ownership costs.

Third, the significance of workplace charging indicates that EV infrastructure planning should extend beyond residential charging. In Kuwait, many residents live in shared family housing, rented accommodation, or areas where private charging access may not be straightforward. Expanding workplace charging at government institutions, universities, commercial centers, and large employment hubs may increase perceived practicality and reduce concerns about charging accessibility. Such infrastructure would be particularly relevant for public-sector employees, who form a large proportion of the national workforce.

Fourth, the importance of battery warranty highlights the need for consumer protection and market-confidence policies. In hot-climate countries, consumers may be particularly concerned about battery degradation, replacement cost, and long-term reliability. Policymakers could work with vehicle dealers, manufacturers, and regulatory authorities to promote minimum battery-warranty standards, transparent battery-health reporting, and clear warranty conditions suitable for Kuwait’s climatic conditions. These measures may reduce uncertainty and improve public confidence in EV ownership.

Fifth, the positive association between EV performance perception and expected adoption suggests that public awareness should not focus only on environmental benefits. Campaigns should also communicate improvements in EV acceleration, reliability, driving comfort, range, and suitability for local travel patterns. Demonstration events, test-drive programs, and public-sector pilot fleets could improve familiarity with EV technology and correct misconceptions regarding performance and usability. The finding that previous passenger exposure was positively associated with expected adoption further supports the value of direct public exposure to EVs.

Sixth, the negative association between perceived complexity and expected adoption indicates that ease of use should be a central component of EV policy. Charging payment systems, mobile applications, public charging instructions, warranty terms, and maintenance requirements should be simple, standardized, and clearly communicated. Public education programs should explain basic EV operation, charging procedures, battery-care practices, and expected ownership requirements in accessible language. Reducing perceived complexity may be especially important for older users and groups less familiar with emerging vehicle technologies.

Finally, the safety-related findings suggest that EV fire-risk communication should be handled carefully. Public concern about EV fire suppression, thermal events, and perceived lower safety should not be dismissed. At the same time, these concerns should be addressed through accurate technical information, emergency-response preparedness, and transparent safety protocols. Kuwait’s emergency services, civil defense authorities, parking-facility operators, and charging-infrastructure providers should develop clear guidelines for EV fire response, battery thermal-runaway incidents, post-fire monitoring, and safe vehicle storage after a fire. In a hot-climate context, visible preparedness may be essential for maintaining public trust.

Overall, the findings suggest that EV adoption in Kuwait requires an integrated strategy. Low electricity tariffs may improve the perceived economic feasibility of EVs, but they should be complemented by charging-infrastructure development, workplace charging expansion, warranty assurance, public education, performance awareness, targeted cost-saving communication, and EV-specific safety preparedness. A comprehensive approach is more likely to support sustainable EV diffusion than a policy framework based only on low electricity prices.

5. Conclusions

This study examined public perceptions and expectations of future electric vehicle (EV) diffusion in Kuwait using a structured questionnaire distributed through online and social media platforms. A total of 1753 valid responses were analyzed, providing insight into the views of licensed drivers across different demographic and socioeconomic groups. The study used descriptive statistics, reliability analysis, group-comparison tests, and principal component analysis to examine general attitudes toward EVs, perceived benefits, practical constraints, and safety concerns. To strengthen the interpretation of the findings beyond descriptive analysis, ordinal logistic regression was conducted to identify the factors independently associated with respondents’ expectations of widespread EV adoption in Kuwait over the next ten years.

The findings indicate that Kuwait has several conditions that may support future EV adoption, particularly low electricity tariffs, high fuel-cost burdens among some respondents, and increasing awareness of environmental benefits. In the full ordinal logistic regression model, low-tariff/delayed-bill perception was the strongest positive predictor of expected EV adoption. Respondents who agreed more strongly that Kuwait’s low electricity tariffs and delayed electricity-bill collection could increase EV use were significantly more likely to expect widespread EV adoption in the future. This finding supports the importance of Kuwait’s low-cost electricity environment in shaping public expectations of EV diffusion. However, the regression results also show that low electricity tariffs should not be interpreted as the only factor influencing EV adoption perceptions.

Several additional factors were significantly associated with expectations of future EV adoption. EV performance perception, environmental benefit perception, workplace charging, battery warranty, previous passenger experience in an EV, and higher weekly fuel expenditure were positively associated with stronger expectations of EV diffusion. These findings suggest that respondents are more likely to expect widespread EV adoption when they perceive EVs as high-performing, environmentally beneficial, supported by charging access, and protected by long-term battery warranties. The positive association with weekly fuel expenditure also indicates that respondents who spend more on gasoline or diesel may be more receptive to EVs as a future alternative.

The results further show that perceived complexity remains an important barrier. Respondents who viewed EV use as complicated were less likely to expect widespread EV adoption. This suggests that public familiarity, user-friendly charging systems, clear ownership guidance, and consumer education are important for improving confidence in EV technology. Therefore, promoting EV adoption in Kuwait will require not only economic incentives but also practical measures that reduce uncertainty and simplify the ownership experience for potential users.

Safety concerns also remain a central issue. The model showed that perceptions related to EV fire-extinguishing difficulty and lower EV safety were significantly associated with expectations of future EV adoption. These results should be interpreted carefully because the dependent variable measured expectations of future market diffusion rather than personal purchase intention. The findings suggest that some respondents may expect EVs to become more widespread despite recognizing safety-related concerns. Therefore, concerns about EV battery fires should not be dismissed. Instead, they should be addressed through transparent technical information, public awareness campaigns, safety regulations, emergency-response training, and EV-specific fire-management protocols suitable for Kuwait’s extreme climatic conditions.

Overall, the findings support a balanced interpretation of EV adoption potential in Kuwait. Low electricity tariffs may create a favorable economic environment for EV diffusion, but widespread adoption will depend on a broader set of factors, including charging infrastructure, workplace charging availability, battery-warranty assurance, vehicle performance confidence, environmental awareness, perceived ease of use, and public trust in EV safety. A successful EV transition in Kuwait will therefore require an integrated policy framework that combines economic, infrastructure, technical, and safety-focused measures.

Future research could build on these findings by examining actual EV ownership and usage behavior rather than relying only on stated perceptions and expectations. Longitudinal or revealed-preference studies would help determine whether perceptions of electricity tariffs, charging infrastructure, vehicle performance, and fire safety translate into actual adoption decisions over time. Technical studies on EV battery performance, thermal behavior, and fire risk under extreme temperature conditions would also provide valuable evidence for safety regulation and emergency-response planning. Finally, comparative studies across Gulf Cooperation Council countries could clarify how differences in electricity pricing, fuel costs, charging infrastructure, climate, and policy frameworks influence EV adoption perceptions and behavior across the region.