1. Introduction

Worldwide, governments have made much of the importance of citizens' responsibility for their local environment. This is characterized by decreased power consumption (Ek & Söderholm, 2010). When it comes to energy, it has been considered an issue and a serious problem all over the world. The efficient utilization of energy in households is a significant key towards national security in energy, climate change prevention, and the reduction of greenhouse gas emissions (Nguyen & Nguyen-Ngoc, 2025). Energy consumption is high in household appliances, so the demand for households to buy energy-saving models is a prime example of how to use energy efficiently and economically (Guan & Lin, 2025; Nguyen & Nguyen-Ngoc, 2025). During the United Nations Sustainable Development Summit, all 193-member states agreed to adopt the United Nations Sustainable Development Goals (SDGs), a global framework designed to address major economic, social, and environmental challenges in an integrated manner and to promote sustainable development by the year 2030 (Shang et al., 2023; Donato & Adıgüzel, 2022; Wang et al., 2021). These goals emphasize the importance of meeting present developmental needs while ensuring that the ability of future generations to meet their own needs is not compromised (Shang et al., 2023; Wang et al., 2021). In response to this global agenda, sustainable consumption—often referred to as green consumption—has increasingly become a key area of focus for governments, businesses, and individuals worldwide. Growing awareness of environmental challenges and resource limitations has led to greater interest in adopting consumption patterns that support environmental sustainability and responsible resource use (Donato & Adıgüzel, 2022; Shang et al., 2023; Wang, Pacho, et al., 2019).

Home appliances like air conditioners, refrigerators, etc., account for the largest components of energy use and carbon emissions (Wang, Pacho, et al., 2019). He added that household appliances contribute to 70% of household CO2 emissions, among which air conditioning, refrigerator, and TV contribute half of the total. Herein, Guan & Lin (2024) and Olsthoorn et al. (2023) urged the adoption of energy-efficient home appliances (EEHA) to reduce energy consumption and carbon dioxide emissions. Thus, the purchasing intention of EEHA encompasses a positive impact on energy use within the household and may cultivate positive buying behavior. Therefore, this purchase needs further investigation (Nguyen & Nguyen-Ngoc, 2025; Lin & Dong, 2023; Olsthoorn et al, 2023). High-efficiency electrical home appliances that are manufactured by modern technology will reduce the usage of power (Tan et al, 2017), such as refrigerators and air conditioners.

Encouraging households to have EEHA is also good for the household sector as this would contribute to environmental sustainability and energy saving (Olsthoorn et al, 2023; Ali et al., 2019; Guan & Lin, 2024). It is generally believed that lower energy consumption and higher efficiency are the most positive, quickest, cheapest, and safest means of relieving environmental degradation and global warming (Zhou & Yang, 2016). Consumers’ demand for five-star-rated energy-efficient appliances at home could also go a long way in determining the purchasing behavior of these energy-efficient consumers because they have more to offer than traditional appliances in terms of energy efficiency and sustainability. Further, they do not need ongoing effort from customers as it is a single investment (Frederiks et al., 2016).

Tan et al. (2017) figured that people can lessen negative environmental impacts by purchasing EEHA. Additionally, energy-saving offerings benefit the economy and environment. Wang et al. (2017) claimed that household appliances contribute to 70% CO2 emissions, whereas TVs, refrigerators, and air conditioners contribute to 50% CO2 emissions. Consumers favor those products of companies that show themselves as eco-friendly and corporate socially responsible (Shang et al., 2023). Some organizations also develop social reputation and realize environmental performance by integrating ethical action into organizational strategy and environmental strategy (Sarker et al., 2019).

Energy-efficient equipment requires less energy than conventional ones for a similar output and hence offers the corresponding performance, convenience, and comfort (Tan et al., 2017). Thus, since traditional appliances are a cause of an increase in emissions and energy use, investment in high-efficiency equipment products is necessary (Joshi et al., 2020). However, the concepts “energy efficiency” and “energy conservation” hold different meanings. Energy efficiency is a reduction in direct energy use through new technology (Poortinga et al., 2003), whereas energy management refers to the change in technology or human behavior aimed at conserving electrical resources. Consumer behavior change is a prerequisite to successful implementation for EEA, but not for EE (Oikonomou et al., 2009).

Countries around the world are focusing on green energy to support and sustain natural resources for a brighter future. In Saudi Arabia, the consumption of electricity accounts for more than a third of its daily oil production. Studies have demonstrated the extent to which buildings in KSA are exposing themselves as a challenge, with the high rate of electricity consumption, which is mainly used for air conditioning (AC) units, has been estimated at 70% of electricity consumed. According to a government report, refrigerators were found to consume approximately twice as much electricity as air conditioners during an average week, highlighting their substantial contribution to household energy consumption (Felimban et al., 2019). In response to growing concerns about residential energy use, researchers have increasingly examined consumer purchasing behavior related to energy-efficient home appliances (EEHA). A considerable body of research has focused on developed economies such as the United Kingdom, the Netherlands, Australia, and the United States (Abrahamse & Steg, 2009; Baldini et al., 2018; Ek & Söderholm, 2010). Similarly, studies conducted in emerging economies—including Malaysia, China, Vietnam, and India—have explored the determinants influencing consumers’ adoption of energy-efficient technologies (Jain & Kaur, 2006; Li et al., 2021; Tan et al., 2017; Zhang et al., 2019). Despite this growing body of literature, research examining these issues within the Saudi Arabian context remains limited. Given that countries differ significantly in terms of socioeconomic conditions, energy policies, and resource availability, consumer preferences and technology adoption patterns may vary accordingly. Therefore, investigating the barriers and motivations influencing the adoption of energy-efficient technologies—particularly photovoltaic-thermal (PVT) technologies—in Saudi Arabia remains an important research gap.

Previous studies have also explored factors influencing consumers’ intentions and behaviors regarding the purchase of energy-efficient home appliances by focusing on individual values and perceptions. For example, Lin and Dong (2023) reported that attitude, subjective norms, and perceived behavioral control significantly and positively affect consumers’ intentions to purchase EEHA in Taiwan. Similarly, Waris and Hameed (2020) applied the Theory of Planned Behavior (TPB) to examine purchase intentions for energy-efficient appliances in Pakistan and found that attitude, subjective norms, and perceived behavioral control were significant predictors of purchasing intention. Furthermore, Harun et al. (2022) confirmed the relevance of TPB constructs in explaining the adoption of energy-efficient appliances in Malaysia, highlighting the significant roles of attitude and normative influences.

Wang, Sun, et al. (2019) find that consumers’ purchase intent was influenced by consumers’ buying attitudes on energy-saving products, personal norms, and subjective norms. Basiru et al. (2024) found that the attitude towards energy-efficient appliances (EEAs) and PBC had a positive and significant impact on the intention to use EEAs. The subjective norm (SN) was found to be insignificant. Nguyen & Nguyen-Ngoc (2025) demonstrate how ATT, PBC, and SN positively influence consumers’ purchase intentions for EEHA in Ho Chi Minh City, Vietnam, as the independent constructs. Arora and Varah (2025) identify the same findings in the Indian context. Furthermore, Jia et al. (2024) extracted related findings in the Pakistan context.

Bhutto et al. (2022) extended the TPB model. They used the model to predict young consumers’ purchase intention towards EEHA in the Pakistani context. According to the results of this study, the self-expressive factor has a low impact on attitude. Meanwhile, the mediation effect remained insignificant with the given insignificant relationship between self-expressive benefits and purchase intention.

Liao et al. (2020) examined the impact of self-expressive benefits in energy-saving products that positively affect energy-saving appliances. Hartmann and Apaolaza-Ibáñez (2012) did not show a positive relationship between self-expressive value vis-à-vis brand attitude and purchase intention. Boobalan et al. (2021) examined the mediating effect of attitude between self-expressive benefit and purchase intention. We clearly found it from mediation analysis that attitude acts as a mediatory variable between self-expressive benefit and purchase intention for the entire sample, at both the country (India & USA) level. But attitude did not emerge as a mediator between relationship self-expressive benefits and purchase intention in the case of the Indian sample only.

Consumers’ desire to buy energy-saving products is deeply affected by their environmental awareness. Increasing public knowledge and providing environmental education are easy approaches to enhance environmental knowledge (Hossain, Fekete-Farkas, et al., 2022). Shahzad et al. (2022) asserted that knowledge was recognized to be the most relevant factor among the factors of green technology adoption, especially for energy efficiency home appliances. Li et al. (2021) found that the impact of environmental knowledge was not significant as a moderating effect between attitudes and purchase intention on energy-saving home appliances. Wang, Sun, et al. (2019) also included and verified that knowledge plays a positive moderating role in the construct of subjective norm and attitude with purchase intention.

As such, in order to fill up the gaps identified above in the literature, this study explores what determines Saudi Arabian purchasing intention of energy efficiency appliances. The current paper contributes to the literature in two aspects: a new setting and novel variables. The present study wishes to observe the essential elements of Saudi Arabian consumers' purchase intention that will expand our understanding of energy efficiency appliances purchase in a developing country context. With respect to the new variables, this study applies the TPB model along with introducing some other constructs like Self-Expressive Benefit and environmental knowledge. The environmental value was introduced as the schema variables of the attitude, subjective norm, and perceived behavioral control on buying intention. As a mediator, the attitude was analyzed between self-expressive benefit and purchase intention. The study investigated purchasing energy-efficient appliances, which is rare in marketing research for integrating both PLS-SEM and ANN.

2. Literature review and hypotheses development

3. Methodology

4. Analysis and Results

4.2 Multivariate statistical assumptions

SmartPLS 4 was employed to analyze the proposed research model due to its suitability for variance-based structural equation modeling (PLS-SEM). This technique is particularly appropriate for exploratory research, complex models involving mediation and moderation effects, and datasets that do not meet multivariate normality assumptions. SmartPLS 4 also offers advanced bootstrapping procedures, improved graphical visualization, and enhanced computational efficiency compared to earlier versions. Given the study’s objective to assess both predictive relationships and indirect effects, SmartPLS 4 provides a robust and reliable analytical platform.

Several different conditions must be met before multivariate analysis can occur. The multicollinearity issue was checked by VIFs and Tolerances, see Table 3. As the VIF values fall within 1.00–1.82, which is below the normal cutoff of 10, there is no multicollinearity problem (Hew & Kadir, 2016).

Table 3Multicollinearity 

Construct	VIF
Attitude	1.44
Environmental Knowledge	1.19
Perceived Behavioral Control	1.82
Self-Expressive benefit	1.00
Subjective Norms	1.76

After checking homoscedasticity based on the standard residual scatter plot (Figure 1), authors observed that the residuals are scattered along a diagonal. Thus, homoscedasticity remains verified. Using a one-sample Kolmogorov-Smirnov test, the normal distribution is tested. Having all p-values below 0.05 indicates that the data distribution needs to be normal. Hence, partial least squares (PLS) variance-based SEM are used because PLS is robust against non-normal distribution in relation to covariance-based SEM (Leong et al., 2019). The authors used Ringle et al.’s (2015) SmartPLS 4 to test the hypotheses in the research framework. Additionally, an ANN analysis was performed in order to rank the standardized importance of the significant predictors according to PLS analysis. Application of SEM-PLS-ANN in two stages would be complementary, as the second stage (ANN) to cover the other business question is not feasible for conducting hypothesis testing on linear relationships, while SEM-PLS can deal well with linear relationships but cannot capture non-linear relationships. The ANN can identify non-linear relationships, but not suitable for hypothesis testing due to “black box” operation (Hew et al., 2016).

Figure 1Residuals vs. Predicted Values 

4.4 Interpretation of Structural Model Results

The structured model results posit insights into the direct, mediated, and moderated relationships among self-expressive behavior, attitude, subjective norms, perceived behavioral control, environmental knowledge, and purchase intention. The authors summarize the results of hypothesis testing below in Table 8.

4.4.1 Direct Relationships

The findings indicate that ATT, SN, and PBC hold a positive and statistically significant relationship with purchase intention. The values of the standardized path coefficient of ATT (β = 0.22, p < 0.001), SN (β = 0.16, p < 0.001), and PBC (β = 0.22, p < 0.001) indicate that people with positive attitude, strong social pressure, and high perceived control will be more likely to form intentions toward purchase behavior as H1 predicted.

Table 8Path Analysis 

Hyp	Structural Path	Beta	T values	P Values	Remarks
Direct Relationships
H1	Attitude ⟶ Purchase Intention	0.22	5.09	0.00	Supported
H2	Subjective Norms ⟶ Purchase Intention	0.16	4.39	0.00	Supported
H3	Perceived Behavioral Control ⟶ Purchase Intention	0.22	4.79	0.00	Supported
H4	Self-Expressive ⟶ Attitude	0.53	14.75	0.00	Supported
Mediated Relationship
H5	Self-Expressive ⟶ Attitude ⟶ Purchase Intention	0.12	4.53	0.00	Supported
Moderated Relationships
H6	Attitude x Environmental Knowledge ⟶ Purchase Intention	0.06	1.21	0.23	Not Supported
H7	Subjective Norms x Environmental Knowledge ⟶ Purchase Intention	0.11	2.76	0.01	Supported
H8	Perceived Behavioral Control x Environmental Knowledge ⟶ Purchase Intention	0.01	0.14	0.89	Not Supported

Formal path assimilations are also presented in Figure 2. In addition, Self-Expressive Behavior (H4) significantly affects Attitude (β = 0.53, p < 0.001), meaning a customer involved in self-expressive behavior is more likely to form a favorable attitude concerning the purchase decision.

4.4.2 Mediated Relationship

The mediation analysis supports H5, showing that Self-Expressive Behavior indirectly influences Purchase Intention through Attitude (β = 0.12, p < 0.001). This result confirms that self-expressive behavior enhances attitudes, which in turn leads to stronger purchase intentions.

4.4.3 Moderated Relationships

Environmental Knowledge does not significantly moderate the influence of Attitude on Purchase Intention (H6) (β = 0.06, p = 0.23), so that Environmental Knowledge does not reinforce or attenuate the effect of Attitude on Purchase Intention. Likewise, Environmental Knowledge is not a significant moderator of the effect of Perceived Behavioral Control on Purchase Intention (H8) (β = 0.01, p = 0.89), indicating that perceived control over purchases does not vary with consumers’ level of environmental knowledge. In contrast, the interaction between Subjective Norms and Environmental Knowledge (H7) is found to be significant (β = 0.11, p = 0.01), indicating that high environmental knowledge reinforces the influence of subjective norms on purchase intention. It indicates that, for purchase intention, the influence of social norms is stronger among those who know about environmental issues. The non-significant moderating effects of environmental knowledge on the relationships between attitude, perceived behavioral control, and purchase intention suggest that these constructs may operate largely independently of consumers’ environmental awareness. Attitude toward energy-efficient appliances is often shaped by personal evaluations, economic considerations, and habitual preferences, which may remain stable regardless of knowledge level. Similarly, perceived behavioral control reflects consumers’ perceptions of affordability, availability, and access, factors that are not necessarily influenced by environmental information. These findings are consistent with previous studies reporting weak or insignificant moderation effects of environmental knowledge in green purchase contexts, suggesting that knowledge alone may be insufficient to strengthen internal drivers of behavior.

Figure 2Structural Model 

4.5 Artificial Neural Network (ANN) Analysis

In addition to a regression analysis and considering non-linear relationships between predictors of PI (Performance Indicator), this research uses an ANN model. The ANN method is based on previous research (Leong et al., 2019) and applies an MLP with a feed-forward back-propagation algorithm. This approach offers richer views of variable importance and predictive accuracy than classic regression models.

4.5.1 ANN Model Architecture and Training

The ANN model used in this work is a three-layer multilayer perceptron (MLP) that can model complex and nonlinear relationships between the independent variables and the dependent variable. Figure 3 provides the graphical depiction of the model. The input layer is composed of five neurons, which correspond to the interesting predictors (SEB, ATT, SN, PBC, and EK), as shown by the statistically highest significance in previous analyses (Leong et al., 2019). The hidden layer consists of ten neurons based on prior methodological recommendations in ANN applications within social science research (e.g., Haykin, 1999; Leong et al., 2019), as well as through iterative testing to balance model complexity and predictive accuracy while avoiding overfitting. by a single neuron of the output layer, which predicted the value of PI (Purchase Intention) as an exchangeable.

Figure 3Artificial Neural Network 

The model was compiled with the Adam optimizer at a learning rate of 0.001, and using Mean Squared Error (MSE) loss to minimize prediction errors. The network’s training was stopped at 1,000 iterations to improve generalization and avoid overfitting, leading to an optimal solution. This methodological procedure is consistent with the recommended neural network applications in social science research (Haykin, 1999), and it offers a strong procedure to test the prediction potential of central behavioral determinants.

4.5.2 Model Performance Evaluation

To evaluate the predictive power of the ANN model, it was validated by using different train test split ratios for generalization performance analysis. This methodology was adopted based on similar studies that also compare the performance of machine learning models using various data partition approaches (Leong et al., 2019). The metrics of performance are: Mean Squared Error (MSE), Root Mean Squared Error (RMSE), and R² Score (Table 9).

Table 9ANN Performance Across Different Training-Testing Splits 

Train-Test Split	MSE	RMSE	R²
50-50	0.2998	0.5475	0.423
60-40	0.3221	0.5676	0.377
70-30	0.2783	0.5275	0.451
80-20	0.2953	0.5434	0.450
90-10	0.3037	0.5511	0.467

The best predictive performance was revealed with the 70–30 split; the lowest MSE (0.2783) and highest R² (0.451). This is consistent with the results reported from ANN-based prediction modeling, as moderate training proportions usually contribute to better generalization (Zhang et al., 2019).

4.5.3 Sensitivity Analysis of Predictors

A sensitivity analysis was performed using this model to assess the relative impacts of each predictor on prediction by perturbing input variables and observing changes in output predictions (Table 10). This is in line with the previous ANN ranking of variable methodologies (Olden & Jackson, 2002).

Table 10Sensitivity Analysis of Predictors (ANN Rankings) 

Feature	Sensitivity	Normalized Sensitivity (%)
ATT (Attitude)	0.00693	100.00
SN (Subjective Norms)	0.00507	73.20
SEB (Self-Expressive benefit)	0.00507	73.14
PBC (Perceived Behavioral Control)	0.00245	35.34
EK (Environmental Knowledge)	0.00131	18.91

Results indicate that ATT (Attitude) is the most relevant of the PI system, before SN (Subjective Norms), but after SEB (Self-Expressive Benefit). A lower effect of EK (Environmental Knowledge) shows that environmental issues have less influence than economic or behavioral factors in comparison.

5. Discussion

The objective of the study was also to integrate additional cognitive antecedents (like Self-expressive benefits and environmental knowledge) in addition to TPB variables. The findings indicated that the developed model is generally acceptable and suitable for understanding product purchases of ‘Energy Efficient Appliances’ as extended TPBs can predict purchasing intention.

The findings reveal that Attitude significantly has a positive effect on the purchasing intention of energy efficiency appliances. It is indicated that the more favorable households feel energy-saving appliances, the likelihood of making such a purchase becomes higher, and this confirms results obtained in (Nguyen & Nguyen-Ngoc, 2025; Basiru et al., 2024). This study also affirmed the proposition that customers’ subjective norms serve the intention of buying energy-saving home appliances. The subjective norms were a meaningful predictor of purchase intention. It is concluded that consumers are influenced by others´ values when buying energy-efficient home appliances. This is in line with results from existing studies, which demonstrated that subjective norms stimulate environmental behaviors and purchasing of green products (Nguyen & Nguyen-Ngoc, 2025; Basiru et al., 2024).

The findings showcase that the PBS factor of PBC positively and significantly affects purchase intention of EEHAs. The PBC strongly influences intention to purchase as an indicator of the degree to which a person feels that an action is easy or difficult to perform. When consumers have enough money, access to products and information for evaluating them in the case of buying energy-saving appliances, they have a higher intention to purchase. On the other hand, if barriers are feared (e.g., high price or low availability, and lack of knowledge), then intention to purchase is reduced. PBC not only directly affects purchase intention but can also function as a moderator that reinforces or weakens other variables, attitude, and subjective norms in this case.

This corresponds to results from Myanmar, Kashmir, and Vietnam, in which Perceived Behavioral Control influenced purchase intention (Nguyen & Nguyen-Ngoc, 2025; Basiru et al., 2024). Self-Expressive Behavior has a material impact on attitude by enabling customers to bring their purchase in harmony with self-concept, values, and social image. Once people come to view purchasing energy-efficient appliances as consistent with their self-concept (e.g., environmentally responsible, modern, socially aware), they develop a more favorable evaluation of and attitude toward such products. This association will elevate emotional responses and reinforce the preference for green alternatives, subsequently facilitating acceptance of energy-efficient products. Hence, self-expressive benefits act as a psychological motivator, influencing consumer attitudes and supporting sustainable consumption behavior. This result concurs with those of others that self-expressive benefit influenced the purchase intention (e.g., Liao et al. 2020).

The mediation analysis provides evidence that self-expressive benefits will indirectly affect purchase intention via attitude. This result suggests that associating energy-efficient appliances with self-expression in terms of the consumers’ personal values or social identity leads to a more positive attitude toward these products. This positive attitude, in its turn, reinforces the intention to buy from these people. This finding underscores the importance of attitude as a psychological conduit that converts self-expressive motives to specific purchase decisions. Similar results are in line with the previous studies (Boobalan et al. 2021).

The results of this study reveal that environmental knowledge does not significantly moderate the relationship between attitude and purchase intention toward energy-efficient home appliances. This suggests that although consumers may possess knowledge about environmental issues, such knowledge does not necessarily strengthen the influence of their attitudes on purchasing decisions. In other words, consumers who already hold favorable attitudes toward energy-efficient appliances may base their intentions primarily on perceived product benefits such as cost savings, efficiency, and long-term economic value rather than on their environmental awareness. Prior research has shown that consumers frequently evaluate green technologies based on utilitarian and economic benefits, which often outweigh purely environmental motivations in shaping purchase intentions (Lin & Dong, 2023; Asif et al., 2023).

This finding aligns with earlier research indicating that environmental knowledge does not always reinforce the effect of pro-environmental attitudes on behavioral intentions. For example, Li et al. (2021) reported that environmental knowledge failed to significantly moderate the relationship between attitude and purchase intention for energy-efficient appliances in China. Similarly, Hamzah and Tanwir (2021) found that environmental knowledge did not strengthen the impact of pro-environmental attitudes on consumers’ purchase intentions in the context of hybrid vehicles. Additional studies also highlight that although environmental knowledge increases awareness, it may not directly translate into stronger purchase intentions without supportive motivational or contextual factors (Rehman et al., 2024; Anwar et al., 2024).

These results also complement more recent studies highlighting the dominant role of psychological and behavioral determinants —particularly attitude— in shaping sustainable consumption decisions (Basiru et al., 2024; Nguyen & Nguyen-Ngoc, 2025). Empirical evidence further suggests that attitudes toward green products are often formed through perceived value, trust, and product benefits rather than solely through environmental awareness (Lin & Dong, 2023; Tanveer et al., 2024). Taken together, these findings suggest that while environmental knowledge contributes to general awareness, it may not substantially intensify the attitudinal influence on consumers’ purchase intentions.

Similarly, the findings demonstrate that environmental knowledge does not significantly moderate the relationship between perceived behavioral control and purchase intention. This implies that consumers’ perceived ability to purchase energy-efficient appliances—such as affordability, accessibility, and availability of products—remains a stronger determinant of purchase intention than their environmental awareness. Even consumers with high environmental knowledge may not exhibit stronger purchasing intentions if they perceive financial or practical barriers associated with energy-efficient appliances. Previous studies examining energy-efficient appliance adoption have also emphasized that perceived behavioral control and situational constraints significantly influence consumers’ purchasing decisions (Mohamed et al., 2025; Lin & Dong, 2023).

This result is consistent with prior research suggesting that structural and situational constraints often limit the translation of environmental awareness into actual purchase intentions. For instance, Hamzah and Tanwir (2021) also reported a non-significant moderating effect of environmental knowledge on the relationship between perceived behavioral control and pro-environmental purchase intentions. However, this finding contrasts with studies that suggest environmental knowledge can reinforce sustainable consumption behavior by enabling consumers to better evaluate environmental benefits and product attributes (Wang, Sun, et al., 2019; Jia et al., 2024). The divergence between these results indicates that the influence of environmental knowledge may be context-dependent and varies across product categories and cultural settings. Recent research also indicates that contextual factors such as environmental concern, social influence, and trust can significantly influence sustainable consumption behaviors alongside traditional TPB variables (Khan & Qureshi, 2025; Tanveer et al., 2024). Therefore, in the context of energy-efficient home appliances, economic and practical considerations appear to outweigh informational factors in shaping consumers’ purchasing decisions.

Overall, the non-significant moderating role of environmental knowledge suggests that knowledge alone may not be sufficient to strengthen the psychological mechanisms that drive sustainable purchasing behavior. While environmental knowledge increases awareness of environmental issues, it does not necessarily translate into stronger behavioral intentions unless it is accompanied by favorable attitudes, perceived benefits, and enabling conditions. Recent studies emphasize that consumers’ green purchasing decisions are often influenced more strongly by perceived economic value, social influence, and product accessibility than by environmental awareness alone (Anwar et al., 2024; Tanveer et al., 2024; Lin & Dong, 2023).

Therefore, the findings of this study highlight the importance of complementing knowledge-based interventions with strategies that enhance positive attitudes and reduce perceived barriers to purchasing energy-efficient appliances. From a policy and marketing perspective, this implies that awareness campaigns should be integrated with economic incentives, clear product information, and improved accessibility to encourage the adoption of energy-efficient technologies. Such integrated approaches may be more effective in translating environmental awareness into actual purchase intentions and sustainable consumption behavior.

6. Theoretical Contribution

This study has several implications for the existing body of literature. First, to the best of the researchers’ knowledge, this is one of the first studies explaining purchasing intention for residential energy-efficient appliances among Saudi Arabian consumers. Saudi Arabia is categorized as a developing country, and the income level is high, but temperature increases lead to high electricity demand. Second, this research expanded TPB by incorporating Self-expressive benefits and environmental knowledge to measure the intention of household energy-efficient appliance purchase. Previous research has further expanded TPB with other variables, namely Self-expressive benefits (Bhutto et al. 2022), and attitude is insignificantly affected by self-expressive. The current study also shows that attitude is not a significant mediator in the self-expressive benefit and purchase intention relationship, so further investigation is required to establish the results. Environmental knowledge was the second variable (Li et al. 2021); however, it did not pay attention to the tendency effect among subjective norms and perceived behavior control on purchasing intention of energy-efficient appliances. Lastly, as far as the study specific to purchasing intention of energy efficiency appliances is concerned, this research is among a handful of marketing studies that merged both PLS-SEM and ANN techniques.

7. Implications (managerial and Policy)

To a certain extent, the findings of this study contribute to academic research and provide significant implications for the state and manufacturers of EEAs. As attitude, subjective norm, and perceived behavioral control have a positive effect on purchasing intentions of Saudi consumers, the government should encourage producers, importers, and sellers of appliances that equally promote energy efficiency products positively. The motivations will also increase the purchasing intention level of Saudi consumers toward energy efficiency appliances, such as the refrigerator, which has low efficiency to save electric power compared to traditional products that consume a high amount of energy and pay huge amounts each month.

Social media communications, such as Facebook, LinkedIn, TikTok, X, and Instagram ads, motivate consumers to exchange old home appliances for energy-efficient ones while also controlling the price level of these devices and increasing environmental awareness towards climate degradation. Domestic appliance retailers may provide monetary incentives (for instance, cashback or discounts) when buying energy-efficient appliances. On the other hand, these products should reach markets near consumers to reduce their access time and cost to make an effort to find and acquire them. The self-expressive benefits also have an impact on consumer attitudes to purchase energy-efficient appliances through identity, ascription, and emotional attachment, resulting in the intention to buy them.

For diversified insights, qualitative methodologies can be explored, for they can probe into the purchase intention of consumers on energy-efficient household appliances and reflect a more profound psychological status of consumers. A model that can be used to examine the moderating effects of consumer characteristics such as self-concept, recycling, and cultural knowledge.

8. Conclusions and Limitations

This study investigates the determinants of buying intention of EEHAs in the Saudi Arabian market. It validates the variables of attitude, subjective norms, and perceived behavioral control, which influence consumers’ intention to buy energy-efficient home appliances. Moreover, the results confirm the impact of self-expressive benefits on attitudes towards purchasing energy-efficient home appliances. It has also been found that environmental knowledge moderates the relationship between subjective norms and PEPU for EEHA consumers. Finally, the findings reveal that the attitude mediates between self-expressive benefits and the buying intention of EEHA items.

The purchase intention of EEHAs has been studied in this research. The topic may also be expanded to repurchase intention in the future to see how consumers are retained. In fact, we employed behavioral intentions to purchase energy-efficient appliances. Further research may include actual behavior, which would yield information on the extent to which intention materializes in terms of action.

Despite its contributions, this study has several limitations that should be acknowledged. First, the sample size, while adequate for PLS-SEM analysis, was obtained using a non-probability purposive sampling technique, which may limit the generalizability of the findings. Additionally, the sample was characterized by a relatively high educational level, which may not fully represent the broader Saudi population. Second, the study focused on respondents from a single geographic context within Saudi Arabia, which may restrict the applicability of the results to other regions or countries. Future research should consider employing probability-based sampling methods, larger and more diverse samples, and cross-regional or cross-cultural comparisons to enhance external validity.