Mediating role of regret feelings in the relationship between consumer expectations, emotions and willingness to buy

Abstract

Purpose

The paper aims to investigate the impact of customers’ expectations, negative emotions and regret on consumers' intention to buy Chinese clothing products in the Palestinian market.

Design/methodology/approach

This paper used a convenience sampling technique. The path relationship of the study model was analyzed by structural equation modeling (SEM) based on partial least squares (PLS-SEM).

Findings

Results showed that regret was affected by the negative feelings that consumers could incur after buying the product. Additionally, negative feelings and regret negatively affected consumers' intention to buy Chinese clothing products, while their expectations positively affected their decisions. However, the mediation effect of regret was approved in the relationship between negative feelings and the intention to buy.

Originality/value

This is to certify, that the research paper submitted by us is an outcome of our independent and original work. We have duly acknowledged all the sources from which the ideas and extracts have been taken. The project is free from any plagiarism and has not been submitted elsewhere for publication.

Citation

Abdallrahman, M. and Darwish, N.A. (2024), "Mediating role of regret feelings in the relationship between consumer expectations, emotions and willingness to buy", Rajagiri Management Journal, Vol. 18 No. 4, pp. 323-336. https://doi.org/10.1108/RAMJ-12-2023-0329

Introduction

Nowadays, we live in globalized markets that offer various choices of products, especially in the clothing sector. The availability of different clothing brands from different origins combined with a wide variety in price and quality levels make it very hard to choose the right brand that could offer the highest value, which increases the complexity of buying decisions. However, after buying and consuming any product, in the post-purchase stage of the buying decision, consumers will decide if they are satisfied or dissatisfied with their choice. In case they make a bad decision and buy what they perceive as an inferior product in comparison to other available alternatives, they will regret their decision, and most probably will not buy the same brand again.

The clothing sector in Palestine is one of the most unique sectors since it is characterized by trends that shift from one season to another; style choices reflect individual preferences and are affected by sociocultural influences. To fulfill customer needs for clothing products, the clothing sector in Palestine imports products from many countries, such as Israel, Turkey, and China. In 2022, the total number of textiles imported to the Palestinian market was 182.6 million dollars, of which almost one-third came from China (International Trade Center, 2024), which gives the “Made in China” label huge attention, representing not just a label but a cornerstone in the Palestinian clothing sector’s supply chain.

Each step of making a purchasing decision is accompanied by choice, information, and ambiguity. Consumers are eager to make the appropriate choices that would match their expectations and be followed by satisfaction rather than regret feelings (Durmaz et al., 2020; Sai Vijay et al., 2020). For a Palestinian family composed of 5 members who spend only around 5% (65$\Month) of their income on clothing products (Central Bureau of Statistics, 2022), clothes made in China which are known for their affordability, diversity, and perceived value (Obeng, 2020), is an appropriate decision to buy on a regular basis as in any other developing country where most individuals are classified as having low to medium income. However, some potential negative emotions, such as feelings of disappointment and regret, which align with customer satisfaction or dissatisfaction will highly affect their repurchase decision.

In the Palestinian market, made-in-China products have an inferior perceived image compared with products from other origins, such as Turkey and Europe, and it is expected that consumers will be dissatisfied with their quality; however, they still buy them because of their affordable price, and it is expected that after a certain number of purchases, they will experience regret feelings that at some level will be strong enough to negatively affect their buying decision.

Previous studies have highlighted the importance of regret influence on buying decisions as a tool to judge customer satisfaction (Biondi et al., 2019; Durmaz et al., 2020; Kumar et al., 2021; Sai Vijay et al., 2020). Despite that, the Palestinian consumers' behavior and attitudes toward “Made in China” products had been underexplored - according to our knowledge - especially in the clothing sector. Therefore, this research attempts to bridge the gap in the current literature by offering a better understanding of post-purchase regret, especially in the unique setting of the Palestinian consumer market. Thus, this research aims to explore the influence of customer expectations and feelings toward “made in China” clothing products on consumers' buying decisions. Additionally, this research will explore the mediation function of regret in this connection and its effect both directly and indirectly on consumer intentions to buy “Made in China” clothing products.

This research’s practical implications will be significant for both consumers and industry stakeholders. By comprehending post-purchase regret and its correlation with purchasing decisions, businesses can enhance their marketing approach, while policymakers can acquire valuable insights into consumer feelings. This, in turn, promotes a conducive climate for sustainable economic growth and development

Conceptual framework

Consumers worldwide face the dilemma of choosing the best brand with the highest significant value every day. Making such a decision is not an easy task due to market globalization (Davvetas and Diamantopoulos, 2018), the spreading of shopping malls, and the increase in buying power, which allows local markets to contain various products to choose from and move societies toward a consumer culture where consumers actively engage with services and products providers (Bushra, 2015; Kumar et al., 2021).

According to Durmaz et al. (2020), repurchase intention refers to the extent of a consumer’s willingness to purchase a product from the same company, considering both the present circumstances and potential situations. According to the expectation-confirmation theory, the degree of determination is influenced by the extent of consumer satisfaction with past purchases (Sai Vijay et al., 2020). Therefore, when the product’s perceived performance meets the client’s expectations, customer satisfaction rises; otherwise, post-purchase regret may happen as a result of negative emotions of disappointment and dissatisfaction with the existing choice.

Customer expectations and post-decisional regret

Consumers frequently compare the brands they buy with the alternatives they choose not to buy after making any brand purchase (Diecidue and Somasundaram, 2017; Kumar et al., 2021; Sai Vijay et al., 2020). This comparison often results in a state of sorrow, anxiety, and psychological discomfort known as post-purchase regret (Diecidue and Somasundaram, 2017; Mojzisch et al., 2020) or cognitive dissonance (Wu and Wang, 2017). This state of discomfort will dominate if consumers perceive their purchasing decisions as wrong (Abdel and Saleh, 2012; Bui et al., 2011; Wu and Wang, 2017), and they will feel regret immensely if they are disappointed with the expected or actual outcomes of the chosen brand or discover that the other options may result with more desirable outcomes (Bui et al., 2011; Connolly and Zeelenberg, 2002; Kutscher and Feldman, 2019; Wen-Hai et al., 2019; Zeelenberg et al., 1998).

Regret can manifest in various forms; prior studies have shown that the perception of regret differs based on whether it arises from actions or inactions and whether it is experienced in the short-term or long-term (Gilovich and Medvec, 1995; Rotman et al., 2017). Short-term regrets are immediate responses to a specific outcome that has been personally experienced, whereas long-term regrets are more abstract and are experienced as thoughts or imaginings of what could have happened. Long-term regrets are only felt when the outcomes of alternative choices are revealed, and it is no longer possible to change one’s decisions (Crotty Susan and Thompson, 2009; Pornpitakpan, 2010). In the short term, consumers often regret action more than inaction, but in the long run, they regret inaction more than action (Yeung and Feldman, 2022).

Thus, the study puts forward the subsequent hypothesis to examine the Palestinian customer expectations toward “Made in China” clothing products quality and the expected consequences of their decision and its relation with post-purchase regret:

H1.

The level of expected decision outcomes (EXP) positively relates to regret (REG).

Negative emotions and post-decisional regret

Regret is an awful feeling. In usual circumstances, Regret is experienced when an individual recognizes that their actions have deprived them of what they value, despite having the option to conduct differently (Price, 2020)

Consumer regret is defined as “cognitive and emotional negative feelings that stem from misfortunes, restraints, losses, violations, faults or mistakes” (Durmaz et al., 2020). García and Curras-Perez (2019, p. 7) also defined regret as “the emotion that we experience when realizing or imagining that our current situation would have been better if only we had decided differently. It is a backward-looking emotion signaling an unfavorable evaluation of a decision. It is an unpleasant feeling, coupled with a clear sense of self-blame concerning its causes and strong wishes to undo the current situation”. Additionally, Sai Vijay et al. (2020) describe regret as “ an unpleasant feeling, coupled with a clear sense of self-blame concerning its causes and strong wishes to undo the current situation.”

From previous definitions, we can conclude that regret is caused by the accumulated negative emotions of sorrow, anger, disappointment, distress, and guilt. In this research, we will consider negative emotions that consumers could encounter if they recognize that they did not make the best buying decision after buying a “Made in China” clothing product and that other alternatives of clothing products made in countries other than China could offer them greater value could affect their regret feelings level. Therefore, the study proposes the following hypothesis:

H2.

The level of consumers' negative emotions (NEF) positively relates to regret (REG).

Regret and buying intentions

From a motivational standpoint, regret, as an emotion, is closely tied to the achievement of specific goals and serves as a strong motivator for taking activities that lead toward those goals (Davvetas and Diamantopoulos, 2018). There is a general agreement that regret is based on a cognitive and emotional basis since, for regret to happen, consumers should think about what might have happened if they chose another brand and then experience feelings of sorrow and regret (Durmaz et al., 2020; García and Curras-Perez, 2019).

According to the regret theory, the consumer’s post-purchase reaction will depend on the disconfirmation of expectations and the expected outcomes of the alternative decision (Biondi et al., 2019) because it has been shown to result in lower consumer satisfaction, decreased intention to repurchase, and increased costs associated with switching brands (Abdel and Saleh, 2012; Bui et al., 2011; Krishen Anjala et al., 2010; Kumar et al., 2021; Sai Vijay et al., 2020; Wu and Wang, 2017). Thus, a sense of regret that arises following a decision has varying effects on consumer behaviors and subsequent decisions (Durmaz et al., 2020). Therefore, to measure the direct effect of post-purchase regret on consumers' intention to buy “Made in China” products, the study proposes the following hypothesis:

H3.

The level of experienced regret (REG) negatively relates to the intention to buy Chinese products (INB).

Consumer expectations and buying intention

The theory of regret proposes that consumers try to avoid feelings of regret (García and Curras-Perez, 2019). However, consumers cannot ascertain the product’s precise value before making a purchase so that they will evaluate expected regret or what is called anticipated regret if the decision is going to be made in the near future (Huang et al., 2022; Liu et al., 2023).

Previous studies demonstrate that anticipated regret can negatively influence consumers' buying decisions and level of satisfaction with their decisions (García and Curras-Perez, 2019; Liu et al., 2023). Customer satisfaction is directly related to customers' expectations, and when the expected product value doesn’t match the perceived value, the customers will be dissatisfied and may regret their decision (Şehirli, 2023; Sujono et al., 2024). Therefore, customer satisfaction is related to customers’ expectations and how they perceive the product quality rather than other factors such as low prices or being a high-quality product (Sujono et al., 2024).

Within the Palestinian market, the majority of consumers have acquired apparel labeled as “Made in China” at least once in their lifetime. These items are readily available in nearly every clothing store. As a result, customers will possess knowledge regarding the prices and qualities of the products, and they will have developed expectations regarding the potential results and repercussions of purchasing these items. Thereafter, to test the effect of expected outcomes of buying “Made in China” clothes products on the intention to buy them and to investigate the mediating role of regret feelings that may have been felt in previous occurrences in this relationship. Based on these objectives, the study proposes the following hypotheses:

H4.

The level of expected decision outcomes (EXP) positively relates to the intention to buy Chinese products (INB).

H4.1.

Consumer regret (REG) mediates the relationship between the level of expected decision outcomes (EXP) and consumers' intention to buy Chinese products (INB).

Consumer emotions and buying intention

Consumers' emotions refer to the distinct psychological reactions of consumers to external events that occur throughout the purchase process. Watson and Tellegen (1985) established a two-factor model of emotions, which categorizes emotions into two groups: pleasant emotions and unpleasant emotions. Pleasant emotions encompass feelings of excitement, refreshing, and satisfaction, whereas unpleasant emotions encompass feelings of disappointment, despair, and anger. However, these two contrasting feelings are not entirely separate, and their combined impact influences customers' purchasing decisions (Bai et al., 2017).

The regret theory also distinguished between regret and other emotions such as anger, disappointment, jealousy, guilt, sorrow, and shame in terms of appraisals, experiential content, and behavioral consequences (Connolly and Zeelenberg, 2002; Wu and Wang, 2017). Therefore, Regret is more than a simple evaluation or judgment; it is naturally loaded with feelings and, therefore, eligible to be considered a genuine emotion (Gilovich and Medvec, 1995).

The regret theory shares similarities with the cognitive dissonance theory in certain aspects. The cognitive dissonance theory proposes that consumers may feel some tension or emotional discomfort if they think their decision was not the best one, and they may blame others for making them make this decision to reduce this discomfort. On the other hand, the regret theory supposes that consumers will blame themselves for the wrong decision (García and Curras-Perez, 2019).

Prior research in psychology and related fields has made many efforts to study the relationship between emotions and buying decisions. These studies show that consumers' decisions are influenced by the expected emotional consequences, whether positive or negative, associated with such decisions. Positive anticipated emotions refer to the happy sentiments that come from the potential to reach a future objective, while negative anticipated emotions refer to the negative sensations that result from the inability to attain a future goal. Thus, stronger emotions increase consumer motivation to engage in behaviors that result in favorable outcomes or help them avoid bad consequences (Bettiga and Lamberti, 2018).

This research dealt with the product “country of origin” not only as a set of geographical coordinates but also to study the emotions that consumers hold regarding the origin of products they intend to purchase since the origin of a brand can evoke positive or negative emotions that influence consumers’ decisions (Orth et al., 2020). Furthermore, we will deal with regret as a genuine emotion, as mentioned in the regret theory, and examine its mediating effect in the relationship between consumer emotions toward “Made in China” clothing products and their intention to buy them. Based on these objectives, the study proposes the following hypotheses:

H5.

The level of consumers' negative emotions (NEF) negatively relates to the intention to buy Chinese products (INB).

H5.1.

Consumer regret (REG) mediates the relationship between the level of negative emotions (NEF) and consumers' intention to buy Chinese products (INB).

Figure 1 shows the proposed research model that shows the main study variables, the supposed relationships and the main hypotheses.

Research methods

For this study, we selected the clothes industry as it is a category from which customers commonly buy and derive everyday benefits. The high frequency of consumption, coupled with low barriers to entry and exit, leads to numerous firms operating in intense competition. As a result, apparel retailers can greatly benefit from the findings of this study in order to enhance their understanding of consumer behavior and implement effective marketing tactics.

This study was conducted with the aim of determining the effect of negative emotions, expectations, and regret felt by consumers after purchasing “made in China” clothing products on repurchase intention. The population under study comprises customers who have bought apparel products labeled as “made in China” at least once in the West Bank and later regretted their purchase. Out of non-probability sample methods, convenience sampling was used. The sample size was set to 350. However, because 51 surveys were not completed correctly, the study was done using data from 299 questionnaires.

The questionnaire had 22 items based on the four key study constructs. The study was conducted online among respondents over the age of eighteen who have purchased “made in China” clothing goods at least once. The digital version of the instrument was created with Google Forms. The Google Form link was shared on Facebook and WhatsApp, and responders were notified via email. All respondents participated voluntarily, and no incentives were offered.

The questionnaire form includes demographic questions as well as scale questions designed to help establish the model. To assess demographic factors, the first section includes questions on gender, age, education, and living location. The second chapter has five questions to assess negative feelings, four questions to assess expectations, seven questions to assess regret, and six questions to assess buying decisions. The questionnaire was developed from three research studies, namely Bui et al. (2011), Creyer and Ross (1999), and Simmonds et al. (2017). These studies were chosen because they provide a comprehensive understanding of the factors influencing consumer behavior in the context of regret. The statements used to evaluate the questions in the second chapter were transformed to a balanced five-point Likert scale and administered to participants on a scale of “1 = strongly disagree” to “5 = strongly agree.”

Originally, the scales were in English. Thus, they were translated into Arabic with the help of a group of academics who are specialists in related subjects. To eliminate variations caused by translations, questions translated into Arabic were retranslated into English by academics with expertise in their respective disciplines to ensure consistency. Following these steps, the scale was accepted for use in the questionnaire. Additionally, AI tools such as Chat-GPT were used to detect language errors and mistakes.

A structured questionnaire was adapted from three research papers, i.e. Simmonds et al. (2017), Creyer and Ross (1999), and Bui et al. (2011). The questions have been further modified to suit the study sampling criteria. The questions have been measured by a 5-Likert-type scale, ranging from “1 = strongly disagree” to “5 = strongly agree”. Each questionnaire contained 26 questions for the dependent and independent variables, whereas the first section, which contained four questions, was designed to gather information about the demographics.

Sample criteria and respondents profile

The study employed a convenience sampling approach and targeted Palestinian consumers. At the end of the data-gathering process, 299 valid questionnaires were gathered. A total of 50.2% of the study sample were females. Most respondents were young persons under 39, comprising 60.2% of the sample. Furthermore, the respondents were found to be highly educated, with 65.9% having an undergraduate degree and 34.1% having a postgraduate degree. The majority of the respondents (54.2%) lived in Palestinian towns, while 34.4% lived in villages, and 6.7% lived in refugee camps.

Partial least squares (PLS-SEM)

In this part, we used structural equation modeling to analyze the research model. We assessed the reliability and validity of the measurement and structural models, as well as the study hypothesis and best-fit model.

Measurement model analysis

The measurement model will be evaluated in terms of internal consistency reliability, discriminant validity, and convergent validity.

Cross loadings and indicators reliability

After reviewing the data in Table 2, we can infer that indicators have greater values for their related latent variables than other constructs. Thus, the requirement for cross-loading is met, and it may be stated that the construct indicators are different. This ensures that the indicators in each construct correspond to the assigned latent variable and support the model’s discriminant validity.

Indicators with loadings less than 0.700 were removed from the analysis if their removal improved the values of convergent validity, AVE, and composite reliability; as a result of this role, all of the remaining factors have loadings that vary between 0.536 (Q21) to 0.951 (Q19), and the majority of the factors have loadings greater than 0.700, as shown in Table 2. Thus, unidimensionality is obtained.

Internal consistency reliability

This study measured internal consistency using composite reliability (CR). CR recognizes that indicators have distinct loadings (Henseler et al., 2015). Table 1 shows that every single one of the indicators has a sufficient reliability value of at least 0.700 (Hair et al., 2017). As a consequence, these findings show that the constructs of EXP, INB, NEF, and REG have adequate internal consistency reliability.

Convergent validity

A regularly employed criterion for convergent validity is the average variance extracted (AVE) introduced by Fornell and Larcker (1981). An AVE value of at least 0.5 suggests acceptable convergent validity (Hair et al., 2011).

The total of the research variables scores meet the convergent validity AVE >0.50 requirements, as shown in Table 1. This shows that every variable in the measurement model can clarify at least 50% of the variation of its indicators on average. Thus, convergent validity is verified.

Discriminant validity

Discriminant validity addresses the extent to which the measurements of different constructs are distinct from each other. The Fornell-Larcker criterion and HTMT test will examine discriminant validity (Hair et al., 2011).

The Fornell-Larcker criterion test demands that a latent variable exhibit greater variance with its allocated indicators than any other latent variable (Hair et al., 2011). The Fornell–Larcker criterion test findings, shown in Table 2, show that discriminant validity has been verified since each latent variable AVE value was larger than the maximum squared correlation with any other latent variable.

The HTMT ratio represents the geometric mean of the heterotrait-hetero method correlations divided by the average of the monotrait-hetero method correlations. In a well-fitting model, the HTMT ratio should be less than 1.0. The HTMT explanation is straightforward. If the indicators from both constructs demonstrate an HTMT value lower than one, their underlying correlation is most likely different from one; therefore, they should diverge. There are two methods of applying the HTMT to examine discriminant validity: (1) as a criterion or (2) as a statistical test (Henseler et al., 2015).

Employing the HTMT as a criterion requires contrasting it to a specified threshold of 0.850. If the HTMT value falls below 0.850, it is possible to conclude that discriminant validity is obtained (Henseler et al., 2015). As indicated in Table 3, all of the HTMT.85 values varied between 0.632 (REG → NEF) as the highest value and 0.148 (REG → EXP) as the lowest value, which suggests that discriminant validity is obtained.

Secondly, the HTMT_inference can operate as the basis of a statistical discriminant validity test. The bootstrapping process allows for the development of confidence intervals for the HTMT. A lack of discriminant validity is shown if the confidence interval encompasses the value of one (Henseler et al., 2015). As demonstrated in Table 3, value one falls out of all confidence interval ranges, which implies that discriminant validity is attained.

Collinearity issues

The variance inflation factor (VIF) values of all the indicators in the measurement and structural models are lower than 4, so there are no collinearity problems.

Structural model analysis

Once the measurement model has been validated, the structural model can be analyzed according to its different indices.

The results of R² reveal that the latent variables NEF, EXP, and REG weakly explain 17.6% of the variation in INB. In addition, each of the latent variables of NEF and EXP has a limited ability to explain 23.2% of the variation in REG.

T-statistics of outer loadings (measurement model)

Outer loadings in measurement show the indicator’s relative role in defining its latent variable. When using bootstrapping, the loadings should have statistical significance (p < 0.05). According to Garson (2016), a measurement model becomes more dependable when the path in the model is stronger.

The T-Statistics for the outer lodgings vary from t = 82.227 (Q19→ INB) to t = 2.043 (Q6→ EXP), all of which are bigger than 1.96, indicating the strong significance of the outer model loadings.

Model's capability to predict (Q²)

We adhered to the suggestion of Hair et al. (2011) and Akter et al. (2011) to use cross-validated redundancy since it utilizes the PLS-SEM estimates of the two types of structural and measurement models for predicting data. If the Stone-Geisser’s (Q²) values are larger than zero, it indicates that the structured model accurately predicts the specific endogenous variable being studied (Garson, 2016). According to the values of Q² provided in Table 4, we can infer that the model properly predicts each endogenous latent construct’s indicators and that the model has an acceptable predictive capacity toward the key latent variables.

The measurement model evaluation revealed that CR was larger than 0.8, and AVE showed the average communality for each latent variable with coefficients greater than 0.5. The Fornell-Larcker criteria and cross-loadings were evaluated for discriminant validity. The square root of the AVE for each construct exceeded the construct’s maximum correlation with any other construct. Cross-loadings for all indicators were greater for their related latent variables than for other latent variables. This proved that the indicators in each construct reflected the assigned latent variable, as well as the model’s discriminant validity. As a result, the measurement model was approved, and the structural model was analyzed. The R² values suggested a low percentage of variation explained by the model, indicating potential for both practical and theoretical importance. The model’s prediction performance and fit were both adequate.

Discussion of the Hypothesis's tests

EXP and NEF → REG

According to the T-statistics results shown in Table 5, customer expectations (EXP) were found to have an insignificant (t = 1.061, p = 0.289) positive effect (β = −0.071) on REG, whilst NEF was found to have a significant (t = 9.263, p < 0.001) positive effect (β = 0.480); the model explained 23.2% of REG variance (r² = 0.232) which considered as acceptable explanation.

According to the results regarding REG, we can conclude that Palestinian consumers who have negative feelings toward Chinese clothing products are most likely to feel regret after buying these products, while their previous expectations regarding the product prices and quality will not affect their level of regret. Thus, the study hypothesis H1 is rejected, and H2 is accepted.

EXP, NEF, and REG → INB

The second endogenous variable to be analyzed is INB, which is affected by EXP, NEF, and REG. The t-values imply that NEF has a significant (t = 5.336, p < 0.001) negative direct effect (β = −0.293) on INB and a significant indirect negative effect (NEF - > REG - > INB) (t = 4.124, p < 0.001, β = −0.147); REG also found to has a significant (t = 4.758, p < 0.001) negative effect (β = −0.306). Additionally, the EXP also indicates a significant (t = 2.410, p = 0.016) direct positive (β = 0.147) effect on INB, and the model explains 17.6% of INB variance (r² = 0.176), which is considered a weak but acceptable explanation.

According to the results regarding consumers' intention to buy Chinese clothes, we can conclude that customers' negative feelings and regret feelings negatively affect their intentions, both directly and indirectly, where negative feelings enforce the regret feelings that affect consumers' intentions. Also, it was found that customers' expectations will positively impact the consumer’s intention to buy Chinese clothes products. Thus, the study hypothesis H3 H4, H5, and H5.1 is accepted, while H4.1 is rejected.

Discussion and conclusions

Chinese products are widely found in the Palestinian market. According to the study results, 97.3% of the respondents had already purchased a Chinese clothing product at least once in the past, and 65.9% of the respondents had currently bought a Chinese product. 89.3% of respondents expect the Chinese products to be of lower quality than other products, and 83% of respondents think that their prices will match their quality, but 88.3% of respondents believe that the Chinese products will be of low quality.

Most respondents (80%) had changed their minds about Chinese clothing products after the trial, and only 2.3% trusted their quality. Moreover, after buying Chinese clothing products, respondents experienced negative feelings such as being uncomfortable 78.9% and angry 82.3 and 59.6% of respondents felt that Chinese clothing products did not meet their needs, and 87% of the respondents regretted at least once after buying a Chinese clothing product.

The study results confirm previous research, such as the paper of Karami et al. (2013), which studied the attitude of Iranian consumers toward apparel products made in China and found that they don’t hold positive attitudes toward them. Also, the study by Guo (2013) found that consumers do not favor buying luxury fashion products made in China because of their brand image. Additionally, the study of Shen et al. (2021) studied the Chinese consumers’ perception and purchase intention of “Made in China” products and found that they prefer “Made in the USA” products over “Made in China” products.

The study results support the study’s assumption that negative feelings, such as being uncomfortable and angry and not meeting customer expectations regarding product value, will lead to dissatisfaction and regret.

According to t-statistics, regret is affected by the negative feelings consumers could incur, such as dissatisfaction or discomfort after buying the product. On the other hand, consumer expectations regarding Chinese clothing products do not significantly affect consumers' regret levels, maybe because they already expected to buy a product of inferior quality at a low price. Additionally, negative feelings and regret negatively affected consumers' intention to buy Chinese clothing products, while their expectations positively influenced their decisions. However, the mediation effect of regret was approved in the relationship between negative feelings and the intention to buy.

This result is consistent with the literature. The study by Durmaz et al. (2020) approved the negative relationship between consumers' post-purchase regrets and their intention to buy fashion products. Also, the study of Sai Vijay et al. (2020) approved a significant impact of regret on brand switching and intention to buy. Additionally, the studies of Bai et al. (2017) and Bettiga and Lamberti (2018) found a significant relationship between consumers' emotions and consumers' buying intentions.

Limitations and future studies

The present study suffers from some limitations due to its small sample size, which limits its ability to generalize its results to the whole Palestinian population. However, future research should include other mediators such as age, gender, and income.

Another limitation corresponds to the study variables. The study tested a limited number of variables, while other variables may interfere with and affect consumers' buying intentions, such as brand name, loyalty, and trust. Other psychological variables like motivation, personality, and lifestyle can also influence shoppers' behavior, and hence, these variables can be taken up in future studies.

The current research mainly focused on the Palestinian market. However, future research can consider other markets in the Middle East region since consumers in these markets share several socio-cultural similarities with Palestinian consumers. Also, future research can focus on studying the consumer regret feelings effect on consumers' behavior toward different product categories or from other countries of origin rather than China.

This research mainly focused on investigating regret and its effect on consumer intentions, but it did not investigate how those consumers who experienced regret dealt with their emotions. So, it is highly recommended that future research study the strategies and techniques the consumers apply to cope with and mitigate purchase regret and study its relationship with different marketing strategies and consumer satisfaction.

Figures

Figure 1

The proposed research model

Table 1

Latent variables

NEF		EXP		REG		INB
AVE = 0.705 CR = 0.507		AVE = 0.572 CR = 0.563		AVE = 0.694 CR = 0.508		AVE = 0.693 CR = 0.865
Indicator	Loading	Indicator	Loading	Indicator	Loading	Indicator	Loading
Q1	0.685	Q6	0.641	Q10	0.559	Q18	0.942
Q2	0.661	Q7	0.846	Q11	0.705	Q19	0.951
Q3	0.798			Q12	0.680	Q21	0.536
Q4	0.696			Q13	0.802
				Q14	0.794

Source(s): Own research

Table 2

Fornell Larcker criterion discriminant validity

	EXP	INB	NEF	REG
EXP	0.750
INB	0.134	0.833
NEF	0.046	−0.286	0.712
REG	−0.049	−0.382	0.476	0.713

Source(s): Own research

Table 3

Heterotrait-monotrait ratio of correlations (HTMT)

	Original sample (O)	Sample mean (M)	2.5%	97.5%
INB → EXP	0.314	9583.150	0.125	0.879
NEF → EXP	0.262	4655.136	0.206	0.870
NEF → INB	0.342	0.361	0.230	0.491
REG → EXP	0.148	3798.572	0.132	0.744
REG → INB	0.425	0.444	0.331	0.561
REG → NEF	0.632	0.635	0.500	0.759

Source(s): Own research

Table 4

Model predictive relevance (Q²)

Dependent variables	Q2_ predict	Predictive relevance
INB	0.079	medium
REG	0.208	medium

Source(s): Own research

Table 5

Path coefficients of the structural model and significance testing results

H	Relationship path	β-value	T Value	Significant (p < 0.05)
H1	EXP → REG	−0.071	1.061	0.289
H2	NEF → REG	0.480	9.263	0.000
H3	REG → INB	−0.306	4.758	0.000
H4	EXP → INB	0.147	2.410	0.016
H4.1	EXP → REG → INB	0.022	0.994	0.321
H5	NEF → INB	−0.293	5.336	0.000
H5.1	NEF → REG → INB	−0.147	4.124	0.000

Source(s): Own research