The effect of consumer ethnocentrism on the acceptance of imported grocery products in Harare, Zimbabwe

Abstract The study investigated the effect of consumer ethnocentrism on the acceptance of imported grocery products in Harare, Zimbabwe. Using a cross-sectional survey of 505 consumers, independent-samples T-test, analysis of variance and structural equation modelling were used to test research hypotheses. Results show that gender, age and education significantly influence consumer ethnocentrism. Income was found to have an insignificant influence on consumer ethnocentrism. Results also show that consumer ethnocentrism had a negative effect on consumer attitude towards imported grocery products. Consumer attitude was found to have a positive effect on consumer intention to buy imported grocery products. Similarly, consumer ethnocentrism was found to negatively influence consumer’s intention to purchase imported grocery products while consumer’s intention was found to positively influence consumer’s actual purchase behaviour towards imported grocery products. These findings have theoretical and practical implications.


PUBLIC INTEREST STATEMENT
Consumers play a vital role in the economic system of a country. They are the ones who pay to consume the goods and services produced. Consumers play a very significant role in driving the sales of a business. A sound understanding of consumer behaviour is therefore imperative as it provides insights to marketers on crafting effective marketing strategies. This study explored the effect of consumer ethnocentrism on the acceptance of imported grocery products in Harare, Zimbabwe. The study established a negative effect of consumer ethnocentrism on the acceptance of imported grocery products. As such, marketers intending to enter the Zimbabwean grocery products market should seriously consider possible ways of dealing with the acceptability of foreign grocery products.

Introduction
Globalisation has intensified the movement of various grocery products across national boundaries. As such, international trade has gained momentum thereby exposing consumers worldwide to various products (Makanyeza & du Toit, 2017). Increased globalisation has resulted in intense competitive pressure in the retail sector the world over (Das Nair, 2018). Nickanor et al. (2020) posit that African markets are increasingly opening up for foreign grocery products in a bid to meet the increasing demands of their growing populations.
The need to survive in this intense competitive atmosphere has motivated firms to increase their attention towards understanding consumer behaviour, paying special attention to consumer ethnocentrism (Casado-Aranda et al., 2020;Vuong and Giao 2020;Zeren et al. 2020). A complete understanding of consumer behaviour is of paramount importance because it helps marketers in designing effective marketing strategies (Makanyeza & du Toit, 2017). Consumer ethnocentrism refers to the beliefs held by consumers on whether it is appropriate or not to buy foreign-made products (Akbarov, 2021;Das & Mukherjee, 2019;Karoui & Khemakhem, 2019;Shimp & Sharma, 1987). Balabanis and Siamagka (2017) posit that consumer ethnocentrism resembles a biased preference for local products at the expense of foreign alternatives.
According to Mlambo (2017), Zimbabwe was once the most industrialised state in sub-Saharan Africa after South Africa and was a chief exporter of manufactured products to its neighbouring countries. Since the beginning of the new millennium, Zimbabwe's economy has been on a downward trajectory, characterised by chronic unemployment, high inflation and rising national debt levels (Makaye & Mapuva, 2017;Mlambo, 2017). This has resulted in imports flooding the retail market thereby negatively affecting the country's balance of payment (BOP) (African Development Bank Group, 2018). The proliferation of imported grocery products like rice, cooking oil, pasta, cereals, dairy, frozen foods, meat and toiletries among other products in the Zimbabwean grocery market has become worrisome. In the global environment, governments can set inward-looking policies to promote local business). In 2011, the government of Zimbabwe launched the "buy Zimbabwe" campaign in a bid to resuscitate the performance of the local companies through spearheading the production and consumption of domestic products (Ndlovu et al., 2017). The government of Zimbabwe tried to regulate the grocery supply side. However, little attention was accorded to the consumers' voice.
Several studies have been conducted on consumer ethnocentrism in both developed and developing countries (Balabanis & Siamagka, 2017;Casado-Aranda et al., 2020;Karoui & Khemakhem, 2019;Maison & Maliszewski, 2016;Makanyeza & du Toit, 2017;C. Pentz et al., 2017;Shimp & Sharma, 1987;Vuong and Khanh Giao 2020). However, the study of consumer ethnocentrism in the Southern African Development Community (SADC) region is still in its infancy. There are relatively few studies that have focused on consumer ethnocentrism and its influence on consumer behaviour. More so, these few studies that have been conducted in the SADC region have recommended that further studies on consumer ethnocentrism be conducted in this region to have a better understanding of this concept (C. D. Pentz et al., 2014;John & Brady, 2011a, 2011bMakanyeza & du Toit, 2017). The main reasons for calls for further consumer ethnocentrism studies include the fact that consumer ethnocentrism varies from country to country and over time. It also varies with the product category. Consumer ethnocentrism cannot be generalised across product categories and from country to country. Consumer tastes and preferences need constant monitoring since they tend to change over time and consumers are dynamic (Akbarov, 2021).
Consequently, the question of consumer ethnocentrism with regards to groceries in Zimbabwe remains unanswered given government policies like "Look East Policy" and the "Zimbabwe is Open for Business" mantra. Consumer ethnocentrism needs further study as it applies to grocery products in developing economies such as Zimbabwe. Although similar studies were conducted on consumer ethnocentrism in Zimbabwe, the focus was only on one product category i.e., imported poultry meat products (Makanyeza, 2014;Makanyeza anddu Toit 2016, 2017). This makes it pertinent to conduct this study to understand whether or not the influence of consumer ethnocentrism on consumer behaviour varies with the product category and over some time as consumer preferences change. Likewise, a study of this magnitude was imperative to enhance the understanding of consumer ethnocentrism. Therefore, this study strengthens the existing body of consumer ethnocentrism knowledge by investigating the effect of consumer ethnocentrism on the acceptance of imported grocery products in Harare, Zimbabwe. The study focuses on a basket of grocery products instead of one product category. The findings of this study provide insights to marketers on designing effective marketing strategies.
The objectives of this study were: (i) to determine the effect of consumer demographics (gender, age, education and income) on consumer ethnocentrism; (ii) to establish the effect of consumer ethnocentrism on consumer attitude towards imported grocery products; (iii) to determine the effect of consumer attitude on consumer intention to buy imported grocery products; (iv) to determine the effect of consumer ethnocentrism on consumer intention towards imported grocery products; and (v) to determine the effect of consumer intention on actual purchase behaviour towards imported grocery products. Balabanis and Siamagka (2017) define consumer ethnocentrism as the beliefs held by consumers about the suitability and morality of purchasing domestic products at the same time rejecting foreign-made products. That is consumer ethnocentrism is the tendency of consumers to prefer locally produced products to foreign products. In the same vein, Agarwal (2020) describe consumer ethnocentrism as a tendency for people to favour locally made products. Ningsih et al. (2019) view consumer ethnocentrism as the inclination of consumers to view and assume one's cultural group as the best. De Nisco et al. (2016) assert that consumers who have strong ethnocentric beliefs tend to evaluate foreign products negatively than those who are less ethnocentric. Consumers who have strong ethnocentric tendencies avoid purchasing foreign-made products although they are of superior quality to local products (De Nisco et al., 2016). Karunaratne and Wanninayake (2019) posit that consumer ethnocentric appeals can predict consumers' inclinations to purchase locally made products instead of imports.

Consumer attitude, intention and purchase behavior
Consumer attitude can be defined as an individual's internal assessment of an object such as a branded product (Akroush & Al-Debei, 2015;Ningsih et al., 2019;Salem, 2016). Zhang et al. (2021) define consumer attitude as a psychological tendency that is expressed by evaluating a particular object with some degree of favour or disfavour. Consumer attitude reveals the psychological predisposition of consumers to like or dislike products (Hung et al., 2016). Aprile et al. (2016) posit that consumers who show negative attitudes toward imported products tend to exhibit the highest willingness to buy locally made products.
Consumer attitude is a multidimensional concept involving a constant interface of emotions, behaviour and beliefs (Ladero, Casquet, and Singh 2015). Zhang et al. (2021) put forth that the basic model of consumer attitude is called the tri-component or the ABC model encompassing three components, namely affect (emotions), beliefs (cognition) and conation (behavioural intention). The cognitive element represents consumers' beliefs, individual's information and knowledge about an object (Ladero, Casquet, and Singh 2015). The affective element of consumer attitude reflects an individual's feelings and emotions with regards to an attitude object (Buchanan et al., 2016).
Conation refers to the behavioural intention which can be termed consumer intention or purchase intention. The conative element can be regarded as one's willingness to buy a certain product or an intention to behave in a certain way (Makanyeza & du Toit, 2017). Salem (2016) views consumer intention as a course of action that one anticipates to follow. Consumer intention is a consumer's conscious plan to purchase or repurchase a product (Makanyeza et al., 2021).
Behaviour is one's noticeable response to a given situation about a given target (Wee et al., 2014). Purchase behaviour is a stage in the consumer decision making process where the consumer makes a purchase (Schiffman et al., 2014). A clear understanding of purchase behaviour is essential and it enables marketers to identify the appropriate marketing strategy (Rana & Paul, 2017).

Consumer demographics
Various studies have been conducted to determine the influence of consumer demographic variables on consumer ethnocentrism (Agarwal, 2020; Hall & Sevim, 2015;Lopez & Zunjur, 2016). As such, a sound understanding of demographic variables is crucial in consumer behaviour studies (Akbarov, 2021;Das & Saha, 2017;Makanyeza & du Toit, 2017;Nguyen, 2017). The effect of consumer ethnocentrism towards imported grocery products varies from individual to individual depending on demographic and socio-economic factors (Agarwal 2020). Gender, age, education and income are the most commonly measured demographic variables in consumer ethnocentrism researches (Akbarov, 2021;Hall & Sevim, 2015;Seitz & Roosen, 2015).

Gender
Gender refers to a set of features distinguishing males from females (Kwok et al., 2016). Theobald et al. (2017) define gender as the socially constructed roles, behaviours, activities and characteristics that a society considers appropriate for men and women. Kwok et al. (2016) conceive that males and females can differ in personalities, attitudes and activities which can influence consumer behaviour. Matić (2013) posits that women are more conservative about upholding social harmony, and positive feelings among the general public. Hall and Sevim (2015) put forth that older consumers tend to exhibit higher levels of ethnocentric tendencies than their young counterparts. Agarwal (2020) established that age is significantly positively related to consumer ethnocentric tendencies. Vadhanavisala (2015) posits that the positive relationship between age and consumer ethnocentrism is based on an increased cosmopolitan lifestyle among the younger population.

Education
Previous researches indicate that individuals become less conservative as their education levels increase and this increases their biases towards foreign products and brands (C. D. Pentz et al., 2014;Vadhanavisala, 2015). Matić (2013) puts forth that less-educated consumers tend to have a stronger sense of awareness of belonging thus favouring domestic over imported products.

Income
There is a correlation between income and consumer purchase behaviour (Hall & Sevim, 2015). Higher-income levels provide consumers with more travel opportunities thereby exposing them to various foreign products (Hall & Sevim, 2015). C. D. Pentz et al. (2014) perceive that the underlying rationale for this phenomenon is that as consumers achieve higher income levels, they tend to become less conservative and patriotic, thus are likely to favour imports over local products. Shankarmahesh (2006) suggest that gender affects consumer ethnocentrism. In a study conducted in Turkey, females were found to be more consumer ethnocentric than males (Balabanis et al., 2001). In the same vein, C. D. Pentz et al. (2014) established that white South African women exhibit greater ethnocentric tendencies than white South African men. There is a general agreement among various scholars that women have higher ethnocentric scores than their male counterparts (Bruning, 1997;C. D. Pentz et al., 2014;Han & Terpstra, 1988;Sharma et al., 1995;Vida & Fairhurst, 1999a). However, Ramsaran-Fowdar (2010) found results conflicting with other scholars whereby Mauritian males were found to be more consumer ethnocentric as compared to their female counterparts. In a study conducted in India, Jain and Jain (2013) established that consumer ethnocentrism was not influenced by gender. Based on this discussion, it was, therefore, hypothesised that:

Development of research hypotheses and research model
H1a: Consumer ethnocentrism is higher in female than male consumers. Hamelin et al. (2011) found a positive relationship between age and consumer ethnocentrism in a study conducted in Morocco. Similarly, Balabanis et al. (2001) found older consumers to be more ethnocentric in Turkey compared to their young counterparts. In a study conducted in India by Jain and Jain (2013), a positive relationship was established between age and consumer ethnocentrism. Bizumic (2019) found that consumer ethnocentrism is strongly correlated with age and older people were found to be more consumer ethnocentric than younger consumers. In support of this, Agarwal (2020) established that age has a significant positive effect on consumer ethnocentric tendencies. Other various studies have shown that older people are more consumer ethnocentric than younger people (Das & Saha, 2017;De Ruyter et al., 1998;C. D. Pentz et al., 2014;Ramsaran-Fowdar, 2010;Sharma et al., 1995). In light of this discussion, it was hypothesised that: H1b: Age has a positive effect on consumer ethnocentrism.
Several studies confirm a negative relationship between educational level and consumer ethnocentrism (Caruana, 1996;Klein & Ettensoe, 1999;De Ruyter et al., 1998;Sharma et al., 1995). In support of this, Javalgi et al. (2005) assert that there is a negative relationship between consumers' educational level and consumer ethnocentrism. Das and Saha (2017) maintain that the level of education is a significant predictor of consumer ethnocentrism. However, C. D. Pentz et al. (2014) found contradicting results whereby white South Africans' educational level did not influence consumer ethnocentrism. Similarly, in a study conducted in India, Jain and Jain (2013) found no significant relationship between educational level and consumer ethnocentrism. In light of the foregoing discussion, it was, therefore, hypothesised that: H1c: Education has a negative effect on consumer ethnocentrism. A negative association between income level and consumer ethnocentrism was noted in various studies (Bruning, 1997;Sharma et al., 1995;Supphellen & Rittenburg, 2001). C. D. Pentz et al. (2014) found a negative relationship between income and consumer ethnocentrism for white South African consumers. However, in a study conducted in Slovenia income did not significantly influence consumer ethnocentrism (Vida & Damjan, 2001). It was, therefore, posited that: H1d: Income has a negative effect on consumer ethnocentrism.
Several studies have established that consumer ethnocentrism leads to negative consumer attitude towards imported products (Bandara & Miloslava, 2012;John & Brady, 2011a;Makanyeza and du Toit 2016;C. Pentz et al., 2017;Shankarmahesh, 2006;Sharma et al., 1995;Vida & Damjan, 2001). In a study conducted in Zimbabwe, consumer ethnocentrism was found to negatively influence consumer attitude towards imported poultry meat products (Makanyeza & du Toit, 2017). Drawing insights from the foregoing discussion, it was plausible to hypothesise that: H2: Consumer ethnocentrism has a negative effect on consumer attitude towards imported grocery products.
The consensus among scholars is that consumer attitude towards a product has a positive influence on consumer intention to purchase that particular product (Makanyeza & du Toit, 2017;Ondang, 2015;Rana & Paul, 2017;Schiffman et al., 2014;Teng & Wang, 2015). In the same vein, consumer attitude was shown to have a strong positive relationship with consumer intention in Croatia (Jeger et al., 2014). A study conducted by Tsang et al. (2011) in Taiwan established that consumer attitude positively correlates to consumer intention to receive mobile advertisements. This is consistent with the Theory of Planned Behaviour (TPB) which entails that consumer attitude is a key determinant of consumer intentions (Ajzen, 1985;Salem, 2016;Teng & Wang, 2015). Akbarov (2021) established that consumer attitude has a positive effect on consumer intention to purchase genetically modified foods. It was, therefore, hypothesised that: H3: Consumer attitude has a positive effect on consumer intention to buy imported grocery products.
A study conducted in Malaysia revealed that consumer ethnocentrism is the most important determinant of consumer purchase intention (Tabassi, 2012). Klein et al. (1998) found a direct negative relationship between consumer ethnocentrism and consumer intention to purchase foreign products. Similarly, Nguyen et al. (2008) established that a negative relationship exists between consumer ethnocentrism and consumer intention to purchase imported products in Vietnam. It was, therefore, plausible to hypothesise that: H4: Consumer ethnocentrism has a negative effect on consumer intention to purchase imported grocery products.
The relationship between consumer intention and actual purchase behaviour was found to be positive and significant (Wee et al., 2014). This is consistent with the Theory of Reasoned Action which suggests that consumer intention determines the performance of a specific behaviour (Ajzen, 1985;Quevedo-Silva et al., 2016;Teng & Wang, 2015). Various studies reported a positive and significant correlation between consumer intention and actual purchase behaviour (Ardekani et al., 2020;Follows & Jobber, 2000;Morwitz & Schmittlein, 1992;Wee et al., 2014). Based on the preceding discussion, it was, therefore, hypothesised that: H5: Consumer intention has a positive effect on actual purchase behaviour towards imported grocery products.
Based on the preceding discussion on the hypothesised relationships, the research model is presented in Figure 1.

Questionnaire design and measures
Data were collected using a structured questionnaire. The questionnaire consisted of five sections. The first section focused on consumer ethnocentrism. The second section focused on consumer attitude towards imported grocery products. This section mainly required respondents to indicate their feelings, emotions and beliefs concerning imported groceries. Consumer intention towards imported grocery products comprised the third section of the questionnaire. Respondents were required to indicate their behavioural intentions and willingness towards purchasing imported groceries. The fourth section required respondents to rate their actual purchase behaviour towards imported grocery products. The fifth and last section of the instrument focused on the demographic information of the respondents. This section considered respondents' gender, age, educational level and income level. Sources of items used in instrument design are presented in Table 1. The items were modified to suit the study at hand. Five-point Likert scale questions were used to measure consumers' responses. Response points were as follows: 1-strongly disagree; 2-disagree; 3-neither agree nor disagree; 4-agree; 5-strongly agree.

Sampling and data collection
The target population for the study comprised consumers in Harare, Zimbabwe. Data were collected from a sample of 600 consumers using a cross-sectional survey design. Harare was selected because it is Zimbabwe's largest city and provides the largest pool of consumers in the country. Consumers were randomly intercepted as they came out of ten major supermarkets in the central business district in Harare. The structured questionnaire was interviewer-administered. Out of the 600 questionnaires distributed, 505 were returned and usable.
The demographic profile of the sample is presented in Table 2. It can be observed in Table 2 that females (54.7%) slightly dominated the sample as compared to their male (45.3%) counterparts. The majority (85.2%) of the consumers surveyed fell into the age group 20 to 49 years. The results indicate that the bulk (73.1%) of the respondents had attained tertiary education. In terms of income, the majority (79.20%) of the surveyed consumers earned less than RTGS $1,499 per month.

Scale validation
Measurement model fit indices, standardised factor loadings, individual item reliabilities (squared multiple correlations), critical ratios, composite reliability and average variance extracted were used to assess convergent validity. Results presented in Table 3 show that convergent validity was

Sources of items
Consumer ethnocentrism CET1. Zimbabwean people should always buy Zimbabwean-made grocery products instead of imports Makanyeza & du Toit, 2017;Sharma, 2015;Sharma et al., 1995;Shimp & Sharma, 1987 CET2. Only those grocery products not available in Zimbabwe should be imported CET3. Buy Zimbabwean-made grocery products; keep Zimbabwe working CET4. Zimbabwean grocery products first, last and foremost CET5. Purchasing imported grocery products is un-Zimbabwean CET6. It is not right to purchase foreign-made grocery products CET7. A real Zimbabwean should always buy Zimbabwean-produced grocery products CET8. We should purchase grocery products made in Zimbabwe instead of letting other countries get rich off us CET9. It is always best to purchase Zimbabwean-made grocery products CET10. There should be very little trading or purchasing of grocery products from other countries unless out of necessity CET11. Zimbabweans should not buy foreign grocery products because this hurts Zimbabwean business and causes unemployment CET12. Curbs should be put on all grocery imports CET13. It may cost me in the long run, but I prefer to support Zimbabwean grocery products CET14. Foreigners should not be allowed to put their grocery products on our markets CET15. Foreign grocery products should be taxed heavily to reduce their entry into Zimbabwe CET16. We should buy from foreign countries only those grocery products that we cannot obtain within our own country CET17. Zimbabwean consumers who purchase grocery products made in other countries are responsible for putting their fellow Zimbabweans out of work (Continued)

Constructs
Items and codes

Sources of items
Consumer attitude towards imported grocery products CAI1. Imported grocery products are good for our health Balasubramanian & Kamakura, 1989;Burton et al., 1998;Makanyeza & du Toit, 2016;Makanyeza & du Toit, 2017 CAI2. Imported grocery products are of high quality CAI3. Imported grocery products are affordable CAI4. Imported grocery products are good CAI5. Imported grocery products are favourable CAI6. Imported grocery products are attractive Consumer intention towards imported grocery products CII1. I intend to buy imported grocery products the next time I make a purchase Hosein, 2015;Vermeir & Verbeke, 2006 CII2. I am likely to buy imported grocery products whenever I buy groceries CII3. I will definitely recommend others to buy imported grocery products CII4. I would not mind buying imported grocery products next time CII5. I plan to purchase imported grocery products on a regular basis CII6. I intend to consume imported grocery products whenever they are available Actual purchase behaviour towards imported grocery products API1. I often buy imported grocery products Ismail & Mokhtar, 2017;Wee et al., 2014 API2. I purchase imported grocery products on a regular basis API3. I always purchase imported grocery products if they are available API4. I frequently buy imported grocery products achieved. The measurement model was estimated using the Maximum Likelihood Estimation (MLE) to achieve more robust parameter estimates. Measurement model fit indices considered consist of CMIN/DF (χ2/Df), Goodness of fit index (GFI), Adjusted GFI (AGFI), Normed Fit Index (NFI), Tucker-Lewis Index (TLI), Comparative fit index (CFI) and Root mean square error of approximation (RMSEA). Table 3 shows that the initial measurement model had less acceptable fit indices. To improve the model, items with poor (<0.4) squared multiple correlations (SMC), namely CET2, CET3,  CET10, CET13, CET14, CET15, CET16, CAI1 and API3 were deleted (Bagozzi & Yi, 1988). The modified measurement model subsequently exhibited acceptable fit indices as illustrated in Table 3. Kline (1998) posit that for a model to be accepted; χ2/DF should be less than 3, and GFI, AGFI, NFI, TLI and CFI should be close to 1 while RMSEA should be less than 0.07 (Hooper et al., 2008).   Table 4 presents standardised factor loadings (λ), SMC, critical ratios (CRs) and composite reliabilities (CRα) used to determine convergent validity. All SMCs were above the minimum cut-off point of 0.4 as set by Bagozzi and Yi (1988) while CRα for all the constructs surpassed the minimum cut-off point of 0.7 (Kim et al., 2013;Segars, 1997). Results presented in Table 4 indicate that all λ for the items were above the required minimum benchmark of 0.6 as proposed by Bagozzi and Yi (1988). CRs were satisfactorily large (CR>2) and significant at p < 0.001 (Reisinger & Mavondo, 2007).
To assess discriminant validity, average variance extracted (AVEs) were compared against squared inter-construct correlations (SICs). AVEs were greater than their corresponding SICs (Fornell & Larcker, 1981). Moreover, Table 5 shows that all AVEs for all the measured constructs were higher than the required minimum of 0.5 (Fornell & Larcker, 1981). Two conditions must be satisfied when assessing discriminant validity (Dadzie et al., 2012;Kim et al., 2013). Firstly, AVE for each latent construct should be greater than the corresponding SICs. Secondly, each AVE should be greater than 0.5. Therefore, the conditions for discriminant validity were achieved.

Testing research hypotheses
H1a was tested using an Independent-Samples T-test. C. D. Pentz et al. (2014) posit that an independent sample t-test is used to assess the statistical significance of the difference between two sample means. Results of the test of differences between means indicate that there is a significant difference (p 0.01) in consumer ethnocentrism between men (M = 3.117, SD = 1.013) and women (M = 3.433, SD = 0.991) [t = −3.524; df = 481.859]. This implies that women are more consumer ethnocentric than their male counterparts. Thus, H1a was supported.
One-Way Analysis of Variance (ANOVA) was used to test H1b, H1c, and H1d. Results show that the mean scores for consumer ethnocentrism increased with age ( . This shows that as consumers get old, they become more consumer ethnocentric as compared to their younger counterparts. Therefore, H1b was supported. In terms of education, results show that the mean scores for consumer ethnocentrism decreased with consumers' level of education (primary education M = 3.781, SD = 1.122; secondary education M = 3.573, SD = 0.902; tertiary education M = 3.176, SD = 1.019; F = 9.212, p < 0.001, DF =2). This implies that as people become more educated; the less consumer ethnocentric they become. Therefore, H1c was supported.
Results show that there were no significant differences in the mean scores of consumer ethnocentr-   Table 6 show that consumer ethnocentrism has a negative effect on consumer attitude towards imported grocery products; consumer attitude has a positive effect on consumer intention towards imported grocery products; consumer ethnocentrism has a negative effect on consumer intention towards imported grocery products, and consumer intention has a positive effect on actual purchase behaviour towards imported grocery products. Therefore, H2, H3, H4 and H5 were supported.

Discussion
The study sought to determine the effect of consumer demographics (gender, age, education and income) on consumer ethnocentrism; to establish the effect of consumer ethnocentrism on consumer attitude towards imported grocery products; to determine the effect of consumer attitude on consumer intention to buy imported grocery products; to determine the effect of consumer ethnocentrism on consumer intention towards imported grocery products; and to determine the effect of consumer intention on actual purchase behaviour towards imported grocery products. The study found that consumer ethnocentrism is higher in women than male consumers, younger consumers are less ethnocentric as compared to older consumers and higher educated consumers tend to be less ethnocentric than the less educated consumers. However, the findings show that consumers' income does not have a significant influence on the level of consumer ethnocentrism. The study also established that consumer ethnocentrism has a negative effect on consumer attitude towards imported grocery products, consumer attitude has a positive effect on consumer intention towards imported grocery products, consumer ethnocentrism has a negative effect on consumer intention towards imported grocery products and consumer intention has a positive effect on actual purchase behaviour towards imported grocery products. By and large, the findings suggest that consumer ethnocentrism has a negative effect on the acceptance of imported grocery products. This is consistent with Casado-Aranda et al. (2020) who indicated that ethnocentric consumers tend to prefer domestic to foreign products.
In terms of demographic variables considered; gender, age and education were found to significantly influence consumer ethnocentrism. This is consistent with previous studies (Bruning, 1997;C. D. Pentz et al., 2014;Sharma et al., 1995;Supphellen & Rittenburg, 2001). However, the study established that consumers' income does not have a significant influence on the level of consumer ethnocentrism. This contradicts the general understanding in literature which suggests that as consumers' income level increases, the less ethnocentric they become (Bandara & Miloslava, 2012;John & Brady, 2011a;Shankarmahesh, 2006;Vida & Damjan, 2001). It is probable that consumers in Zimbabwe, a developing country, might not have much choice on the foreignness or localness of grocery products given their low incomes. This finding supports the views of Jain and Jain (2013) who found that income does not influence consumer ethnocentrism in India.
The findings of this study have implications for theory, practice and future research.

Theoretical implications
Various studies have been conducted on consumer ethnocentrism both in developed and developing countries. However, there is a dearth of studies that have examined the effect of consumer ethnocentrism on the acceptance of imported grocery products in SADC countries like Zimbabwe. This study was, therefore, conducted to close this knowledge gap and contribute to the extant literature. The findings suggest that consumer ethnocentrism has a negative effect on the acceptance of imported grocery products. This implies that, given an option, ethnocentric consumers will not buy imported grocery products. Instead, they will buy locally made grocery products. This substantiates the theoretical underpinning that ethnocentric consumers tend to prefer domestic to foreign grocery products (Casado-Aranda et al., 2020;Jeger et al., 2014;Ondang, 2015;Teng & Wang, 2015;Wee et al., 2014). Similarly, the study shows that the effect of consumer ethnocentrism on consumer attitude, purchase intention and actual purchase behaviour seems to be consistent across product categories and countries. Considering prior research on consumer ethnocentrism in Zimbabwe (Makanyeza & Chikazhe, 2017;, the findings suggest that consumer ethnocentrism and its influence on consumer behaviour is relatively stable in the Zimbabwean environment.

Practical implications
The need to get an understanding of the effect of consumer ethnocentrism on the acceptance of imported grocery products was the motivating force behind this study. A thorough comprehension of studies of this nature is imperative to marketers as it helps them in crafting appropriate marketing strategies. One major takeaway from this study is the notion that consumer ethnocentrism and its influence on consumer behaviour is relatively stable in the Zimbabwean environment. This brings some degree of predictability when it comes to designing marketing strategies that address consumer ethnocentrism.
Gender, age and education were found to significantly influence consumer ethnocentrism. As such, marketers are advised to consider these demographic variables when designing marketing programmes. Marketers can segment their imported grocery product markets based on these demographic factors.
The findings established a negative effect of consumer ethnocentrism on the acceptance of imported grocery products. As such, marketers intending to enter the Zimbabwean grocery products should seriously consider possible ways of dealing with the acceptability of foreign grocery products. One such strategy is to consider collaborating with local Zimbabwean suppliers instead of going it alone so that the imported grocery products can be viewed as a part of the local products. Local grocery producers can also take advantage and emphasise localness when marketing their domestically produced grocery products.

Limitations and implications on future research
The study was confined to Harare, the capital city of Zimbabwe. This is likely to raise generalisability issues. It is, therefore, recommended that future research be conducted in other cities in Zimbabwe and other emerging economies to improve the generalisation of the findings. Further research can be done for specific grocery items to compare and contrast with the general grocery findings.