A. Introduction

Cultural Dimensions are presented in the framework proposed by Grett Hofstede (1980, 2001) and classify national cultural variations into 6 dimensions (CFI Team, 2022). This essay will highlight various limitations brought forth by several scholars. The model itself can be disregarded as a misguided pursuit to measure the immeasurable (MacIntyre, 1971). The model can also be compared with newly introduced portrayals of national culture (Schwartz, 1992), instead of considering it as an ideal model for the classification of culture. The model can also be contrasted with richer conceptions of culture (Geertz, 1973; McSweeny, 2002). To accept the claims by Hofstede that categorize something as extensive as culture into just 5 dimensions and apply it successfully across various fields of study is shortsighted. Furthermore, the primary data on which the model is based is collected from employees from a single organization - IBM (McSweeny, 2002). Thus, the diversity across age, race and gender is finite. However, this criticism is defended by Hofstede stating that, what was measured was the differences between national cultures and any other data set of ‘functionally equivalent samples on national populations’ would have provided the same information (Hofstede, 2001). A combination of two surveys, one around 1968-9 and the other around 1971-3 (McSweeny, 2002) was used to base Hofstede’s model. Upon close examination it was discovered that the average representation per country was small and in certain cases it was negligible (McSweeny, 2002). No attempt was also made to include nations that were otherwise neglected due to a lack of representation and a demanding recruitment process on behalf of a multi-national organization. Even among the 66 countries covered, data from only 40 countries was utilized to characterize national cultures (McSweeny, 2002). Also, complete representation is not guaranteed by just considering a large group of respondents (Bryman, 1988). A diverse but focused group of participants would have been suited to determining and categorizing national cultures. Hofstede (1990) also states that individuals were not compared but central tendencies from each country. A statistical average based on individual views from a country was regarded in the formation of cultural dimensions (McSweeny, 2002). This method only considers the prominent responses across each individual from a country and discards the rest of them. Also, the occupational backgrounds of the respondents were limited to the marketing and sales departments (McSweeny, 2002). The absence of diversity is yet another aspect that can be challenged, as data from just one occupational background may or may not be successfully applicable across different industries and multitudes of occupations.

​

B. Academic Analysis

National culture is not just a single culture or the sum of all cultures within a nation, but something that just differs the citizens of one country from the other (Hofstede,1980). Despite the grounds by which a country's population is divided, Hofstede (1980) claims they share a common unique culture, and it is carried out by every citizen. For instance, the United Kingdom comprises of at least 3 nations – England, Scotland, and Wales. Therefore, according to Hofstede, the United Kingdom will be considered as a single nation with one culture (McSweeny, 2002). In addition to this, the cultural identification of each respondent is assumed to be the same as the ‘organizational,’ an ‘occupational,’ and the ‘national’ (McSweeny, 2002). Another assumption made by Hofstede was the uniformity of organizational culture globally. For instance, the organizational culture of a long-standing branch in say Texas was the same as that of the newly established branch in Turkey (McSweeny, 2002). The fundamental flaw with this claim is not that there is a single uniform organizational culture (Risberg, 1999; Parker, 2000) but that there is only one organizational culture. There is no consideration of any other form of culture within the organization - IBM (McSweeny, 2002). However, almost 10 years after the first publication, Hofstede has acknowledged that there is cultural variation within the components of the same organization. This admission of acceptance of multicultural organizations contradicts an essential assumption and undermines the entire national cultural mapping (McSweeny, 2002). The exhaustive literature that contends the acknowledgement of multiple, dissenting, emergent, organic, counter, plural, resisting, incomplete, and contradictory cultures in organizations are effectively ignored (Jelinek et al., 1983; Smircich, 1983; Spender, 1998). Hofstede also postulated that every member within the same occupational category shared the same culture despite being part of the organization (McSweeny, 2002). In this case, the individual cultural identification outweighed the occupational culture. For instance, every German ‘laboratory clerk’ (1980), had the same occupational culture as that of every Bangladeshi laboratory clerk (McSweeny, 2002). Another supposition of Hofstede in terms of organizational culture was that of continuity. The belief that national and occupational cultures are permanent and finished consequences of early socialization (McSweeny, 2002). This assumption is critical as it matches respondents based on their occupation and the answers manifest national cultural values (McSweeny, 2002). Another problematic assumption is that employees of an organization, irrespective of their nationalities, will have the same educational background. Even if it was assumed that they share similar academic qualifications, the content of the course itself differs (McSweeny, 2002). Thus, the values and ideas that shape the perception of culture by means of education are the same for each individual, which is implausible. Another fundamental flaw is that a uniform organizational culture was based on the premise that the respondents never changed occupations outside or within the organization (McSweeny, 2002).

Moreover, Hofstede assumes that regardless of the educational background, entry qualifications, regulations, etc., each member of the same occupation across the world shares the same occupational culture (McSweeny, 2002). For instance, the possibility that a marketer in an organization in the United Kingdom has a similar culture as a marketer in an organization in Malaysia. This is an agreeable assumption that fits perfectly well with Hofstede’s outcomes (McSweeny, 2002). Also, occupational practices are affected by national culture and the diversity of the nation should be enough to create differences in occupational or organizational culture (McSweeny, 2002). Also, if the national culture was the same for a country, there would not be considerable differences between the individuals’ responses (McSweeny, 2002). Despite having acknowledged this shortcoming, Hofstede argues that the responses were an ‘average’ or ‘central tendency’ (McSweeny, 2002). Several varied responses were averaged and converted into a single national response and then compared with other nationally classified data to label the national cultural differences (McSweeny, 2002). Hofstede also maintains that there is not only a single national culture, but that culture can be accurately uncovered based on data from a single organization (McSweeny, 2002). Due to the selective recruitment of IBM employees, mainly the middle class (Hofstede, 1980), the data set only considers individuals’ views from a narrow perspective. Other socio-economic backgrounds are not even taken into account while calculating a ‘central tendency’ (McSweeny, 2002). Lytle et al. (1995) have stated that: ‘Hofstede’s (1980) data was representative of a very limited segment of the overall national population’. Thus, there is no adequate support for the claim that an average tendency within an organization is the same as the national average tendency, by just utilizing data from a single organization (McSweeny, 2002). Furthermore, the construction of the questionnaire was designed to fit with Hofstede's methodology (McSweeny, 2002) and most, if not all, responses would produce notable differences (Schwarz, 1992). The difference would then be used to ‘identify’ and label a particular ‘culture,’ that would fit into the prior classification that formed the data set (McSweeny, 2002). Another possibility must be considered that if Hofstede proposed an entirely distinct set of questions for the survey, then he would have discovered different “national” cultures (McSweeny, 2002). Several attempts have been made to identify national cultures using different questions as compared to Hofstede and that have indeed produced different results (d’Iribarne, 1991; Schwartz, 1992; Lytle et al., 1995). Hofstede even acknowledges and offers no rebuttal that ‘other dimensions in relation to equally important problems of mankind which were not found...as the relevant questions were simply not asked’ (Hofstede, 1980). There are scholars who have identified dimensions other than those introduced by Hofstede. For instance, Schwartz (1994), has introduced seven culture-level dimensions which are, he states, ‘quite different’ than Hofstede. To support the earlier criticism by McSweeney (2002), Robinson (1983), states that the dimensions are a 'hodgepodge’ of items and ‘few of which relate to the intended construct’. Another challenge highlighted is questions about the motives of the respondents. If they were aware of the survey's end purpose, they could have altered their responses for personal gains (promotions) or the betterment of their departments (McSweeny, 2002). Hofstede (1984) states that the responses were under the ownership of IBM and the respondents were aware that their managers expected the development of corrective strategies. He also claims that some surveys were completed in groups rather than by individuals (Hofstede, 1984). Prior knowledge of a survey whose responses will be used to implement corrective strategies can be seen as an incentive to answer in an acceptable manner as opposed to answering in a truthful manner. The composition of the dimensions by Hofstede is also argued by Dorfman and Howell (1988). They argue that the Uncertainty Avoidance Index is composed of three items that reflect disparate constructs: ‘level of perceived stress, length of time the individual believes s/he will work for the present company - IBM and beliefs regarding whether rules should be broken.’ It’s problematic as it attempts to measure non-interactive values (Dorfman and Howell, 1988). On the other hand, the literature by Schwartz (1992), among others, points to the ‘dynamic relations among values' (p.47) instead of values that are categorized into four (later five) ‘independent dimensions (Hofstede, 1984; McSweeny, 2002).

The dimensions are bi-polar; each of them is characterized to be diametrically opposite of the other. For example, the dimension of ‘individualism’ and ‘collectivism’ are treated as opposite poles (McSweeny, 2002). The co-existence between the two dimensions is blatantly ignored. The suggestion by Triandis (1994) that the two dimensions can coexist and sometimes some are emphasized, depending on the situation, is also overlooked. He also states that everyone carries both individualist and collectivist tendencies. For instance, most Catholics believe in the contracting beliefs of ‘free will’ and the ‘will of God.’ Hofstede’s dimensions deny such co-existence and is therefore blind to crucial cultural qualities (McSweeny, 2002). To confidently classify a national culture into highly contrasting dimensions without considering the existence of both together or in some cases, in between (McSweeny, 2002), raises serious concerns. The issue of whether there is an equal understanding of the meaning of specific expressions by respondents is also questioned (McSweeny, 2002). As noted by (Schwartz, 1994), the comparisons are insignificant if there is no equivalence of meaning and since Hofstede did not address this issue, the extent to which the dimensions are equivalent across cultures is unknown (Schwartz, 1994). Also, the acceptance and application of Hofstede’s model is implausible as it equates nation with culture (Baskerville, 2003). Also, culture cannot be divided into component systems or even as different values quantitatively. Instead, it is viewed as a combination of symbols and meanings (Baskerville, 2003). An endeavour to replicate the original Hofstede’s study on accounting research by Chanchani (1998) on two populations – India and New Zealand, has failed to prove that the indices can actually be replicated (Baskerville, 2003). The analysis revealed that the rankings of three out of five dimensions were opposite to that of Hofstede (Chanchani, 1998). One of the dimensions of Hofstede’s model – the Power Distance Index (PDI), focuses on the degree of acceptance of inequality in a relationship, particularly between a boss and an employee (Alion, 2008). Alion (2008) claims that the questionnaire was manipulated to use Western culture as a comparison for non-Western culture and was selective in its representation of inequality within the West. Also, the questions do not capture racial or colonial inequalities (Alion, 2008). Additionally, according to Hofstede, hierarchical inequality is something “we inevitably find” (1980). The choices in the questionnaire were therefore already based on the premise of universal inevitability and the respondents were not asked if they too agreed with this notion but were forced to answer as it was
(Alion, 2008).

C. Practical Implementation

Despite the harsh criticisms and problematic approach of Hofstede’s methodology, his model is still applicable and widely accepted across various industries. This essay will now consider the use of Hofstede's classification of cultural dimension and its use by corporations in the field of advertising by undertaking an example of a fast-moving consumer good (FMCG), by means of video adverts from two countries – Sri Lanka and Canada. The product that is considered is Nestlé's Nescafé Gold coffee. Relevant screenshots are attached to strengthen and visualize the difference between the two advertisements. Making use of the country comparison tool on Hofstede Insights (Hofstede Insights, 2022) which accurately assigns a score for each of the dimensions proposed by Hofstede (1980, 2001), a bar chart is produced as shown in Figure 1, (Hofstede Insights, 2022). As seen in Figure 1, (Hofstede Insights, 2022) the second dimension – individualism/collectivism, the difference between the scores of the two countries – Canada and Sri Lanka, is more than 20 units, with 80 and 35 units respectively. A country's individualism score is the degree to which interdependence is maintained by members of the society (Hofstede Insights, 2022).