Like any other cultural elements or dimensions in organisation theory, Ubuntu has been conceptualised and measured in various ways: several conceptualisations of Ubuntu and what comprises its antecedents of compassion, survival, solidarity and respect and dignity (Broodryk 2005; Khoza 2005; Mbigi & Maree 1995; Tutu 1995). Ubuntu has begun to gain prominence in a myriad of academic fields and discourses, both in South Africa and beyond in knowledge domains as varied as law, media, theology, education, sport, servant leadership and public policy ethics among others (House et al. 2004; McAllister 2009; Qobo & Nyathi 2016; Richardson 2008). Mbigi (2000) and Nussbaum (2003) perceive Ubuntu as an expression of an African view of the life world anchored in Africa’s own persons, cultures and societies, which is difficult to define in a Western context. Common to all the conceptualisations of Ubuntu found in the literature is a link with South Africa’s Zulu-Xhosa aphorism, Umntu-ngumntungabantu which means that ‘each individual’s humanity is expressed in relations with others’ (Battle 1996:99). Simply put this means, ‘a person is a person through other people’ (McAllister 2009:2; Poovan et al. 2006:17). This latter aphorism according to Mangaliso (2001) conveys the notion that a person becomes a person only through his or her relationship with, and recognition by, others. Tutu (2004) gives credence to this particular Ubuntu aphorism by pointing out that it is one unique concept that Africa can offer to the world. Therefore, this article argues that the contribution that recognises the value of Ubuntu to service management knowledge is through its extension of existing theories of individual-collectivism cultural dimensions for directing high service quality team performance. More practically, Ubuntu supports group solidarity in team work and, in turn, a collective achievement orientation by employees to attaining organisational goals. This aphorism suggests that Ubuntu is a phenomenon that can promote common understanding of work as a collective output between a manager and his work teams by promoting that as a team, they help and care for each other as they would members of one family or community as suggested by (Tutu 2004).

The Ubuntu aphorism is further adopted widely by southern African Nguni speaking people having equivalent phrases extending beyond South Africa to other sub-Saharan African countries. Broodryk (2005) reflects that the concept of Ubuntu is expressed in many different African languages, always emphasising a relationship with collectivism as a community. In Kenya, the concept of Ubuntu is talked about as ‘Utu’ referring to the idea that every individual’s action should be performed for the benefit of the whole community (Broodryk 2005:219; Oppenheim 2012:370). In Uganda, the concept is known as Obuntu-bulamu translating to an individual’s human generosity, and harmonious interaction with, one’s own community (Oppenheim 2012:370). Mangaliso (2001:24) defines Ubuntu as humaneness – a pervasive spirit of caring and community, harmony and hospitality, respect and responsiveness – that individuals and groups display for one another. Mangaliso (2001) goes further to suggest that Ubuntu is invariably invoked as a scale for weighing good versus bad, right versus wrong and just versus unjust. The most prevalent understanding of Ubuntu in the literature is when it is considered ‘an African humanness which is characterised by compassion or caring, sharing, communist democracy and related predispositions’ (Khoza 2005:269). Khoza’s view was taken up by Broodryk (2005:13) who described Ubuntu as a ‘comprehensive, ancient African worldview which pursues primary values of intense humanness, caring, sharing and compassion, and associated values’. In public spaces, Ubuntu is perhaps best represented by public figures who regard Ubuntu as the essence of being human, which extends beyond African ways of life to include hospitality, kindness, warmth, welcoming, generosity and caring about others (Dandala 1996; Mthembu 1996; Tutu 1995). The article therefore asserts that, in the context of sub-Saharan Africa, communism (also referred to as collectivism) becomes understood as having group meaning expressed through Ubuntu (Broodryk 2005).

More important than language similarities, however, are to consider the differences between the various conceptualisations of Ubuntu. These differences involve the varying social ethic characteristics reflected in Ubuntu as a mode of conduct by individuals in contemporary life in terms of an individual reflecting antecedent traits such as hospitality, sharing, generosity and selflessness, that is, the behaviours that are expected to results from Ubuntu (Louw 2002; Mangaliso 2001; McAllister 2009; Theletsane 2012). Linked with such expected behaviours is group solidarity which is crucial to the survival of African communities who have historically survived through group care by the community for the community (Mbigi & Maree 1995). In this sense, Ubuntu proponents (Battle 1996; Ndaba 1994; Sono 1994) seek to increase understanding that the ideal value of Ubuntu is to connote a compromise by encouraging an exclusion of oppressive communalism but rather allowing a person to grow and prosper in a group relational setting by providing ongoing contact and interaction with others. There is a lack of consensus on the definition coupled with the scarcity of psychometrically robust measurement scales (Brubaker 2013). With the exception of limited empirical investigations (Brubaker 2013; Mabovula 2011; Mbhele 2015; Sigger et al. 2010), relatively little attention has been given to the development of measurement scales for Ubuntu that conform closely to the conceptualisation of the Ubuntu concept. Sigger et al. (2010) is a study that operationalised Ubuntu in terms of business management. Other studies tended to focus on leadership alone and not the role of collectivism (manager and workers) (Mangaliso 2001; Mbigi 1997, Mbigi & Maree 1995; Msila 2008; Nussbaum 2003; Poovan et al. 2006). Even though Sigger et al. (2010) need to be acknowledged in the development of measurement scales for Ubuntu, their stated scales reveal several limitations. Sigger et al. themselves (2010) acknowledged that their proposed items for the four Ubuntu elements met with a lack of understanding of the meaning of these items by respondents. Without firstly the development of a consensus on what Ubuntu is conceptually, based on the identification of the antecedents for each of the four Ubuntu elements, reliable scales cannot be constructed. Louw (2002:8) refers to Ubuntu as ‘an account of innovative constructions, which, is inevitably coloured by our (post)modern values, beliefs and biases’. Louw (2002) argues that any attempt to answer the question, ‘what is Ubuntu?’ would be misguided, positing that the more important question is: ‘How should Ubuntu be understood and utilised for the common good of all Africans, and of the world at large?’ (Louw 2002:8). These questions underpin establishing a consensus about the composition of Ubuntu in the context of business management practice.

This article focussed on the four elements of compassion, survival, group solidarity and respect and dignity as principles in common with other studies (Brubaker 2013; Mangaliso 2001; Poovan et al. 2006; Sigger et al. 2010). These elements and their antecedents as complemented by many authors are now discussed individually.

The first step involved the collection of data by interviewing 25 purposefully selected (15 South African hospitality frontline managers and 10 South African hospitality academic lecturers or instructors) as research participants, and analysing this data. South African hotel frontline managers were chosen as they are characterised by their role in interfacing through management practice or training with frontline employees ensuring consistent service quality delivery for customers (welcoming strangers). South African academics teaching and training in South African universities were chosen as they have insight into the usefulness of Ubuntu as a management style in a South African hotel context. Criteria for selection for this research were individuals knowledgeable as follows:

Initiating permission for interviews to be held required that contact be first made with the head office of each university or hotel group. The hotel chains approached for the research were wholly South African owned (to ensure employee service training reflected South African management aspects and employee cultural diversity in service teams) with a presence in each of the nine provinces (to ensure diversity in opinions of managers on the role of Ubuntu in their management styles). Two South African hotel groups agreed to be involved, four hotels for the first and three hotels for the second hotel group. Lecturers from four South African public universities agreed to be involved.

Invitation letters to participate were sent via email. The types of frontline manager in a hotel are broad, so to gather diverse perceptions, managers from front office, housekeeping, restaurants and banqueting or conferencing were approached, that is, front office (five participants), food and beverage (seven participants) and housekeeping (three participants). Esterberg (2002) recommends a sample minimum size of 12 as adequate for generating themes for exploratory analysis. The one-on-one interviews using a semi-structured interview schedule that interrogated the value in management of the four principal Ubuntu themes in hospitality service management were tape recorded and transcribed verbatim and content analysis conducted. The content analysis conducted on the transcribed interviews (Figure 2) established additional 16 items that described the four-key Ubuntu concepts. These items found support in the Ubuntu scale development literature (Brubaker 2013; Sigger et al. 2010; Strauss et al. 2016).

Application of a Delphi technique assists researchers to arrive at ‘effective decisions about the data being interpreted in situations that present insufficient information for scale development’ (Hasson, Keeney & McKenna 2000:1008). A Delphi technique can be viewed as a ‘method for consensus-building’ among a group of ‘experts or knowledgeable participants’ (Van Dun, Hicks & Wilderom 2016:4). The Delphi survey was implemented to review the trustworthiness of 16 proposed Ubuntu item measures derived from stage 1 interviews. The decision about who to invite to participate entailed a purposeful selection of participants with expert knowledge of Ubuntu (Hasson et al. 2000). Eight purposefully selected academic participants were invited to provide their expert input on the proposed 16 scale items. The Delphi study participants included six men and two women. Three of them were Xhosa speaking, two Ndebele, one Zulu and two English speaking thereby eliciting a wide knowledge base. Five participants through their upbringing and associated culture understood the concepts of Ubuntu and three had a good theoretical knowledge of Ubuntu. Employing a sample of between six is not rare in Delphi studies. Previous research (Gustafson et al. 1973) used a sample size of four participants involving two rounds; Nambisan, Agarwal and Tanniru (1999:374) with six research participants with three rounds. Accordingly, the period recommended for the participants to complete the iterative rounds of Delphi study ranges between 1 and 3 months (Okoli & Pawlowski 2004:10). Round 1 was to determine individual opinions of each of the 16 scale items from the eight participants (Hasson et al. 2000:1011). Participants were requested to judge individual concept items in terms of their ability to measure the four constructs of Ubuntu in management practice. Hasson et al. (2000) draw attention to the detail that in applying the Delphi survey-consensus method, constitute collective group consensus from the participants should be in the range 51 % – 80 %. An average of 68 % – 82 % group consensus was attained through each of rounds 1–3, as illustrated in Figure 3.

Two of the eight participants did not return their responses for inclusion in the analysis of rounds 2 and 3, resulting in a lower but acceptable response rate. In the second round, participants were requested to rate each of the 16 item statements (amended from round 1 feedback) and rate an additional 10 items added from participant round 1 feedback. Six of the 10 items corresponded with the collectivism element of Ubuntu identified by Hofstede (1983) and House et al. (2004). The 10 additional items were among the measures added to the original 16 items that participants felt should be split into individual items. The intent in this round was to assess the extent of agreement (consensus on the measurement scales) among the participants, and to resolve any further disagreements (consensus development) (Jones & Hunter 1995:377). These participants provided input and opinions finally agreeing with the pool of 26 Ubuntu item measures in round 3.

Approaches proposed by Hinkin (1995) and King (2004) were next applied to design the Ubuntu questionnaire. Accordingly, Hinkin (1995), Sigger et al. (2010) and Strauss et al. (2016) recommend that scale development uses a theoretical definition of the phenomena of interest (Ubuntu four concepts), applying an established typology (management) to guide item generation. These authors further note that in the absence of sufficient theory to guide the development of a measure, an inductive approach should first be used based on a sample of participants who can provide descriptions of the phenomena principles. This has been applied in this research design in stages 1 and 2 already described. The resulting items are as follows:

Ubuntu – Survival (four items) – At this hotel …

Ubuntu – Respect and dignity (four items) – At this hotel …

Ubuntu – Group solidarity (six items) – At this hotel …

Ubuntu – Compassion (six items) – At this hotel …

Ubuntu – Collectivism (six items) – At this hotel …

Defining Ubuntu in a manner that allows for its measurement has to date been limited to a focus on the Ubuntu four principles (compassion, survival, group solidarity and respect and dignity), ignoring the role of collectivism integral to each. This article firstly synthesises various concepts of Ubuntu by earlier academic studies creating a conceptual consolidation of past research. Then, following the lead of House et al. (2004), this article has argued that organisational productivity requirements by managers such as assuring service quality excellence should consider the positive implications of Ubuntu in the context of service management. Thirdly, this article highlights the critical role of acknowledging and encouraging collectivism in attaining an Ubuntu style business management. Finally, the proposed measurement scale provides researchers with a fresh perspective of what is relevant in a scale to measure the degree of Ubuntu adopted in management scenarios.

It is acknowledged that it may not be easy to implement Ubuntu as a management style as some of this research’s participants noted that Ubuntu may be perceived as a culture that supports only a certain group of people, or that it may be difficult to implement an ancient wisdom like Ubuntu in a modern society. These issues aside many aspects of the discussion in this article highlight that Ubuntu has a role to play in service organisations especially those in South Africa.