A generation is established when its members jointly experience a formative and similar event (Weber & Urick, 2017). Millennials are part of a generation that stands out for being known as "the children of technology" (Tapscott, 2008). Although there is no consensus on when this generation begins and ends, Kim (2018) and Saeed et al. (2018) refer to those born between 1981 and 1995. Other authors differ from these years and include even those born in 1999 (Stein & Martin, 2015), or between 1981 and 2001 (Howe & Strauss, 1992). After them, 2001 marks the start of the next generation, known as Generation Z.

In countries such as the United States, it is one of the most diverse generations and also one of the most numerous (Hartigan, 2010). By 2015, Millennials represented 37% of the workforce in Canada (Mahmoud et al., 2020), and estimates indicate that they will constitute 75% of the global working population by 2025 (Stein & Martin, 2015). Hence, companies should understand their characteristics and behavior in order to attract them, as they represent the largest working generation (Kim, 2018). The Millennial generation has several distinctive characteristics that differentiate it from others (Mahmoud et al., 2020). Compared to Generation X and previous ones, Millennials are more extroverted, aware of a globalized world, technological, narcissistic, and self-centered and have more self-esteem (Lyons & Kuron, 2014; Twenge, 2014).

While within this generation there may be certain variations due to individual identities (Joshi et al., 2010), they share common characteristics. First, they are concerned about social causes and are especially interested in them (Stein & Martin, 2015; Abdullah et al., 2022), which is reflected in several aspects of their lives. For example, they buy clothes not only for fashion but also to support social causes (Groysberg & Abbott, 2016; Konstantinou & Jones, 2022), are more socially aware compared to other generations (Salam et al., 2022), and are more informed (Stein & Martin, 2015). Second, they buy and acquire products and services through social networks (Dabija et al., 2017), which have had a strong influence on both their apparel preferences and how they buy (Groysberg & Abbott, 2016). Millennial consumers have a strong preference for the internet (Dawar, 2016) and consider technology to be an essential part of their lives (Lebowitz, 2018).

The third common characteristic is the type of information sources they use. This generation is more informed than previous ones (Stein & Martin, 2015; Nnambooze & Parumasur, 2016), who relied on "traditional" sources, such as print newspapers and television—where the immediacy of the news is not the priority. Their access to technology has generated a fourth distinctive feature: their dependence on smartphones and the use of social networks, which allow them to satisfy their need to post photos, personal news (such as products they like and buy) and share everything that happens in their lives (Sashittal et al., 2015).

Specifically, regarding work, Millennials differ from other generations in multiple aspects. First, they are used to receiving feedback at work more frequently than others (Stein & Martin, 2015). Second, they demand greater flexibility and independence in their work schedules (Stein & Martin, 2015) and vacation time (Twenge et al., 2010). Third, they have an entrepreneurial approach, which is reflected in the great importance they place on business strategies (Pontón Deluquez & Márquez López, 2016). Fourth, they cyberloaf, i.e., they use technology excessively and for personal purposes at work (Lim, 2002). It has been reported that, on an eight-hour workday, they can spend up to two of them using technology for personal reasons (Zakrzewski, 2016). They also waste more than twice as much time as Baby Boomers (Conner, 2013), i.e., those born between 1946 and 1964 (Mahmoud et al., 2020).

The fifth aspect that differentiates Millennials from other generations is that they find it difficult to separate their personal life from their work life. Millennials build social relationships and communicate more easily with friends and strangers online (Thayer & Ray,2006). In addition, they see the use of technology as part of their identity (Taylor, 2014). Sixth, their learning style differs from those of previous generations because they do not receive guidance or structured formal programs anymore (Kim, 2018). In addition, they believe they can access all kinds of information, which is always available just a few clicks away on the computer (Kim, 2018). Seventh, they consider themselves to be multitaskers (Kim, 2018) able to switch their attention between different media (e.g., laptops, smartphones, television) at least 27 times per hour on average, an additional 60% above other generations (Steinberg, 2012). Although other characteristics differentiate this generation from previous ones, those seven mentioned above are considered the most important. Using data collected during the COVID-19 pandemic, Rattanapon et al. (2023) proposed that millennials could be more engaged in an organization when there is greater job fit but lower group involvement. Meanwhile, job fit may convince Gen X workers to stay in their jobs for the long term.

Motivation has been defined in multiple ways. According to Reeve (2005), motivation can be understood as the needs, thoughts, and emotions that guide a person's behavior (see Figure 1). Human motivation is also defined as an emotional state generated in a person as a consequence of the influence that certain motives exert on his or her behavior (Koenes, 1996). According to Moody and Pesut (2006), “Motivation is a values-based, psycho-biologically stimulus-driven inner urge that activates and guides human behavior in response to self, other, and environment, supporting intrinsic satisfaction and leading to the intentional fulfillment of basic human drives, perceived needs, and desired goals” (p. 17). Different variables influence motivation, such as personality or the properties of the perceived environment, which will cause changes in the motivation generated (Martínez, 2001). In general, motivation can be defined as the process of emergence, maintenance, and regulation of acts that produce changes in the environment and that are consistent with certain internal constraints, i.e., plans or programs (Bueno, 1993, cited in Soriano, 2001). According to Ryan and Deci (2000), “To be motivated means to be moved to do something".

Figure 1

Figura 1. Jerarquía de las fuentes de motivación

The Work Motivation Questionnaire (WMQ, known as CMT in Spanish; Toro, 1992) has been used to diagnose motivation at work. It was designed and validated in Colombia, and it has been employed since 1984. It measures 15 variables grouped into three conceptual categories. The validity and reliability of this instrument have been measured in the past (Toro Álvarez, 1998a). There have been five versions of this questionnaire over time to improve it, and the items that are representative of each variable have been standardized. Additionally, considering the psychometric conditions, reliability, and validity of this instrument, it is evident that, with each update, the psychometric indicators and the use of scales have been optimized. The WMQ has been used in different Latin American countries (such as Puerto Rico, Venezuela, Chile, Brazil, and Colombia, among others) in undergraduate and graduate research, business consulting, and scientific articles (Toro, 1991; Álvarez Ramírez, 2012; García Rubiano & Forero Aponte, 2014).

The motivational variables measured by the WMQ (Figure 2) are divided into three conceptual categories or dimensions: internal motivational conditions (intrinsic motivations), preferred means of obtaining desired rewards at work (obtaining rewards), and external motivational conditions (extrinsic motivations) (Toro Álvarez, 1998a; Toro Álvarez, 1998b). Each one of these dimensions contains five variables. Intrinsic motivations include Achievement, Power, Affiliation, Self-Actualization, and Recognition. Obtaining rewards encompasses Dedication to the Task, Acceptance of Authority and Rules, Requisition, and Expectation. Finally, extrinsic motivations are Supervision, Work Group, Job Content, Salary, and Advancement Opportunities. These three dimensions of the WMQ and their variables are detailed below.

Figure 2

Figura 2. Desglose de las categorías conceptuales medidas por el WMQ de Toro

This study implemented the WMQ because it was designed in Latin America and has been validated in several countries in this region. An adapted WMQ was submitted for evaluation and approved by three expert judges in organizational psychology, industrial engineering, and statistics. After the review, the instrument was adjusted, changing the wording in five of its questions. The final form had 33 questions. There were 15 questions about motivation, grouped into three dimensions: Intrinsic Motivations, Obtaining Rewards, and Extrinsic Motivations (Toro, 1992). Each variable was measured using five statements, and there were 18 variables for sociodemographic and employment characterization.

Before the instrument was administered, it was also validated in a pilot test with 30 students from the Universidad Panamericana in Mexico to determine if each one of the questions was adequately understood. Based on this pilot test, it was determined that the questions were well formulated, and the average response time was 40 minutes.

The clustering procedure was applied as a two-step method. First, individuals were grouped into clusters, highlighting the distinctive variables that differentiated them. This step followed the approach presented by Rubiano-Moreno et al. (2019) for ipsative variables. Second, after the clusters had been determined, logistic regression was applied to determine the driving factors that explained the phenotypic differences among groups. The following paragraphs detail these two steps.

The first step employed a method that incorporates the concept of dissimilarity (Sørensen, 1948). First, the optimal number of clusters (k) was determined via the average silhouette method. Afterward, a dissimilarity matrix was obtained in which every element d_ij represented the dissimilarity value between individuals i and j. Subsequently, the centroids of the clusters were assigned as the k individuals with the highest dissimilarity value among them. After that, the remaining n-k individuals were grouped into the cluster that offered the centroid with the lowest dissimilarity value. The procedure stopped when there were no individuals left to assign.

In the second step, after the clusters had been identified, the score obtained in the centroid for each characteristic was used to characterize the profiles considering their sociodemographic variables and, if possible, describe the preferences of each group (Rubiano-Moreno et al., 2019). This way, it was possible to define the profile of each cluster.

Logistic regression analysis was conducted to identify the driving factors in each cluster. Depending on the number of clusters created, logistic regression can be either binary or multinominal. In this regression, the dependent variable (output) was denoted by the cluster to which each individual was assigned (the k-th cluster is selected as the reference value). Meanwhile, the sociodemographic variables and questionnaire answers were treated as predictors.

Logistic regression model:

(1)

Model to be fitted:

(2)

Where πj=P(Yj=1), and xj represents each socio-economic variable. The model in Equation (2) was introduced to identify the socioeconomic variables whose association with the constructed groups was statistically significant. A measure of goodness of fit called Akaike Information Criterion (AIC) ¹ is commonly used to compare the fit between models (the lower the value, the better the fit), but it was not implemented in this case and is only presented here for the sake of completeness.

The sampling was non-probabilistic with two criteria: (1) being part of the Millennial generation and (2) having work experience. The selection criteria did not include aspects related to socioeconomic or education level to broaden the coverage of the study.

The form was uploaded to the Question Pro® platform, which provided a link to fill out the questionnaire. It was administered in Guadalajara, Jalisco, Mexico. Subsequently, the quality of the collected data was evaluated, and a total sample of 197 responses was obtained. Finally, the data were tabulated in Microsoft Excel® for subsequent analysis using an algorithm programmed in C language.

The first step to analyze the data was to calculate the optimal number of clusters that should be used according to the methodology. Figure 3 shows that the optimal number of clusters was two because it maximized the value of the average silhouette, a metric commonly used for clustering methods (Pollard & Van Der Laan, 2002). Subsequently, clustering was performed as shown in Figure 4 and taking into account the 15 that explain the factors that influence workers' motivation, i.e., Achievement, Power, Affiliation, Self-Actualization, Recognition, etc. (Toro Álvarez, 1998a).

Figure 3

Figura 3. Número óptimo de clústeres para los Millennials

The sociodemographic variables classified by cluster in Table 2 show that Cluster 1 concentrates the youngest population. Consequently, they have the lowest number of children, and their main economic support comes from their parents. In the rest of the variables, both clusters exhibited similar characteristics.

According to the information in Tables 3 and 4, Age and Main Breadwinner in the Household are driving factors that can be used to differentiate between the two clusters. In the case of Age, the older the individual, the higher his or her probability of belonging to Cluster 2. On the contrary, high values in Main Breadwinner indicate a lower probability of belonging to Cluster 2 (i.e., a high probability of belonging to Cluster 1).

The radar chart in Figure 4 indicates that the Intrinsic Motivations of Millennials in Cluster 1 are not so much Achievement, Self-Actualization, or Recognition. Instead, they focus on influencing groups or having control over situations through Power, and they like to maintain relationships with people around them to achieve Affiliation with the group. In terms of Obtaining Rewards, they are characterized by dedicating time and effort to the tasks assigned to them, and they care if their performance contributes to the fulfillment of the objectives of their group, always accepting authority. Finally, in relation to Extrinsic Motivations, they focus on participating in collective work where varied tasks can be performed and there is freedom in the way in which activities are carried out within the organization. These characteristics clearly define the "cooperative" profile of Cluster 1. The following paragraph describes the motivational profile of Millennials in Cluster 2.

Contrary to the previous group, Cluster 2 includes the oldest Millennials, which implies that they have the largest number of children and are the main economic support of their families (see Tables 2, 3, and 4). With respect to their Intrinsic Motivations, they continuously seek to improve their skills and abilities within the organization to improve their performance. They also like to lead their work teams to meet goals. Regarding Obtaining Rewards, they have a hard time accepting authority, but they focus on performing their tasks enthusiastically because they expect the organization to notice their effort and reward it. Finally, in relation to Extrinsic Motivations, their main difference with respect to Cluster 1 is that they are more interested in the economic retribution they get for doing their job. These characteristics make these Millennials more "competitive" than their counterparts in Cluster 1.