Background

In 2016, a car and truck dealership with branches in several states of the country requested the design of a test for the selection of salespeople due to an unsatisfactory staff turnover situation and because they wanted a better option than the tests that AMEDIRH (^{2021, Spanish: Asociación Mexicana de directores de Recursos Humanos}) offers to its clients, the Terman (^{Society of Applied Psychology, 2000}), the Cleaver (^{Society of Applied Psychology, 2000}) and the PIPG (^{Gordon, 1994}). Therefore, the objectives of the research are to identify the characteristics of successful salespeople and to design a valid and reliable test for their selection.

Tests to predict sales success include the inventory and profile of the seller or IPV(^{Manual Moderno, 2016}) and the 16 Personality Traits Test or 16 PF (^{Catell & Mead, 2008}). The SPI manual claims that the test has defined 10 salesperson personality traits and, through extensive studies, is valid and reliable. However, it does not offer evidence of its predictive validity. Similarly, the 16 PF does not provide evidence of predictive validity, but it does identify two traits associated with sales: the O- (serene adaptation) and the M- (objectivity).

^{Arias (2018)} paints a “bleak” picture of the use made of tests in Mexico and probably in other countries of the continent. It is common for those responsible for personnel selection to use tests without checking whether they are reliable, valid, and appropriate for the particular positions that require the right personnel. Arias comments that it is common to use tests designed in other regions, at different times, and in another language, and that those responsible for recruitment and selection blindly trust in their predictive power. Consequently, this author recommends previous validity, reliability, and standardization studies and considers other techniques, such as interviews and work samples. Thus, the objectives that users and designers of new tests are expected to achieve are to master contemporary theoretical principles and psychometric concepts, to recognize common misconceptions that lead to misuse of tests and misinterpretation of test results, and to illustrate how seemingly complex and sophisticated concepts and methodologies can be understood and in easily understandable terms. Statistical topics, ranging from standard deviation to factor analysis, are illustrated in the article.

^{Arias (2018)} paints a “bleak” picture of the use of tests in Mexico and probably in other countries on the continent. It is common for those responsible for recruiting personnel to use tests without verifying whether they are reliable, valid, and appropriate for the particular positions that require the right personnel. Arias comments that it is common to use tests designed in other regions, at different times, and in another language, by those responsible for recruitment and selection who blindly trust their predictive power. Consequently, this author recommends prior studies of validity, reliability, and standardization and considers other techniques, such as interviews and work samples. Thus, the objectives that users and designers of new tests are expected to achieve are to master contemporary theoretical principles and psychometric concepts, recognize common misconceptions that lead to the misuse of tests and misinterpretation of test results, and to illustrate how seemingly complex and sophisticated concepts and methodologies can be understood in easily understandable terms. The article illustrates statistical topics ranging from standard deviation to factor analysis.

The test user can only adequately evaluate a test if he or she is familiar with the main steps of test construction and is aware of the main psychometric characteristics of tests, especially as they relate to norms, reliability and validity. By test user, in contrast to test constructor, the authors mean essentially anyone who has the responsibility of choosing tests or interpreting scores and using them as a source of information for making practical personnel selection decisions. Test users (test applicators) include teachers, students, human resource managers or human capital managers in the private or government sectors, consultants and researchers.

Many current criticisms of testing are directed at the tests and the misuse of tests by unqualified users. Management professors, human resources, psychometrics, and organizational psychology professors can play a dual role in helping to disseminate accurate information about tests and thus combat misuse and misinterpretation. First, they can contribute through the knowledge they pass on to their students in their classes; second, they can serve as specialists in answering special questions that are sure to be asked of them by organizations, users, colleagues, and their students. Therefore, teachers, consultants, and human resources or human capital managers, whatever their fields of expertise, need to master knowledge about the technical properties of tests, in particular about:

Many current criticisms of tests is directed at tests and their misuse by unskilled users. Management, human resources, psychometrics, and organizational psychology professors can play a dual role in helping to disseminate accurate information about tests and thereby combat misuse and misinterpretation. First, they can contribute knowledge that they pass on to their students in their classes; second, they can serve as specialists in answering special questions that are likely to be asked by organizations, users, colleagues, and their students. Therefore, professors, consultants, and human resources or human capital managers, whatever their fields of expertise, need to master knowledge about the technical properties of tests, in particular about:

Theoretical considerations

The tests that are commonly applied in personnel selection have mostly been constructed from an inductive and empirical perspective, supported by the classical test theory or CTT (^{Nunnally, 1978}), which starts from the elaboration of item banks and the content and statistical analysis, as occurs in the cases of the inventory and profile of the salesperson or IPV (^{Manual Moderno, 2016}), the big five questionnaires or BFQ (^{Barbaranelli & Caprara, 2000}), Gordon’s personality profile and inventory (^{Gordon, 1994}), the Minnesota Multiphasic Personality Inventory or MMPI (^{Zambrano-Guerrero, Caicedo-Padilla, and Matabanchoy-Tulcán, 2015}), except for some cases derived from a theory such as the Myer-Briggs typological inventory test or MBTI (^{McCaulley & Martin, 1995}). The factors identified by statistical analysis are called constructs or latent variables and give rise to theoretical propositions. The test in this paper also derived from a qualitative exploratory, inductive approach, and the classical test theory is a model that establishes as a first step the identification of behaviors, attitudes, and beliefs associated with previously observed characteristics, such as the case of successful salespeople.

The tests that are commonly applied in personnel selection have mostly been constructed from an inductive and empirical perspective, supported by the classical theory of tests or CTT (^{Nunnally, 1978}), which is based on the elaboration of item banks and content and statistical analysis, as is the case of the inventory and profile of the seller or IPV (^{Manual Moderno, 2016}), the big five questionnaires or BFQ (^{Barbaranelli & Caprara, 2000}), the Gordon personality profile and inventory (^{Gordon, 1994}), the Minnesota Multiphasic Personality Inventory or MMPI (^{Zambrano-Guerrero, Caicedo-Padilla and Matabanchoy-Tulcán, 2015}), except for some cases derived from a theory such as the Myer-Briggs typological inventory test or MBTI (^{McCaulley & Martin, 1995}). The factors identified through statistical analysis are called constructs or latent variables and give rise to theoretical propositions. The evidence in this work also derives from an exploratory, inductive and qualitative approach, and the classical theory of evidence is a model that establishes as a first step the identification of behaviors, attitudes and beliefs associated with previously observed characteristics, as is the case of successful salespeople.

^{Novak (1966)} introduced the classical theory of tests, but Spearman, in 1904, was the first to decipher the influence of measurement error and propose the calculation of reliability. Later, ^{Lord and Novick (1968)}, ^{Nunnally (1978)} and ^{Allen and Yen (2002)} explained the theory in detail.

In summary, the test design was based on an exploratory and inductive approach and was conducted in accordance with classical test theory. In the initial stages, no particular theory was considered, and in the final stage, the relationship between 15 items was examined and the emerging construct was proposed to be named “optimism.” However, now it is time to describe the research methodology followed for the design of the salesperson test, in accordance with classical test theory.

Research design

This research is mixed since the data were collected through individual interviews and focus groups, examination of job applications and application of a standardized online questionnaire; consequently, it uses a qualitative phenomenological technique and a correlational study that seeks to establish the relationship between variables.

Procedure and time-space framework

Eight stages are presented, which began in 2016 and concluded in 2018, and were carried out in the distributor's offices and agencies in Mexico, Querétaro and Hidalgo. The initial contact was the human resources director, who requested advice to design a valid test to select vehicle sellers and address the high level of voluntary staff turnover.

The study was conducted in eight stages:

Stage 1 (2016, January): Familiarization with the talent acquisition process through a focus group with the human resources department.

Stage 2 (2016, February and April): Focus groups and interviews with salespeople, managers, and executives in the State of Mexico, Querétaro, and Hidalgo.

The qualitative method was based on conducting unstructured individual interviews and seven focus groups (2016) with 5 managers, 8 human resources representatives, 13 agency managers, and 41 salespeople working in the corporate offices and 7 agencies. Records (job applications) of the 100 best salespeople in terms of commissions and the 100 worst salespeople were also analyzed to discover possible demographic characteristics that differentiate successful salespeople.

Stage 3 (2016, May): Evaluation of the instruments used to select sales representatives. The validity of the tests applied in 2015 and 2016 (Cleaver, Terman, and PIPG) in the selection of 147 sales representatives was analyzed, and it was found that their correlation with average daily sales commission is not relevant. This finding confirms the need to design a customized test and conduct the study.

Stage 4 (2016, June): Design of the screening instrument according to the 11 constructs identified in the qualitative research.

Stage 5 (2016, July): Administration of the online test to active salespeople.

Based on the findings of the qualitative method, the quantitative method consisted of implementing a standardized questionnaire with 11 factors and 63 items.

Stage 6 (2016, August): Analysis of the tool’s effectiveness in terms of reliability and validity.

Stage 7 (2016, September to September 2018): Administration of the 63-item test to new candidates.

Stage 8 (2018, October): confirmation of the validity of the test 2 years after its use (sample of 415 salespeople hired in 2016, 2017, and 2018).

Internal locus of control

In 1966, Rotter proposed the construct that stems from the assertion that successful people, and consequently successful salespeople, have control over the process and closing of the sale. It depends on them rather than on circumstances. This trait describes people who perceive that most achievements and events depend on their behavior, skills, and knowledge; internal locus of control is a belief that the person is in control of his or her life and therefore attributes his or her achievements and failures to himself or herself. On the other hand, people characterized by an external locus of control hold that success and failure are the result of luck or external factors beyond their control. An example of a statement item is: Most of the sad situations that occur in people's lives are due, in part, to bad luck.

Self-efficacy

Interviews and focus groups reveal that a successful salesperson is one who considers to be competent and possesses the necessary skills to sell. The concept of self-efficacy expectancy (^{Bandura, 1977}) refers to the belief in one's ability to adequately manage specific goals and life stressors. Self-efficacy expectancy can influence feelings, thoughts, and actions. People with low expectations tend to display low self-esteem and negative feelings about their abilities. Regarding thoughts, perceived self-efficacy facilitates cognitions regarding one's own abilities, and these thoughts act as motivators for action. An example of a formulated item is: I can find a way to get what I want, even if someone opposes it.

Self-monitoring

Interviewees affirm that successful salespeople correctly perceive customer expectations and know how to act flexibly and consistently with their expectations. Self-monitoring is a personality trait that refers to the ability to regulate behavior to adapt to social situations (^{Snyder, 1974}; ^{Snyder & Gangestad, 1986}). Individuals who closely monitor themselves are classified as high self-monitors and often behave in a manner that is highly responsive to social cues and their situational context. Those with high self-monitoring can be considered social pragmatists who project images to impress others and receive positive feedback. In contrast, those with low self-monitoring do not engage in expressive monitoring to the same degree nor do they share a similar concern for situational relevance.

People with low self-monitoring tend to exhibit expressive control consistent with their internal states, beliefs, attitudes, and dispositions, regardless of social circumstances. They tend to be less observant of the social context and consider expressing an image different from their internal states to be undesirable falsehood. Therefore, they are unwilling to self-monitor or adjust their behavior accordingly. They are often aggressive, stubborn, and insistent with others. An example of a formulated statement is: I find it difficult to imitate the behavior of others.

Resilience

Interviews and focus groups reveal that a successful salesperson is a person who views himself or herself as competent and has the skills necessary to sell. The concept of self-efficacy expectancy (^{Bandura, 1977}) refers to the confidence in one's abilities to adequately handle specific goals and life stressors. Self-efficacy expectancy can influence feelings as well as thoughts and actions. People with low expectations tend to show low self-esteem and negative feelings about their abilities. Regarding thoughts, the perception of self-efficacy facilitates cognitions that refer to one's abilities, and these thoughts act as motivators for action. An example of a formulated item is: I can find a way to get what I want, even if someone opposes me.

Extraversion

Salespeople have been described as sociable and social, and in some cases have become friends to such an extent that their clients invite them to meetings, parties, birthdays, and funerals. Extraversion is a typical trait characterized by concentration of interest on an external object. Extroverts are interested in the external world of people and things; they try to be more friendly and aware of what is going on around them. They are cordial, sociable, assertive, active, and thrill-seeking (^{Costa & McCrae, 1992}). In this context, they are called dynamic, active, dominant, bold, go-getters, the opposite of underachievers. Extroverts can be described by their dynamism and dominance, where dynamism refers to energetic and dynamic behaviors, while dominance describes the ability to assert oneself. An example of a statement item is: I like meeting and dealing with new people.

Open-mindedness

Another trait that emerges from the interviews and focus groups is the flexibility to help customers with diverse characteristics and attitudes, even customers who appear to lack sufficient resources to purchase a vehicle because of their way of dressing and speaking. People who score high in this category are cultured, informed, interested in new things and experiences, and open to contact with diverse cultures (^{Costa & McCrae, 1992}). Openness to culture indicates an interest in reading, acquiring knowledge, and staying informed. Likewise, open-mindedness refers to a positive attitude toward new things, perspectives, cultures, styles, values, and ways of life. People who score low in open-mindedness are considered closed to experience and tend to be conventional in their attitude and behavior. They prefer familiar routines to new experiences and generally have a narrower range of interests. Openness has moderate positive relationships with creativity, intelligence, and knowledge. An example of an item is: Everything new excites me.

Emotional stabilityxxx

Interviews and focus groups indicated that a good salesperson is a keen observer who inquires about customers' needs and desires and avoids closing a sale impulsively. Emotional stability is a trait related to the control of states of tension that arise from emotional experiences and the ability to maintain control even in circumstances of conflict, danger, or discomfort (^{Costa & McCrae, 1992}). Emotional stability refers to a person's ability to remain stable and balanced. At the other end of the scale, a person with a low level of stability tends to experience negative emotions easily. An example of a formulated statement is: I worry and feel very nervous in emergencies.

Conscientiousness or responsibility

In interviews, successful salespeople are described as tenacious and persistent, taking charge of the process, beginning with an inquiry into the customer's needs and interests, proposing and demonstrating appropriate vehicles, proposing different purchase plans, and closing the sale or subsequent follow-up. There are also anecdotes from salespeople who have sold to grandparents, parents, and children, providing reminders about when maintenance is due, checkups, payment of maintenance fees, and offers on new vehicle models. The trait of conscientiousness can be defined by conscientiousness and perseverance. The former consists of meticulousness, love of order, and reliability; the latter refers to the conscientiousness and persistence with which people perform their tasks (^{Costa & McCrae, 1992}). An example of a stated item is: I tend to leave most of my activities to the last minute.

Achievement motivation

One of the primary motivations of salespeople is achieving monthly vehicle sales goals and the associated commissions. Salespeople are aware of their objectives, their own sales, and those of their colleagues. The need for achievement drives the desire to excel and strive for success. People with a predominant need for achievement seek challenging tasks (neither too easy nor too complex), reject gambling, and prefer to be in control (^{McClelland, 1988}). An example of an item is: I like being assigned complex tasks.

Need for power

In the interviews and focus groups, it was stated that the role of salesperson carries prestige with family, friends, and the distributor, since sales work is a difficult skill that few people can master. This need motivates individuals to get others to behave differently from their natural behavior. It is the desire people have to control the means that allow them to direct, influence, and dominate others (^{McClelland, 1988}). An example of an item used is: I like the idea of ​​being a boss.

Need for affiliation

Finally, this need also characterizes successful salespeople, given that it was frequently mentioned in the interviews that a good salesperson is someone who easily establishes social relationships, lasting relationships based on trust and friendship, which they manage to consolidate with the client. The need for affiliation is an impulse that drives a person to relate to others; it is a social need. For ^{McClelland (1988)}, it is a critical need, since people with a predominance of the need for affiliation are characterized by showing great concern for the recognition of others and preferring group work. They are more likely to be assigned teamwork, where maintaining relationships is more important than making decisions and achieving goals. Interest in affiliation is considered important for understanding the needs of others and creating a favorable work environment. Therefore, employees with a high need for affiliation seek satisfaction in social and interpersonal activities. An example of a formulated item is: I like being in the company of those who work with me.

Following the definitions in the previous paragraphs, the eleven scales were designed by writing 63 items, 45 positive and 18 negative (to be recoded), according to the 5-interval Likert-type scale, where 5 is the pole of high appreciation (totally agree).

Demographics

This section describes the characteristics of 453 salespeople hired between 2016 and 2018 in terms of their age (mean 36.9 years and standard deviation 11.9 years), gender, and the brand they were hired for. Table 1 shows that the top three brands (whose data are kept confidential at the distributor's request) account for the largest share of salespeople (85.4%), and that women represent 34.5% of the workforce.

Criterion and factorial validity of the 11 factors

In order to test the predictive validity of the 11 factors measured by 63 items, the test was administered to 453 candidates between 2016 and 2018. The test results were correlated with the salespeople’s average daily commission, which was achieved six or more months after hiring. No correlation achieved significance at the .05 level or lower; therefore, it was decided to perform an exploratory factor analysis to identify associated underlying factors using principal component extraction and varimax rotation.

In Table 2, a main factor was identified in which 15 items load, and the Kaiser-Meyer-Olkin tests of sampling adequacy and Bartlett’s Sphericity obtained satisfactory results of .964 and 4 252.3 (p = .000); the variance explained is 55.4%; the factor loadings of the items range from .898 to .318, according to the order presented in Table 2. It is worth mentioning that four items correspond to emotional stability (ES), three to locus of control (LC), three to extraversion (EX), two to conscientiousness (TE), one to open-mindedness (AM), one to resilience (RE) and one to self-monitoring (AUM). It should be noted that 14 items go in the opposite direction to the measured factor, and only item EE2r.16 positively measures emotional stability. Nonetheless, this item is recorded to add it to the other 14 items and obtain a total score whose minimum possible score is 15 points and the maximum score is 75 points. According to Table 3, the minimum score observed is 15, the maximum score observed is 73, the mean is 48.1, and the standard deviation is 15.5. The reading of the 15 items reveals that due to their interrelation, they refer to a construct called optimism, which is explained below.

To test the predictive validity of the 11 factors measured by 63 items in Spanish, the test was administered to 453 candidates between 2016 and 2018. The test results were correlated with the salespeople's average daily commission, obtained six or more months after hiring. No correlation reached a significance level of 0.05 or lower; therefore, it was decided to conduct an exploratory factor analysis to identify the underlying associated factors using principal components extraction and varimax rotation.

In Table 2, one main factor was identified with a loading of 15 items, and the Kaiser-Meyer-Olkin sampling adequacy test and Bartlett's sphericity test yielded satisfactory results of 0.964 and 4252.3 (p = 0.000); the explained variance was 55.4%. The item factor loadings range from .898 to .318, according to the order presented in Table 2. It is worth mentioning that four items correspond to emotional stability (ES), three to locus of control (LC), three to extroversion (EX), two to conscientiousnes( C ), one to open-mindedness (OM), one to resilience (RE), and one to self-monitoring(SM). It is worth noting that 14 items are contrary to the measured factor, and only item EE2r.16 positively measures emotional stability. However, this item is recorded to add it to the other 14 items and obtain a total score whose minimum possible score is 15 points and the maximum is 75 points. According to Table 3, the minimum score observed is 15, the maximum score observed is 73, the mean is 48.1, and the standard deviation is 15.5. The reading of the 15 items reveals that due to their interrelation, they refer to a construct called optimism, which is explained below.

According to the mean reported in Table 3 and Figure 1, two bell curves are observed, separated by an intersection at 48 points; 48 points is proposed as the cutoff point. Salespeople with low average commissions obtain high scores on the test, and salespeople with high average commissions have low scores on the test. Salespeople with low average daily commissions are located in the right bell, and salespeople with high average daily commissions are in the left bell.

Source: created by the authors

Figure 1 Distribution of the scores obtained in the 15-item Factor 

Nevertheless, the correlations between Factor 1 (15-item test) and average daily commissions are:

In all cases, the correlations are significant (** p < .01), thus confirming that the 15-item test has construct and criterion validity. Figure 2 shows the correlational dispersion between the test score (horizontal axis) and the average daily commission (vertical axis). Each point represents one salesperson or more salespersons: the higher the test score, the lower the commission.

Source: created by the authors

Figure 2 Dispersion of scores obtained in the 15-item test (X-axis) and average daily sales commissions (Y-axis) 

Taking into account Table 3 and Figure 1, Table 4 compares the average daily commissions of salespeople scoring below 48 and above 48. The 243 salespeople who score above 48 points on the 15-item test have an average daily commission of MXN 209.60, and the 172 salespeople who score below 48 points have an average daily commission of MXN 534.80, which is a 2.55 times higher commission gain.

Reliability cronbach’s alpha and standard error of the measure

Regarding reliability, the obtained coefficient of .94 is highly satisfactory (Table 5), and the standard error of measurement (SEM) is 3.8 or 4 rounded points (for the reliability of .94 and the standard deviation of 15.54); two standard errors of measurement (recommended to have a 95% confidence interval) is 7.6 or 8 rounded points of measurement error.

The SEM was calculated using the following formula:

SEM is the standard error of measurement; SD is the Standard Deviation, and r is the reliability coefficient. When calculated, the SEM is expressed in the same units as the test scores. Therefore, if the 2SEM is applied, 8 points should be subtracted and added to the score obtained by a sales candidate to interpret and predict their sales potential.

Norms (percentiles)

Table 6 shows the percentiles corresponding to the scores obtained in the 15-item test. It should be remembered that high scores correspond to low commission performance, so scores of 48 points or less correspond to salespeople who have higher average daily commissions. Another appropriate way to interpret the table is to consider it as a pessimism scale, where an applicant who scores more than 48 points is considered pessimistic, and applicants with 48 points or less are considered optimistic.

Voluntary staff turnover

Table 7 presents the ratio of annual voluntary staff turnover from 2013 to 2018, which excludes departures due to dismissal, disability, retirement, or death and includes departures due to resignation or abandonment of work; the ratios appear in the last column, for example in 2013, 4.3 salespeople resigned or abandoned their job for every salesperson hired. There is a consistent downward trend in the staff turnover ratio, with 2018 being the lowest. The main reasons for the drop are better recruitment measures, which consisted of recruiting at least three applicants per vacancy and applying the new test.

Finally, a LISREL confirmatory factor analysis was performed with a sample of 224 aspiring salespeople to corroborate the findings of the exploratory factor analysis. The statistics and Figure 3 confirm that the fifteen items are grouped into one factor; an acceptable relative Chi-square of 1.93 was obtained since it falls between a rank of 2 to 1; this Chi-square is obtained by dividing the Chi-square of 164.31 by 85 degrees of freedom; the following statistics are also interpreted as acceptable: RMSEA = .06, NFI = .96, CFI = .98 and GFI = .91.

Source: created by the authors

Figure 3 Confirmatory factor analysis. The 15 items are identified with rectangles and the letter P. The numerical order of the items is respected, as presented in Table 2