Master Thesis
MSc Business Administration
The Role of Categorization in the Application
Process of Elite Art Schools
Fabienne Marie Nigg
11375507
June 2017
Abstract
Categorisation literature states that people rely on categories when assessing and evaluating organisations, products and other individuals. Current research uses this concept to explain why certain applicants in the labour markets have better chances of being hired. Thereby the most common approach of researchers is to distinguish between specialists who concentrate on a single category and generalists who concentrate on different combinations of categories. However, there are different opinions regarding what profile is more favourable. One view argues that applications who demonstrate a specialist profile are preferred. On the contrary, a conflicting view states that applicants who exhibit a generalist profile are favoured. By applying these opposing views, this thesis focuses on the short-term and long-term effects of being categorized as a specialist or generalist by selection committees, in the context of the selection to a visual art program. The empirical setting of this thesis is the Rijksakademie van Beeldende Kunsten in Amsterdam. The data from two application years is used to test the relationship between an applicant’s profile and his or her probability of acceptance and further his or her future artistic ranking. Contrary to this thesis’s expectations, the findings suggest that neither a specialist nor a generalist profile confer specific benefits.
Table of Contents
1. Introduction ... 1
1.1 Purpose and Relevance of the Thesis... 1
1.2 Approach and Structure ... 6
2. Theoretical Framework ... 6
2.1 Categorization of Individuals ... 6
2.2 Specialist Advantage ... 9
2.3 Generalist Advantage ... 12
2.4 The Specialist – Generalist Continuum ... 18
2.5 Categorization and Future Ranking ... 21
3. Research Design ... 26 3.1 Data collection ... 26 3.2 Dependent Variables ... 29 3.3 Independent Variables ... 30 3.4 Control Variables ... 31 3.5 Method ... 33 4. Results ... 33 4.1 Descriptive Statistics ... 33 4.2 Correlations ... 42 4.3 Regressions ... 43 4.4 Alternative Analyses ... 49
5. Discussion... 52
5.1 Findings... 52
5.2 Limitations ... 57
6. Conclusion ... 58
1. Introduction
1.1 Purpose and Relevance of the Thesis
The majority of university applicants face the same nerve-racking procedure: motivational letters are rewritten several times, documents are carefully prepared and the CV is adapted to the requirements. After this careful and time-consuming preparation, the most frustrating moment is when one is receiving the rejection letter. But why did one get rejected? More and more elite art students are asking themselves this question. The numbers of art students applying to renowned institutions is rising and the application processes are becoming more and more selective (Solomon 1999). Googling “how to get accepted to an elite art school“ yields 80 Million results, each one of them containing tips and tricks on how to increase one’s chances of acceptance. However, the secret formula is yet to be found.
The categorization literature stemming from the organisational and sociological research could shed some light on this matter or at least deliver a partial explanation. By categorizing our surroundings, we are able, among other things to identify and assess industries, and organizations and their products (Ruef and Patterson 2009). Categorization also helps people to assess others not only in daily life but also in formal selection processes that can be found in labour markets (Zuckerman et al. 2003, Merluzzi and Phillips 2016) and universities.
A well-established practice among researchers of the categorization literature is to differentiate between generalists and specialists. Although different descriptive terms for these two categories exist, they all describe the same thing: specialists show a strong link to a particular category, whereas generalists engage in more than one category at the same time (Hsu 2006; Jensen and Kim 2014). A more detailed approach stems from the work of Leung (2014), who states that not only do specialists and generalists differ from each other, but
category experiences. As a result, Leung (2014) distinguishes between: incremental generalists, who have gained experiences in related categories and erratic generalists, who demonstrate expertise in various unrelated categories.
Extant literature so far makes contradicting statements regarding whether being a generalist or being a specialist is advantageous. On the one hand, a group of researchers states that the category-spanning performed by generalists has several negative side-effects, ranging from misunderstanding, over non-consideration and repulsion or devaluation (Zuckerman 1999; Smith 2011; Hsu, Hannan and Kocak 2009). Having a specialist profile, especially in the labour market, demonstrates to the employer that the candidate has the qualifications and skills to work in the respective field (Zuckerman et al. 2003). Specialists are thus not only more likely to be hired (Zuckerman et al. 2003) but also receive higher wages (Neale 1995) and have higher chances of promotion (Ferguson and Hasan 2013).
On the other hand, another group of researchers has identified certain circumstances in which demonstrating a generalist profile is actually beneficial (Pontikes 2012, Kovacs and Hannan 2010, Ruef and Patterson 2009, Hannan 2010, Merluzzi and Phillips 2016, Wang and Murnighan 2013). Leung (2014), states that a person with generalist profile can be considered multi-talented, as long as the moves between the different categories take place incrementally. The authors Merluzzi and Phillips (2016) indicate that generalists are preferred over specialists in labour markets where strong institutionalized selection mechanisms are set in place and the supply of specialists exceeds demand. Such factors apply especially in many professional labour markets (Merluzzi and Phillips 2016).
The main focus of this thesis is to examine how individuals (as opposed to organisations and products) can be categorized and what effects this has on the individuals’ career in the short- and long-term, within the empirical context of the selection to an elite visual art program in the Netherlands.
Due to the considerable differences between the art labour market and the non-art labour market a simple extension of the findings of the categorization literature is limited. The art market is characterized by an oversupply of producers, demand uncertainty, high rates of unemployment and consequently high risks (Menger 1999; Caves 2003). Nowadays most of the workers in the arts can be distinguished on the basis of their work conditions: on one hand, there are performers, such as actors, musicians and dancers that are employed temporarily but mostly do not have any job security. On the other hand, there are creative artists, for instance painters, sculptors, writers and composers, that work independently as self-employed contractors without job security (Throsby 1994). However, not only does the art labour market differ from the non-art labour markets, but there are also considerable differences between the creative disciplines themselves. Especially the fine arts differ from other disciplines such as architecture, writing, film and music. For instance, graduates from the fine arts face the risk of unemployment, twice as much as graduates from other creative areas (Comunian, Faggian and Jewell 2011). Only about 23% are able to find a job in the creative industries and those usually face poor working conditions and low salaries. In comparison, the sectors architecture and writing offer graduates more job stability and remuneration (Comunian, Faggian and Jewell 2011). The mentioned differences demonstrate that the generalization of the findings of the categorization literature is not straightforward and may be errant.
As already mentioned, this work seeks to examine how and to what extent an individual’s profile (specialist vs. generalist) has an impact on their success. In the empirical setting of the application process of an elite art program, short-term success can be defined as acceptance to the program, and a (high) ranking in an artist ranking as long-term success. Up to this date, the application process of elite art programs has not gained much attention and limited literature on art education and its importance for a career in this field exists and most
of the statements are rather contradictory. Authors such as Menger (1999) consider formal training as unsuited for the selection of talents and abilities. Other scholars emphasize the importance of an artist’s formal training as a preparation for his or her career (e.g. Martin 2007, Throsby and Hollister 2003) and some even present evidence of the training’s influence on future successes in the art industry (e.g. Galenson 2005a). Outside the general arts market, education is considered as important and worthwhile and following the conventional wisdom a good education and/or training improves one’s job chances. In other words, whereas in the arts, education is rather an option, it is a necessity for the workers in more traditional professions (Filer 1990).
Based on the above arguments, the following research question is derived:
To what extent does being categorized as a specialist or generalist by a selection committee affect applicants’ likelihood of acceptance to an elite visual art program and the applicants’ future artistic ranking?
Since the above-mentioned studies deliver different and contradictory explanations of what kind of profile is more advantageous on the individual level, this work will reconcile the opposing perspectives by empirically testing both. In addition, this work distinguishes, based on Leung (2014), between two different “kinds” of specialists and two different “kinds” of generalists, namely diffuse and focused specialists and incremental and erratic generalists. Therefore, a further aim of this study is to examine these differential effects on the acceptance and future performance.
The thesis’s objectives are twofold: On one hand, it aims to deliver a better understanding of the application processes of elite art programs. On the other hand, the thesis will examine the importance of the applicants’ profiles (specialist vs. generalist) in the application process and future artistic performance, measured by an artistic ranking. By examining the two contradictory perspectives dominating the categorization literature, the
insights gained within the scope of this work will thereby complement and enrich the existing literature concerning the categorization of individuals. Furthermore, this thesis delivers not only new insights on categorization in unstructured selection processes, but also applies the theory in a new context, namely the fine arts, which is different to the settings that have been studied so far (e.g. the labour markets of actors and MBA graduates). In addition, the findings will contribute to the literature regarding the importance of education in the fine arts. With respect to practice, this thesis informs artists as well as elite art schools to what extent specialisation in one particular category is beneficial and therefore a good indicator of success.
The empirical setting of this thesis is the Rijksakademie van Beeldende Kunsten in Amsterdam. Approximately 25 new residents are chosen annually for a two-year elite art program (also referred to as a residency), in which artists are supported and assisted in their work by providing the necessary facilities, a work budget and workshops. Applications can be submitted by upcoming artists, preferable but not obligatory between 25 and 35 years old with a Master of Arts degree. Candidates that are applying for the residency are asked to hand in their personal information, past education history, two motivational letters, concerning their content of work as well as their objectives, and past work examples (Rijksakademie 2017). The Rijksakademie is very selective: from the large pool of applicants only about 1,5% is accepted per year. The application procedure itself consists of four rounds and candidates are judged by a jury of experts from different fields of the fine arts. In the first two rounds, the pre-pre-selection and the pre-selection round, the Dutch and foreign candidates are split into two groups and evaluated separately from each other. Thereafter, in the selection round all the remaining candidates are assessed and approximately 60 to 75 candidates are then invited to Amsterdam for an interview with the selection committee. On the website of the Rijksakademie it is stated that during the procedure the institution focuses
on the vision, motivation and work of the artist as well as his or her artistic quality and room for development (Rijksakademie 2017).
1.2 Approach and Structure
The thesis is divided into six chapters. After an introduction with regard to the subject matter, the theoretical framework of the research is specified. This section pays attention to the foundations of the categorization theory and what effect categorization plays at the individual level such as the labour market. First, a general overview of the categorization literature is given, which depicts the main concepts and definitions. Afterwards, the two contradictory opinions that emerged over the years on this subject are specified. Towards the end of the theoretical framework, the long-term impact of categorization on an artists’ career, measured by an artist ranking, is discussed.
The third part of the thesis is devoted to the research design, where the research objectives and the research instruments are explained. Subsequently the research findings are presented. The end of the thesis forms the discussion of the findings obtained in the study, comparing them to the theoretical background, as well as considering implications for the theory and practice.
2. Theoretical Framework
2.1 Categorization of Individuals
Categories and the process of categorization has been used in social sciences for a long time before it gained a foothold in organisational theory. Categories were mainly used by sociologist to describe human behaviour and cognitive frameworks (Hannan 2010 and Hsu,
Kocak and Negro 2010). By applying the concept of categories to the business field it is possible to understand the network of relationships and resources available to an organisation as well as an organisation’s identity, market behavior and performance outcome (Hannan 2010 and Hsu, Kocak and Negro 2010). In the beginning, organisational theorists focused primarily on the effects of categories on the firm level (i.e., Zuckerman 1999, Hsu 2006 and Negro, Hannan and Rao 2008 Negro, Hannan and Rao 2008), however, the concept was soon extended to individuals, more precisely to applicants in the labour market (i.e., Zuckerman et al. 2003, Leung 2014 and Merluzzi and Phillips 2016).
Zuckerman (1999) was one of the authors who sparked the interest for categories in the field of organisational theory. With his work, he illustrated the application of rules and norms in a more refined way (Hsu, Kocak and Negro 2010). Earlier literature on categorization, looked at categories solely as elements of the external environment. Zuckerman (1999) was one of the first researchers that raised awareness for category systems as “taken-for-granted constructs that influence market behavior and market outcomes” (Hsu, Kocak and Negro 2010, p.7). Following Hsu, Hannan and Kocak (2009), categories can be defined as “social facts”: they help to better understand experiences by clearly defining the market boundaries and depicting what lies within and beyond those boundaries.
The main idea of the categorization concept is that an audience (for instance consumers, critics or a selection committee), assigns a (category) label to a candidate, which can be an organisation, a product or an individual. In turn, schemas, associated with the labels, specify “what it means to be a full-fledged member of that category” (Hsu, Hannan and Kocak 2009, p. 152). Zuckerman (1999) observes this behavior in the stock market and additionally addresses the consequences of category spanning. According to his findings, candidates compete for the selection by the audience. In order to form a relationship with the audience, candidates have to undergo a two-stage process, in which categories play an
essential role: first, they demonstrate certain basic characteristics that they have in common, to be eligible for the selection by the audience. Thereby, a clear demonstration of category membership helps the audience to arrive at a decision. Put simply, when comparing apples with apples, something that looks like an orange will be ignored. Those candidates and their products that are not identified as belonging to one of the categories will be penalized with illegitimacy. Second, the audience evaluates the remaining legitimate players and selects the one that is most attractive. To increase one’s appeal, the candidate seeks to differentiate themselves from each other, while still showing conformity with the category they belong to (Zuckerman 1999).
In order to be considered as member of a category, candidates have to demonstrate common characteristics or features (Zuckerman 1999). Earlier works identified categories mainly on the basis of the candidate’s environment or product (Hsu and Hannan 2005; Hsu, Kocak and Negro 2010), such as industry membership (Zuckerman 1999), movie genres (Hsu 2006) and wine labels (Negro, Hannan and Rao 2008). However, category membership can also be conveyed by features such as, a candidate’s self-description, his or her behaviour or presentation (Pontikes 2012). It is important to note that the condition that something is considered by the audience as category is that the audience demonstrates a high level of agreement on the common characteristic and/or features of the candidates (Hsu, Kocak and Negro 2010). Additionally, category labels can be imposed in two ways. On one side, they are enforced explicitly, by specific actors such as critics or gatekeepers who determine if something belongs to a category or not. On the other side, implicitly by the audience that expects that the candidates in a specific category exhibit certain features (Pontikes 2012). Having said this, category membership depends largely on the audience that examines the relevant candidates. As a consequence, different audiences categorize candidates differently or even in conflicting ways (Hsu and Hannan 2005).
It is important to note that the boundaries of categories cannot be considered as rigid but rather as vague. The reason for this is that candidates only fit into defined categories to a certain extent and will have additional characteristics that allow them to be a member of multiple categories. This also points to the fact that membership in a category cannot be considered as binary, since category members can be associated with a category to a greater or lesser degree (Hannan 2010).
The nonconformity to a specific category is the result of a candidate that violates the boundaries of the relevant category, as identified by the audience. This is the case when a candidate tries to engage in multiple categories. Consequently, one can distinguish between specialists and generalists (Hsu 2006). Jensen and Kim (2014) define specialists as candidates that engage in a single taste position and therefore are limited to a distinct market category. Generalists are on the contrary candidates that engage in multiple taste positions, thereby bridging or spanning categories, and as a result concentrate on different combinations of categories. Even though various works discuss the phenomena of categorization, different terms can be found that describe specialists and generalists. For instance, Ruef and Patterson (2009) describe generalists as hybrids whereas Merluzzi and Phillips (2016) refer to them as candidates with diffused profiles. To avoid confusion and for the sake of simplicity, this thesis will mostly stick to the terms specialists and generalists.
2.2 Specialist Advantage
The prevailing opinion in the literature concerning categorization is that candidates, who fail to be identified with a category or are identified with a variety of categories, face some kind of penalty or disadvantage. As Zuckerman (1999) points out, the audience associates labels or features with a specific category and candidates that do not clearly fit in the considered category and its schema are ruled out. The consequences of not fitting into a category include
disregard, misunderstanding and devaluation (Smith 2011). There are two possible explanations for why generalist behaviour involves economic and social costs. As proposed by Zuckerman (1999) category-spanning increases the information costs for the audience, and deviation from the categories limits the consideration of the candidates in the first place. Contrary, Hsu, Hannan and Kocak (2009) suggest that the spanning of categories is linked with the development of capabilities, which in turn affect the attractiveness of the candidate experienced by the audience. By engaging in multiple categories, candidates may develop a quantity of capabilities but at the same time they lack the focus and effort to concentrate on each single category. As a result, category-spanning affects the quality of their capabilities and expertise. This does not only put them at risk of over-diversifying their resources but also hinders them to outperform in any category and consequently to attract enough appeal from the audience (Hsu, Hannan and Kocak 2009). Taking both arguments into account, it can be concluded that having a specialised or focused profile is favoured by the audience over presenting oneself as a generalist (Negro, Hannan and Rao 2010).
Extant literature has analysed the specialist advantage in various contexts. For instance, Hsu (2006) analysed at the role of audiences’ perceptions of films with varying niche widths and concluded that movies that include more genres are able to attract more people. However, they also have difficulties in generating true appeal among their audience since for them it is difficult to grasp the movie’s identity and fulfil their expectations (Hsu 2006). Additionally, Negro, Hannan and Rao (2008) found proof of the specialist advantage in the wine industry. Quality wines made by wineries that specialised in a particular style of wine received a better critical evaluation than wines made by winemakers, who offered a variety of styles (Negro, Hannan and Rao 2008).
The notion that specialists are favoured over generalists is also applicable to the labour market and its individuals. Zuckerman et al. (2003) argue that a job applicant’s
specialisation is a good indicator of his or her competence. The reason for this is that employers do not have access to credentials and endorsements, which could attest an applicant’s skills. Therefore, the employer has to rely on the candidate’s past job history “[…] that met some threshold of competence” (Zuckerman et al. 2003, p.1026). According to the authors, labour market categories are a good indicator of the specific skill of the worker as well as his or her underlying ability. In other words, a person, who can demonstrate a certain amount of experience in a single labour market category, is considered as having enough skill to participate in this category. If this is not the case and the person presents a more differentiated profile, he or she runs risk of being considered as dilettante, someone who lacks the skills to perform in one specific category. However, according to Zuckerman et al. (2003) specialisation is advantageous at the start of a career but it may be disadvantageous in the long-term. By specialising in one category, experienced workers can be considered less-skilled, since they consistently demonstrate the same set of skills (Zuckerman et al. 2003). Earlier research has also found evidence for the specialist advantage. Neale (1995) analysed the relationship between specialisation and wages. The results indicate that, workers that switch jobs in the same industry earn more than workers that find new work in a different industry. The reason for this is that companies prefer workers, who have a specific set of skills, which are characteristic of a certain industry. Workers who are new to an industry, hence generalists, exhibit rather general skills, are not acquainted with the work and therefore are worth less in the eyes of the hiring organisation (Neale 1995). Further evidence, for a preference for specialists, stems from the work of Ferguson and Hasan (2013), which examined the promotion probability of workers in the elite bureaucratic service. The authors concluded that workers with a diverse work history have lower chances of promotion than their specialised co-workers. Reason for this is that a specialised profile demonstrates general ability (Ferguson and Hasan 2013).
Based on the arguments put forward by the authors Zuckerman et al. (2003), Ferguson and Hasan (2013) and Neale (1995), it can be expected that in the application process of elite art schools, specialists are favoured over generalists. On one hand, Zuckerman et al. (2003) mention that early-year actors who specialise in one category are more likely to be hired. The applicants to elite art schools can be considered as at the beginning of their career. The target group of the Rijksakademie, the research setting of this thesis, are between 25 and 34 years old and the residency is designed for individuals with the intention of undergoing further training (Rijksakademie 2017). On the other hand, the candidates’ applications to elite art schools do not include strong credentials or endorsements. The jury of the Rijksakademie has only access to the candidates’ personal information, past education history, two motivational letters and past work examples. Following Zuckerman et al. (2003) this means that a specialisation, is therefore important in order to facilitate valuation on behalf of the jury. In other words, candidates that demonstrate a specialised profile in their past studies, attract the jury’s attention and demonstrate competence (Zuckerman et al. 2003). Consequently, the following hypothesis can be derived:
Hypothesis 1: Specialists are more likely to be accepted at an elite art education institution than generalists.
2.3 Generalist Advantage
Initially, categorization literature considered that specialists are always favoured over generalists, since the spanning of categories leads to confusion, lack of appeal and devaluation (Kovacs and Hannan 2010; Negro, Hannan and Rao 2010 and Hsu 2006). However, more recent works contradict this assumption and argue that there are some conditions under which being a generalist does not involve serious disadvantages. In
emerging classification systems that have not gained social legitimacy, a membership in different categories will be tolerated since the boundaries of the categories themselves are still in flux (Ruef and Patterson 2009). Furthermore, Kovacs and Hannan (2010) discover that the consequences of category spanning depend on the contrast of the categories. If categories do not stand in stark contrast to each other, hence exhibit fuzzy boundaries, the social and economic costs of being a generalist will be restrained. On the contrary, spanning categories in an environment where categories have rather sharp boundaries will lead to confusion about the candidate’s identity on the part of the audience and consequently lower his or her appeal (Kovacs and Hannan 2010). Additionally, Kovacs and Hannan (2010) stress out that activists, who are according to their definition “experts on the relevant dimensions” (p. 7), are actually appreciative of generalists. The reason for this is that activists are more acquainted with the relevant categories and in general perceive the category boundaries as fuzzier as do other audience members (Kovacs and Hannan 2010). Pontikes (2012) arrived at a similar solution and emphasizes that it depends on the audience if generalists attract appeal or are penalized. An audience that has influence over the candidates and acknowledges novelty prefers generalists over specialists.
Further criticism on the specialist advantage points out that in the studies of Zuckerman (1999), Negro, Hannan and Rao (2008) and Hsu (2006), consumers make short-term purchases or review a film or wine based on category perception, whereas in the labour-market the focus lies rather on long-term employment and factors such as recruitment, training and replacement costs (Merluzzi and Phillips 2016). Additionally, Zuckerman et al. (2003), who to a great extent support the specialization benefit, pose some limitations to the results. According to them, category spanning is not per se a bad thing. The observed effects of exhibiting a specialist profile are more common at the beginning of an actor’s career. As the actor advances in his career, the benefits of having a focused profile decrease. Being
categorized as a specialized actor could be considered in the long run as being untalented, because the actor repeatedly replicates the same skills. Furthermore, the authors noted that specialization only confers an advantage in those settings where quality signals such as credentials and sorting mechanisms are not available or insignificant (Zuckerman et al. 2003).
The authors Merluzzi and Phillips (2016) extend and develop these limitations and argue that specialists are not always favoured and as a matter of fact specialists are discounted in certain settings. This is a likely scenario in a context characterized by three decisive factors: (1) strong institutionalized screening mechanism, (2) candidate profiles with direct investments that signal their value, and (3) a high supply of focused candidates relative to demand. These conditions are particularly typical for many professional labour markets. If strong institutionalized screening mechanisms are in place, the importance of demonstrating specialization decreases as the previous assessments serve as a sound indicator of quality. These screening mechanisms include requirements, such as previous work experience, high grades in certain subjects and credentials. Furthermore, labour markets in which “candidates have publicly chosen the past investments that constitute their profiles” (Merluzzi and Phillips 2016, p. 92), generalists will be favoured over specialists. In other words, a job applicant that invests in a diversity of experiences will stand out from the masses of applicants who have focused on becoming specialists and therefore will attract greater appeal among employers. Key condition is that the employer is not searching for a specialist that has a rare and complex knowledge. Last but not least, supply and demand influence the kind of applicant profile that will be given priority. The more job seekers with focused profiles who are in search for a job, the less valuable specialists become and the more likely it becomes for generalists to secure a job (Merluzzi and Phillips 2016). In their study, Merluzzi and Phillips (2016) examined and verified their assumptions in the labour market of Master of Business
Administration graduates. The authors defined specialists and generalists based on their work experience prior to the MBA program, MBA concentration and extracurricular activities during the program and internships. Specialists, or in other words focused individuals, had a career in the finance sector prior to the program, specialized in finance, joined a student club concerned with finance and completed an internship in the finance industry (Merluzzi and Phillips 2016). According to the authors’ opinion, these results are not limited to the MBA labour market but rather can be adopted to “[…] other professional elite graduate programs, in which the admissions requirements to top-ranked programs are equally demanding and provide an effective pre-screening of candidates” (Merluzzi and Phillips 2016, p. 117). One example of such a similar setting is the National Basketball Association (NBA). Wang and Murnighan (2013) looked at specialists and generalists in the NBA. Specialists are defined in this context as three-point shooters, thus mastering a specific skill. On the contrary, generalists are players who are more flexible and have diverse shooting skills. The authors found that when selecting and rewarding players for any team, a generalist bias persists. In other words, even though the team is in need of a specialist, generalists are favoured and even receive higher salaries and fan appeal. This effect becomes even stronger when comparing specialists and generalists directly, the bias increases. Contrary, while evaluating them separately the bias decreases (Wang and Murnighan 2013).
One could argue that the selection systems in elite art schools are not as institutionalized as in the MBA context that Merluzzi and Phillips studied. In the MBA recruiting market, students have to go through highly selective admission processes to be accepted at the respective schools (Merluzzi and Phillips 2016). By contrast, in the art field a comparable test or pre-selection has not existed until this date. However, it is possible to draw comparisons between the application processes of elite art education institutions and the application process of MBA or PhD elite institutions, since they both are very strong and selective. Art
schools such as the U.C.L.A only accept one out of 32 applicants, in doing so it is more competitive than for instance the Harvard Business School, which accepts one out of every 10 applicants (Solomon 1999). The selection systems implemented in the application process for such institutions (art and non-art related) can be considered as gatekeeping devices. In other words, they are a mechanism that controls access to a specific profession. According to Brear, Dorrian and Luscri (2008, p.93f) it comprises of “[…] the identification of evaluative criteria and process, and the accountability of the gatekeeper to apply the criteria and take responsibility for the evaluative decisions”. Selection and thus gatekeeping to a study program therefore ensures the profession’s and graduate’s quality and the study program’s integrity, as well as sheltering the community’s interest (Brear, Dorrian and Luscri 2008). It is important to note that the evaluation of applicants is far from being perfect. The decisions of whether one gets accepted or not, is based on incomplete information, restricted time and personal and collective standards of judgment (Posselt 2014). Elite education institutions usually receive more applications than they have study places. Consequently, they are in a position where they do not only implement selection mechanisms in order to safeguard the above-mentioned features, but additionally are able to choose some students over others based on certain personal characteristics (Killgore 2009). An application to an elite institution includes among others an undergraduate transcript, letters of recommendation and a personal essay. Additionally, academic merit, such as standardized test scores and a GPA, is the basic prerequisite to be accepted at an elite institution (Pratt 2015; Merluzzi and Phillips 2016). From this perspective, the selection system in place at elite education institutions seems straightforward. However, Killgore (2009) notions the following:
Elite colleges do not simply reward superior individual talent and effort wherever it appears in the population. They pick and choose the very best students they need. And
their admissions practices are not intended to be transparent, as a strictly merit-based system would require (p. 482).
Posselt (2016) makes similar statements. According to her work, quality is a “moving target”. The conditions of being accepted are prone to change and are not bound to a certain standard.
The selection process of elite art schools may not be considered as highly institutionalized due to the lack of pre-selective scores and tests. However, they can be considered as strong and very selective, as can be seen on the example of the U.C.L.A above (Solomon 1999). Having said this, elite art schools have definitely implemented processes, like other regular elite education institutions, to ensure that only the very best are accepted. It could also be reasoned that elite art schools, such as the Rijksakademie, have multi-staged admission processes as a substitute for missing pre-selective scores and tests that among others, can be found in the MBA recruiting market.
Furthermore, there are other additional characteristics that support the argument for the specialist discount in the elite art school application process: Merluzzi and Phillips (2016) mention that hiring managers in the MBA context, consider specialized candidates as less appealing. On one hand, specialized candidates are more common than their generalist counterparts. On the other hand, generalists that exhibit accomplishments in various areas, demonstrate talent and commitment beyond a basic interest and background in a specific area (Merluzzi and Phillips 2016). The Rijksakademie data indicates that the number of specialists is much higher than the number of generalists applying, suggesting the possibility that generalists could be perceived as more interesting. This argument is supported by the works of Kovacs and Hannan (2010) as well as Pontikes (2012). These authors indicate that experts as well as an audience that honours novelty, prefer generalists over specialists. It can be
reasoned that both are found in a selection system of elite art schools. From the research setting the Rijksakademie, it is known that the jury consists of experts from different fields of the fine arts. Likewise, it can be assumed that the jury selecting applicants, is on the search for novelty. Additionally, Zuckerman et al. (2003) argues that specialization is beneficial, if the only information available about job applicants is his/her job history. Having said this, elite art schools require from their application candidates much more than just their past job history. In the case of the Rijksakademie, applicants are invited to hand in besides personal details and former studies, work examples and two motivational letters describing their content of work as well as their work objectives (Rijksakademie 2017). The above argumentation leads to the second hypothesis:
Hypothesis 2: Generalists are more likely to be accepted at an elite art education institution than specialists.
2.4 The Specialist – Generalist Continuum
Leung (2014), takes a similar position as Merluzzi and Phillips (2016), Kovacs and Hannan (2010) and Pontikes (2012) and criticizes the assumption that drawbacks result from any category spanning. According to the author, previous research that insists on the strict separation of generalist and specialists, has two major weaknesses. First, it is known that generalists can be considered as either multi-talented or unsuccessful in any of the multiple categories they engage in. However, extant literature regards both of them similarly disadvantaged. Nevertheless, Leung (2014) stresses that some applicants, given a fixed portfolio of multiple category experiences, are favoured over others. Consequently, employers are differentiating between candidates with distinct category experiences. The second shortcoming of previous research is that it conceptualizes a job applicant’s profile as
solely a function of the most recent (occupational) activities and in doing so neglects the sequence and nature of accumulated experiences (Leung 2014). Leung (2014) distinguishes between three prototypical profiles: specialist, incremental generalist and erratic generalist. Incremental generalists have consecutively gained experience in related categories whereas erratic generalists demonstrate expertise in various unrelated categories. According to the author, workers with an incremental generalist profile are favoured over workers with a specialist or erratic generalist profile. The reason for this preference describes Leung (2014) as follows: a person who moves incrementally between job categories demonstrates commitment, since it requires substantial effort to collect a consecutive linked set of job experiences. On the contrary, a candidate with an erratic job history moves between very dissimilar categories and thereby creates the impression that he or she is not fully dedicated to the category he or she applies to. As a result, the type of generalist profile matters and has a profound impact on the selection process (Leung 2014).
The notion of Leung (2014) that incremental generalists are perceived as more committed than erratic generalists and specialists, is in line with the work of Merluzzi and Phillips (2016). These authors argue that a consistent performance in one category is disadvantageous for job applicants. Combining Merluzzi and Phillips (2016) reasoning with Leung’s (2014) distinction of the generalist profile, the following hypothesis can be proposed:
Hypothesis 3: Incremental generalists are more likely to be accepted at an elite art education institution than erratic generalists and specialists.
The argumentation of Leung (2014) implies that a candidate’s profile can change over time. An erratic generalist is able to become an incremental generalist, if he or she tries to
accumulate experiences in related categories. However, Leung (2014) defines specialists as candidates who do not move between categories, but rather remain or even atrophy in their specific category. Since candidates do not turn immediately into specialists by working once in a category, this raises the question whether specialists also differ in their degree of specialization as generalists do. As a result, the writer of this work argues that a specialist that recently started to specialize in a specific area, is different from a specialist that has already collected various work experiences in this field. To avoid any confusion the work will address these two types of specialists as diffuse specialists and focused specialists. Diffused specialist, may have decided to specialize in a particular area, but since he or she still only has limited experience in the field, they may still have some experience and knowledge from other unrelated categories. On the contrary, focused specialists are more extreme in their specialization. Candidates with such a profile have extensive experience in one field and can be considered as proven experts in their respective field. To ensure a better understanding of the matter, the four different profiles, erratic generalist, incremental generalists, diffused specialists and focused specialists are placed on a continuum, ranging from experience in related categories to experience in unrelated categories. The continuum is depicted in illustration 1.
As mentioned above, Zuckerman et al. (2003) stress that having a specialist profile, compared to having a generalist profile, is advantageous at the beginning of one’s career. Conversely, too much specialisation (focused profile) has a detrimental effect for the future career, since a person appears to be single-skilled (Zuckerman et al. 2003). Following this notion of Zuckerman et al. (2003) and the distinction of specialists based on Leung (2014), it cannot be expected that incremental generalists are favoured over erratic generalists and specialists. But rather, diffused specialists are favoured over focused specialists and generalists as a whole. Therefore, the following hypothesis can be derived:
Hypothesis 4: Diffuse specialists are more likely to be accepted at an elite art education institution than focused specialists and generalists.
2.5 Categorization and Future Ranking
In the past years, the topic of education in the creative and cultural industries has not received a lot of attention. In general, it is rather difficult to find research dedicated to this area and most of the studies and papers available present divergent opinions. On one side, researchers downplay the importance of artists’ education. Menger (1999) as well as Janssen (2001) argue that education per se does not pose a barrier to entry to the profession as an artist. On the contrary, education is seen as a rather imperfect mechanism for the selection of high talent and ability. It is the concept of learning-by-doing that is considered to help an artist to gain knowledge about his or her occupation as well as develop and enhance his or her skill set (Menger 1999). Janssen (2001) argues that autodidacts are very common in the arts, basically because everyone can enter the industry and call him- or herself an artist. There are no training institutions or associations that protect the artistic profession and that award its graduates with recognition (Janssen 2001). Filer (1990) states that most of the artists simply
make the decision to not undergo training or education in the arts. According to him, artists experience their first zenith at a very young age. As a result, artists may decide to skip education in order to capture the rewards of their talent (Filer 1990). The study of Karhunen (1996) provide mixed results. An education in the arts and therefore a degree does not have an effect on employment. However, it does have an effect on the mode of employment. The majority of the examined graduates work freelance and only one-third of them have a permanent job (Karhunen 1996).
Historically, it is understandable that art education was not a necessity for working as an artist. However, nowadays the situation has changed and artists need to be academically certified so a Master in the fine arts is almost a must (Solomon 1999). This is also reflected by the data of the Rijksakademie. Autodidacts are a rarity, the applicants with a Bachelor and/or a Master in an art-related studies are the majority. Various scholars give reasons for the importance of an education in the arts. Becker (1982) states that an education is necessary for new artists to understand and pick up the art specific language, terms and conventions in order to fit into the arts world in the first place and to be able to explain their own work to other people. This notion is confirmed by Martin (2007), who asserts that art schools pave the artists the way for their integration into the market. Among others they are trained in performing exhibitions, promoting their work to legitimating bodies and applying for grants (Martin 2007). The work of Throsby and Hollister (2003) stresses the importance of an education in an artist’s career. The majority of artists, that were part of their study, had some type of formal training. In particular, visual artists and craft practitioners are inclined to undergo formal training at an education institution, such as, among others, universities or technical and further education schools. Those who completed formal training consider it one of the most important means of preparation for their work as an artist (Throsby and Hollister 2003). Some studies even consider vocational training at an artistic education institution to
have an influence on the future success of an artist. Galenson (2005b) points out that several of the greatest artists attended an (elite) art school. They were not only affected by the content of the studies but also by teachers that challenged and inspired them. Especially, some schools, like the Yale School of Art, produce consistently successful artists that sell their works for high prices (Galenson 2005a).
Still unknown is whether formal education in the arts increases the future performance of its students. As Towse (2006, p. 878) puts it “You can’t make a silk purse out of a sow’s ear”. This refers to the fact that art schools may have put selection systems put in place in order to pick out the best of applicants. However, there is no guarantee that they assess the quality of all their students with all certainty (Towse 2006). In this case, factors such as heterogeneity and the absence of objective information about quality, play not only a big role in the artistic labour market but also in the assessment process of art school applicants.
As can be seen the work about formal education and training in the arts is rather sparse and sometimes contradictory. However, especially the more recent works in this field of research are assuming a positive relationship between the attendance at an (elite) art education institution and future career prospects (Galenson 2005b; Galenson 2005a and Martin 2007). Research on common elite education institutions and their long-term effects on their students’ performance is also quite rare and rather inconsistent. For instance, Brand and Halaby (2006) identify a positive link between the attendance at an elite education institution and future career achievements and occupational status. Anelli (2016) comes to similar conclusions: the attendance at an elite institution has a positive effect on the graduates’ annual income. Whereas, Clark and Bono (2016) found that only women profit from an elite school attendance in respect to their labour market productivity and incomes. However, results like these must be treated with caution. Students that attend elite schools could be
selected based on unobservable characteristics, which, rather than the quality of the institution, may have an effect on future earnings (Anelli 2016).
The literature about elite education institutions gives reason to believe that elite schools confer an advantage on their graduates in respect to their future vocational success and earnings. In the context of the creative and cultural industries, such a link has not been examined in detail. A positive influence of formal arts training on future success has been proven or indicated by several studies (Galenson 2005b, Throsby and Hollister 2003; Martin 2007). Additionally, Galenson (2005a), mentions that numerous important artists graduated from diverse elite art schools but unfortunately did not undertake a closer examination of this link. Therefore, this thesis proposes the following hypothesis:
Hypothesis 5: Accepted applicants to an elite art program are more likely to have a better artistic ranking compared to those applicants who were rejected.
In the case that specialists, or generalists, are more likely to be accepted at an elite art school (hypothesis 1 and hypothesis 2) and that graduates from elite art education institutions have higher chances of succeeding in their career (hypothesis 3), it can be expected that being categorized as specialist or generalist has a long-term effect on the artists ranking. Although Leung (2014) argues that a candidate’s profile is prone to change over time, it can be expected that this happens at a relatively slow pace. Reason for this is that an applicant’s profile is an accumulation of category experiences and a change in profile requires one to gain several experiences in the one category (for a more specialised profile) or diverse categories (for a more generalist profile). As a result, it can be assumed that applicants in a certain category at the time of application stay in that category for a given period. The following hypotheses that subsequently can be derived are:
Hypothesis 6: Applicants that exhibit a specialist profile, when applying to an elite art program, are more likely to have a better artistic ranking in the future compared to those applicants who exhibit a generalist profile.
Hypothesis 7: Applicants that exhibit a generalist profile, when applying to an elite art program, are more likely to have a better artistic ranking in the future compared to those applicants who exhibit a specialist profile.
All in all, the concept of categorization allows to gain a better understanding of why certain job applicants have a better chance of being hired. The early categorization literature predicted a clear connection between category-spanning and penalties by the audience (among others Zuckerman 1999, Smith 2011 and, Negro, Hannan and Rao 2010). However, over the last years this notion has drawn criticism. Particularly the authors Merluzzi and Phillips (2016) suggest that being a specialist is actually disadvantageous in professional elite graduate programs, given certain key factors. One of these factors is a strong selection mechanism, which for example can be found when MBA graduates are searching for a job (Merluzzi and Phillips 2016). Given these contradictions in the extant literature this thesis contrasts both perspectives and proposes contradicting hypotheses, on the one hand assuming a specialist and on the other hand a generalist advantage. This notion is complemented by the work of Leung (2014), who argues that not only do specialists and generalists differ from each other, but people within the generalist group exhibit different career trajectories. As a result, Leung (2014) distinguishes between incremental and erratic generalists, arguing that incremental generalists are more likely to find a job than erratic generalists. Incremental generalists, who gain experience in related categories, will have a higher appeal than erratic
generalists who demonstrate expertise in various unrelated categories and therefore communicate lack of commitment. Following this argument, this thesis proposes that there are also differences within specialists: diffuse specialists, who are at the beginning of specialising, and focused specialists, who have already gathered enough experience and skills in the one particular category. Based on the first pair of contradicting hypotheses, this thesis argues that first, incremental generalists are favoured over erratic generalists and specialists and second, contradicting this assumption, that diffuse specialists are favoured over focused specialists and generalists (Leung 2014).
3. Research Design
3.1 Data collection
In order to answer the research question, this thesis will take a quantitative approach and will make use of: (1) the secondary data provided by the Rijksakademie, (2) the Linguistic Inquiry Word Count, (3) the Artfacts artistic ranking.
Rijksakademie. This study will use the data from the application process from two
years, namely 2009 and 2011. The databases from both years display each candidates’ personal information such as gender, place of residence and main working area as well as previous education and years of graduation. Furthermore, the data contains the applicants’ content of work and objectives, which they submit in text format. Applicants for whom not all data was available (n = 29), or double entries existed (n = 20), were excluded. Furthermore, applicants that submitted their motivation letters in a different language than English (n = 13)were excluded as well. This results in the final number of 1303 applicants in 2009 and 1814 applicants in 2011.
Linguistic Inquiry Word Count. In order to make use of the motivational letters
submitted by the candidates a computer program, Linguistic Inquiry Word Count (LIWC) was used to code the articles. LIWC is a text analysis tool that scans text files and compares the words to pre-defined categories that are based on built-in dictionaries. The categories themselves represent various emotions, cognitive processes and personal traits. For instance, the word “cry” belongs to five different categories: Negative Emotions, Overall Affect, Verb, Sadness and Present Focus. However, LIWC does not only indicate to what categories a word belongs to but also indicates with a point scale or percentage, depending on the category, how well the word matches the relevant category (LIWC 2017). LIWC processes in this manner every word in a given text. The analysis output summarizes these results and illustrates to what extent each category is represented in the text. To give an example, the LIWC analysis of a text could show that 2% of all words in the text belong to the category Positive Emotions (Tausczik and Pennebaker 2010). Several studies so far have made use of the LIWC analysis: Alpers et al. (2005) studied the conversations in breast cancer support groups; Bohanek, Fivush and Walker (2005) compared with the help of LIWC positive and negative memories of emotional events; and Pennebaker and Lay (2002) analysed approximately 10 years the language use of the former New York mayor Rudolph Giuliani in order to assess changes in his personality. For this thesis five variables were obtained from LIWC: positive emotions, analytic, clout, authenticity and emotional tone.
Artfacts Ranking. Since this study also examines what effect the categorization of
applicants has on the artist’s future artistic ranking, data from a web based artist ranking site www.artfacts.net, called “Artfacts” is used. This ranking classifies artists according to their exhibition success. The Artfacts ranking includes artists that had at least one gallery representation in their past. In order to rank the artists, Artfacts makes use of a point system, where every exhibition in a gallery or a museum earns the artist points. Thereby the
reputation of the institution plays an important role. An exhibition at an institution with a high reputation will allow the artist to gain more points than an exhibition at a more or less unknown gallery or museum. Consequently, an artist with a high number of “exhibition points” will have a high-ranking position and a low-ranking number (Artfacts 2017). To give an example, in 2017 Andy Warhol is ranked number 1 with 54.944 exhibition points followed by Pablo Picasso on rank 2 with 38.683 exhibition points. The Artfacts ranking can be considered as very extensive and included in 2017 around 570.000 artists (deceased and alive), 760.000 exhibitions and 21.000 galleries all from 192 countries (Artfacts Beta 2017). The Artfacts ranking has already been used within the scope of various studies, among others the study of Zorloni and Ardizzone (2016), which studied the competitive dynamics in the contemporary art market; and the authors Quemin and Hest (2015) who examined three major artist rankings, including the Artfacts ranking and their national concentration effects.
For the purpose of this study, the artists’ ranking four years after the start of the training at Rijksakademie and thus two years after the artists’ graduation has been used. That is, for the application year 2011, the ranking from 2015, and for the application year 2009, the Artfacts ranking from 2013 has been used. By using a ranking that exhibits a time lag between the start of an artist’s training and the year of ranking it can be ensured that a possible effect of the training on an artist’s ranking is taken into account. Controlling the artists’ ranking after a relatively short time after their graduation ensures that one of the reasons for their success is their education at the elite art school. If this study would focus on later dated rankings it cannot be ruled out that other factors have an influence on the artists’ ranking.
3.2 Dependent Variables
Hypotheses 1 to 4 test the influence of a candidate’s specialist or generalist profile, respectively a candidate’s focused specialist or incremental generalist profile on acceptance to an elite art programme. Accepted applicants receive a value of 1, while the rejected applicants are the base category and receive a value of 0.
Hypothesis 5 tests whether a graduation from an elite art program has an influence on the artist’s future ranking. Hypotheses 6 and 7 examine the association between applicants’ specialists and generalists profiles, ceteris paribus, and their future artistic ranking. The future artistic ranking of the applicants is derived from the Artfacts ranking, which ranks artists according to their exhibition success, as measured by the number and location of exhibitions as well as the reputation of the gallery or museum where the exhibition is held (Artfacts 2017). For the purpose of this thesis four categories were created that summarize all the artists within a certain rank. To form these categories, the logarithm of the candidates’ ranking was taken. Following four categories were formed: the first category, ranging from 0 to one standard deviation below the mean; the second category, ranging from one standard deviation below the mean to one standard deviation above the mean; and the third category, ranging from one standard deviation above the mean till the lowest rank. The fourth category consists of all the candidates that were not listed in the Artfacts ranking in the relevant year. All these non-ranked candidates a value of 25 was assigned, which would equal a very low ranking. Simultaneously it ensures that there is a big enough gap between the ranked and the non-ranked artists so that the results are not negatively influenced. For an easier understanding the categories contain the following rank range (logarithm of the rank):
• 1 = 0 – 4,10658 logarithm of rank • 2 = 4,10659 – 5,01502 logarithm of rank
• 4 = 25 (not-ranked)
3.3 Independent Variables
The independent variable in the study is the generalist profile variable, which equals 0 if the individual has a generalist focus and 1 if he or she can be characterised as a specialist. Following Merluzzi and Phillips (2016), someone can be considered a specialist, when this person specialised in his or her former education in one particular area. Conversely, this means that a person with any variation in those activities can be considered as a generalist. Applied in the context of this thesis this would mean that an applicant can be regarded as a generalist if one of two conditions apply:
• The applicant has no prior education in an art-related study.
• The applicant studied besides an art-related studies, also an art-unrelated studies. In the context of this research art-related studies are considered either as studies completed at a university that can be clearly identified as an art university (for instance Royal Academy of Arts) or at an art and design university faculty (for instance Fine Art Faculty at the University of Lisbon). Following this characterisation art-related studies of applicants of the Rijksakademie were among others fine arts, graphic design, photography, art history and architecture. The art-unrelated studies included for instance communication, philosophy, journalism and comparative literature. Some applicants did an exchange year at a foreign university, year or semester abroad, these studies was not considered as a separate study.
The study also assumes that the specific type of generalist or the specific type of specialist has an influence on the acceptance to elite art schools. These variables are namely:
incremental generalist, erratic generalist and diffuse specialist, focused specialist. These
into two subgroups, namely focused specialists and diffuse specialists. These four sub-categories, applied to the context of the Rijksakademie are defined as:
• Erratic Generalist: the applicant has no prior education in one or more art-related studies
• Incremental Generalist: the applicant has graduated from an art-related studies and from one art-unrelated studies
• Diffuse Specialist: the applicant has graduated from one art-related studies • Focused Specialist: the applicant has graduated from two art-related studies. For the purpose of this study, dummy variables were created for these four categories. In order to test hypotheses three and four, the data set will be split up into two data sets: one containing only candidates with a generalist profile, and one only containing candidates with a specialist profile. In the former, the dummy variable incremental generalist will be used, with the variable erratic generalist as the reference category. In the latter, the dummy variable diffuse specialist will be used, with the variable focused specialist as a base category.
3.4 Control Variables
The acceptance to an elite art school and an artist’s future ranking could also be predicted by other variables. The study controls for a number of variables: gender, age, Dutch
nationality and main working area. Regarding nationality, candidates where categorized as
either Dutch (value of 0) and non-Dutch (value of 1).
Similarly, for the variable main working area four dummy variables were created. The four dummy variables are based on the distinction made by the Rijksakademie and are:
painting, drawing/graphics, photography, and video/film/sound/performance. The category sculpture/installation functions as the base category.
In addition, certain characteristics in the applicants’ application letters (content of work and objectives), which are reflected in the data obtained from LIWC, will be tested to ascertain whether they influence a candidates’ acceptance. Overall, five variables have been extracted from the application letters, namely positive emotions, analytic, clout, authenticity and emotional tone. Thereby, LIWC distinguishes between individual and summary variables. The individual variable, positive emotions, is characterized as followed:
• Positive emotions: words that express favourable feelings (examples are: happy, good)
It is important to note that percentage of the total words relating to each of this variable is measured rather than the number of respective words (LIWC 2017). In other words, a text that has a value of 50% in the category positive emotions, consists one half of words that express favourable feelings.
The last four variables are summary variables that represent a 100-point scale, ranging from 0 to 100. This point scale represents a continuum indicating a certain style of emotions, cognitive processes and personal characteristics.
• Analytic: reflects the analytical thinking of the writer, high numbers represent logical and hierarchical thinking whereas lower numbers refer to informal and personal thinking.
• Clout: higher numbers refer to a knowledgeable and confident author, whereas lower numbers represent a more timid and anxious perspective.
• Authenticity: higher numbers indicate that the written text is associated with an honest and personal style, on the contrary, lower numbers exhibit a more reserved and guarded style.
• Emotional tone: a number around 50 reflects a lack of emotionally or ambivalence. Numbers that are higher stand for an optimistic and positive writer, whereas lower numbers disclose rather sadness, hostility and anxiety (LIWC 2015).
3.5 Method
Given the binary nature of the dependent variable in hypotheses 1-4 a probit regression was performed to examine the effects of a candidate’s generalist profile on his or her chance of acceptance at an elite art school. To assess whether the candidate’s profile at the time of application has an influence on the candidate’s future ranking as an artist (Hypotheses 5-7), an ordered logistic regression was performed.
It is important to note that the probit regression in SPSS uses the highest value, or respectively the last category, of a variable as the reference category. Consequently, for the dummy variables gender, dutch nationality, main working area, generalist - specialist profile as well as the sub-categories erratic generalist, incremental generalist, diffuse specialist and focused specialist, the value 1 is the reference category. However, this does not apply to the binary dependent variable, acceptance, for which 0 is the reference category.
4. Results
4.1 Descriptive Statistics
Table 1: Descriptive summary of the application data
With both application years considered together, 3117 candidates applied for a residency at the Rijksakademie. Of these 3117 applicants only 54 were accepted. The total number of women and men was almost equal, with slightly more women (50.9%) applying. When dividing the candidates into age groups it becomes apparent that the age group 25-34 is the most numerous, with 2316 (74.3%) candidates. The number of candidates belonging to the age groups of 55-64 years (4 candidates) and 65-74 years (2 candidates) was in comparison very low. The number of candidates within each main working area varies considerably. Candidates specializing in sculpture and installation represent 1100 candidates, more than
one third of the sum of applicants. The main working areas photography (10.4%) and drawing/graphics (8.4%) form the minority, with considerably lower numbers of candidates compared to the other areas. Most candidates were categorized as specialists (87.6%), whereas generalists constituted the minority with 385 (12.4%) applicants.
Table 2: Min., Max. Means and Standard Deviations
Table 2 illustrates minimum, maximum, means and standard deviations. The average ranking of the applicants in the Artfacts ranking is 18021,6 (minimum 0, maximum 153704). The mean age of the applicants is 31,73 years (minimum 20, maximum 72).
Table 3: Descriptive summary sorted by year
Table 2 contains a descriptive summary of the data sample divided into the two application years. In 2009, 1303 applicants (627 women and 676 men) applied to the Rijksakademie, in 2011 the number rose to 1814 applicants (903 women and 911 men). Furthermore, candidates with a Dutch citizenship made up 1/10 of all applicants, both in 2009 and 2011. Data from each of the years was also consistent with regards to age groups. The age group 25-34 years constitutes the majority, whilst candidates from the age groups 55-64 and 65-74 are rarely applying. In both years, candidates specializing in sculpture/installation dominated the application process, whereas candidates specialized in drawing/graphics were in the minority, constituting only around 8%.
Illustration 2: Candidates’ Profile categorized by year
Illustration 4 visualizes the distribution of specialists and generalists among the applicants. In both years, when specialists were compared to generalists, specialists made up the majority with 88,7% in 2009 and with 86,9% in 2011. When taking a closer look at the subcategories of specialists and generalists it becomes apparent that candidates with an erratic generalist profile were a rarity in both years (2009: 2,3%; 2011: 4,8%). The same applies to the incremental generalists (2009: 9%, 2011: 8,3%). The diffuse specialists as well as the focused specialists dominated the application process of the Rijksakademie in both years. In 2009, the difference between diffuse specialists (24,7%) and focused specialists (64%) was rather large. Two years later in 2011, the numbers of diffused specialists (37,5%)
Table 4: Descriptive summary of accepted and rejected candidates
Table 3 illustrates the data regarding the candidates accepted and rejected at the Rijksakademie. Among accepted candidates, men are in a slight majority with 59.3%. Looking at the candidate’s nationality it becomes apparent that Dutch candidates make up a third (31.5%) of all candidates. Furthermore, the table illustrates the distribution of the accepted candidates by main working area. The largest working areas among the accepted candidates are in descending order: sculpture/installation (27.8%), painting (25.9%), video/film/sound/performance (24.1%) and photography (18.5%). What is striking is that, in the two application years only two candidates specializing in drawing/graphics (3.7%) were accepted.
Illustration 3: Accepted Candidates’ Profiles
When looking at the accepted candidates’ profiles it becomes apparent that those candidates that exhibit a generalist profile are in the minority with 11.1% compared to the candidates with a specialist profile (88.9%). When splitting them into subgroups, the categories diffuse and focused specialists are almost balanced with 25 candidates and 23 candidates each. The subcategory with the least candidates are erratic generalists with only two candidates, followed by the category incremental generalist, with four candidates.
