• No results found

2. Structure of the educational system: institutional practices and social

2.1.2. Selection function

2.1.2.1. Assessment practices

To evaluate students’ level of competence, the educational system primarily relies on on students’ grades or other products of normative assessment (i.e. such as letters in alphabetical order, percentages, or value judgments) as an indicator. This form of assessment is widely used in the education system (Hamilton, 2003). The explicit aim of normative assessment is to provide students and parents with clear information about their level of competence (Williams, Pollack & Ferguson, 1975). However, when compared to other assessment methods, normative assessment presents an additional feature which is to create a visible and quantifiable way of evaluating students’ performance. On the one hand, the visibility created by this method is its main allure, and on the other, it can become its main shortcoming (Pulfrey, Buchs, & Butera, 2011). This specific feature of normative assessment gives the possibility to compare and rank the students. Whereas this characteristic can be helpful in a professional context, its advantages tend to disappear when used on students who are still in the

process of learning. Many scholars who have discussed the implications of normative assessment for learning find that there are some disadvantages to its use (Butler & Nisan, 1986; Cliffordson, 2008; Jasso & Resh, 2002).

2.1.2.1.1. Normative assessment

2.1.2.1.1.1. Effects of social comparison on learning and performance

The high visibility of grades (or of other forms of normative assessment) allows students to easily compare themselves with each other. The very way in which learning and evaluation situations are structured in classrooms often favors this comparison process (Pepitone, 1972; Rosenholtz & Simpson, 1984). Students work on the same tasks at the same time with identical instructions. Even though comparison processes are triggered automatically by the mere presence of others, the visibility of differences in performance afforded by normative assessment exacerbate social comparison among students comparatively to ther assessment methods (Goudeau, 2016; Hayek, Toma, Oberlé, & Butera, 2014).

On the one hand, social comparison has been known to produce positive outcomes (Festinger, 1954). For instance, a study conducted by Huguet, Dumas, Monteil and Genestoux (2001) has shown that students who engage in modest upward comparison presented a higher level of improvement over the school year when compared to students who did not compare themselves. Indeed, engaging in upward comparisons could be beneficial for performance as it could provide useful information about how to improve (see also Butera & Darnon, 2017; Buunk & Ybema, 1997; Taylor & Lobel, 1989). This type of comparison could also lead students to set higher standards of success, which in turn would increase motivation to make the necessary efforts to achieve these new goals. However, the beneficial effects of upward comparison seem limited to chosen comparisons (Dumas & Huguet, 2011, Major, Testa, & Bylsma, 1991). Forced comparisons on the other hand, appear to be threatening for self-image (Huguet et al., 2009; Muller & Butera, 2007). In a competitive context, social comparison can elicit a threat to the students’ feeling of self-competence (Mugny, Butera, & Falomir, 2001; Selimbegovic, Quiamzade, Chatard, Mugny, & Fluri, 2007). This phenomenon, called competence threat, has been shown to take place in situations in which social comparison is presented as being the result of competence disparities (i.e. some students are more competent than others) instead of a way to focus on the epistemic value of diverse answers from students (Butera, Darnon, Buchs, & Muller, 2006). This leads students to engage their attention on worries about their own competence and self-worth when they could be devoting all of their cognitive resources on the task (Buchs, Butera, Mugny, & Darnon, 2004). The competence

threat increasingly occurs if the target of the upward comparison is considered a standard (Muller & Fayant, 2010). Moreover, the visibility provided by normative assessment can also be detrimental to low-achieving students’ performance. As previously mentioned, Monteil & Huguet (1993) demonstrated that when assessment is visible, the performance of low-achievers is reduced compared to that of high-achievers. The difference of performance disappears when visibility of assessment is removed. In short, the advantages provided by normative assessment (i.e. emphasis on social comparison) may be useful for selective educational systems as it provides an easily identifiable hierarchy, but it can also hold negative consequences for students who could be focusing their attention on comparing themselves to other students rather than concentrate on the task at hand.

2.1.2.1.1.2. Effects of grades on motivation and competitive behaviors

Furthermore, social comparison, induced by normative assessment, reduces perceptions of task autonomy and intrinsic motivations and encourages performance-avoidance goals, which are associated with a host of maladaptive consequences in school (Pulfrey et al., 2011; Pulfrey, Darnon, & Butera, 2013). Specifically, performance-avoidance goals are linked to negative effects on self-esteem, perception of control and performance (Cury, Elliot, Da Fonseca, & Moller, 2006; Harackiewicz, Durik, Barron, Linnenbrick, & Tauer, 2008). Therefore, instead of promoting improvement, task autonomy and focus on the task, all of which are important factors to guarantee good learning conditions, normative assessment, through social comparison, enhances comparison-related considerations. Specifically, the use of grades is associated with perception of competitive social comparisons. In a series of studies conducted by Hayek and colleagues, the presence of grades interfered with the capacity of individuals to consider information coming from others in an unbiased way and hampers information sharing (i.e. cooperatively useful behaviors) in cooperative tasks (Hayek et al., 2014; Hayek, Toma, Oberlé, & Butera, 2015; Hayek, Toma, Oberlé, Guidotti, & Butera, 2017).

2.1.2.1.1.3. Effects on intergroup stratification

The visibility provided by normative assessment might not only hinder individuals’ levels of performance but can also have an effect on intergroup relations and groups’ level of performance. In the quest for social recognition, social groups also compete over available academic resources. In the current educational system, grades also represent a form of rewards which provide students with status, and privileges. Those privileges can be symbolic (i.e. good reputation) or material (i.e. academic opportunities in the tracking system; Felouzis &

Charmillot, 2013). These academic rewards are made visible by normative assessment, allowing others to observe which social groups are able to succeed and are given high status.

As mentioned in the previous chapter, current school settings and expectations can generate differences in groups’ level of achievement (Croizet & Claire, 1998; Darnon et al., 2017; Stephens et al., 2012). Yet, by ignoring certain advantages enjoyed by members of privileged groups, educational systems visibly accentuate the differences in performances with normative assessment. This can potentially render educational context more threatening and make it harder for members of unprivileged groups to attain academic benefits, such as access to higher tracks or diplomas (Felouzis & Charmillot, 2013). Indeed, the aforementioned studies conducted by Monteil & Huguet (1993) showed that high-achieving and low-achieving students alike are more likely to underperform after a negative feedback if the evaluation is rendered visible. Importantly, Goudeau & Croizet (2017) demonstrated that the visibility of performance can be transposed to social comparison affecting social groups, as their studies found that when achievement was visible, low-SES students produced lower performance.

By clearly and visibly identifying through assessment which students are worthy of academic rewards, school highlights which are the high and low-achieving students. These categories can become intertwined with certain expectations of academic performances (Resh, 2009). For instance, Unzueta and Lowery (2010) showed that participants, regardless of their own racial background, expect racial differences on mathematics test achievement, such that they predict that Asian-Americans would get better results than Whites followed by Blacks and Latinos. As privileged groups are continuously overrepresented in the high-achieving categories, this state of affairs perpetuates the representation of a social order, in which certain social groups “rightfully” deserve rewards allocated by the education system (Sabbagh, Resh, Mor, & Vanhuysse, 2006). Visible discrepancies in different social groups’ performance contribute, in a meritocratic selective context, to the notion that certain groups deserve their high-status and belong in the school culture (Resh, 2009).

In sum, in a competitive context, normative assessment can accentuate the performance differences between various social groups, creating in the process a social scale of academic achievement justified by meritocratic principles regulating the distribution of grades. As such, normative assessment can become a social regulation tool, which by increasing social comparison processes between individuals (and by extent their social groups), contributes to psychological processes linked to the maintenance of existing hierarchies.

2.1.2.2.2. Formative assessment

This wide range of criticism has led several scholars to contemplate other methods of assessment which could counteract some of the negative aspects created by normative assessment; one of which is formative assessment. Formative assessment is a method that has been widely discussed by scholars in educational sciences (Clark, 2012; Dixon, Hawe, & Parr, 2010; Williams, 2010). Generally, formative assessment consists of a detailed feedback on the students’ performance. This feedback focuses on the students’ personal mistakes and provides them with information on how to overcome the difficulties they encounter. Its personalized setting allows students to target their own strengths and weaknesses, and focus them on their own progression (Hamilton, 2003). Some authors such as Black and Wiliam (1998) define this type of assessment more broadly to include any activities undertaken by teachers and students to get information that can be used to alter teaching and learning. This definition would therefore include: teacher observation, classroom discussion, and analysis of student work (including homework and tests). The central component of formative assessments is that the information collected from tests is used to adapt teaching and learning to meet students’ needs. In sum, the main objective of formative assessment is to identify gaps between desired goals and current knowledge and skills to guide the students towards the goal (Boston, 2002). This method avoids pressuring students to compare themselves with others. It also provides more specific information on performance than normative assessment does and has been linked with positive effect on learning (Black & Wiliam, 1998; Frey & Schmitt, 2007; Marcotte & Hintze, 2009). According to an extensive review of 250 articles and chapters, the learning gains, as measured by comparing the average improvement after formative assessment with control groups of students who performed on the same test, produced effect sizes ranging from .4 to .7. Low-achieving students and students with disabilities benefited more from this assessment method than did other students (Black & Wiliam, 1998). Indeed, formative assessment could be in particular useful to low-achievers because it focuses the student on improvement and effort (Boston, 2002). Finally, formative assessment increases the chance for students to devote a great deal of effort to their learning outcomes and it has been shown to best support academic success (Dalbert, Schneidewind, & Saalbach, 2007).

Considering the advantages provided by formative assessment in terms of learning outcomes (i.e. the fulfilment of the main objective of education), one might wonder if it is the additional selection function of normative assessment which explains its prevalence. Even though other evaluation methods have been demonstrated as being strong educational tools, they do not facilitate the second function of the education system. Normative assessment, unlike

formative assessment, is a useful tool to serve the function of selecting students (Cliffordson, 2008; Jasso & Resh, 2002). This reason could potentially explain their continuous widespread use despite some of the observed negative consequences for students.