The current deployment of digital tools in schools makes it possible to carry out, via digital platforms, national assessments of the knowledge and skills acquired by students.

In 2017, for the first time in France, all sixth grade students were evaluated in French and mathematics via a digital platform. It was the first time that such an operation was conducted in Europe on this scale (810,000 students in more than 7,000 schools).

Four national assessments at the beginning of the cycle (CP, CE1, 6^eand 2de) and one in the middle of the cycle (for CE1) are generalized at the start of the 2018 school year.

• These so-called "benchmark" assessments are intended to provide teaching teams with a precise "overview" of each student's skills in French and mathematics in order to help teachers implement personalized support.
• They also allow for local management: the aim is to provide the ministry's decentralized authorities (rectors, DASEN, IENs) with indicators that will enable them to better know and understand the results of schools.
• Finally, they make it possible to measure the national performance of the school system and, ultimately, to evaluate the impact (at different scales) of education policies.

In 2018, two national assessments were conducted online: the sixth grade assessment (in French and mathematics, as in 2017) and the positioning tests at the entrance to the second grade (which aim to identify achievements and needs in French and mathematics, 740,000 students).

The assessments at the beginning of CP (which aim to assess students' achievements in French and mathematics) and at the beginning of CE1 (which aim to assess reading, writing and numbering skills) were, on the other hand, carried out last September "on paper" (via workbooks). The Ministry is experimenting in 2018 with taking these two tests on tablets.

Standardized tests and automatic controls

Thus, for the assessment of 6^eIn October, the evaluation consisted of two tests, in French and mathematics, each lasting 60 minutes (10 minutes of preparation and 50 minutes of testing):

In French, each student in 6^e was to solve exercises on :

• understanding of the written word: literary texts, documents, images
• oral comprehension: audio and video supports
• knowledge of language study: spelling, grammar, vocabulary

• his knowledge of numbers
• calculation and problem-solving skills
• knowledge of geometry and measurement

National performance and individualized feedback 

Anonymized, the results are used by the Direction de l'évaluation, de la prospective et de la performance (DEPP), which will publish a summary around the summer (as it did last year for the 2017 national assessment of sixth graders).

An efficient observation instrument...

In a report devoted to the organization of the evaluation function of the education system, the deputies Régis Juanico and Marie Tamarelle-Verhaeghe, underline the "irreplaceable" character of national evaluations.

• "They have become necessary because any school system, if it does not want to be guided by beliefs - or ideological convictions - must measure the effects of its action to verify that they are in line with its objectives.
• "They are made possible, at a reasonable cost, thanks to advances in computer technology. The Cour des Comptes estimates that the cost of carrying out the evaluation (putting the tests on line, correcting them, distributing the results to the students and schools) for 810,000 sixth-grade students is 430,000 euros, or 0.50 euros per student.
• "Unlike traditional assessments developed by teachers for their classrooms, they are subject to quality controls, particularly with regard to their psychometric properties, i.e., their ability, if designed according to scientific standards, to report on the cognitive achievements of students when they answer several items.
• "They also provide an accurate view of student learning because, although they are subject to measurement errors, these inaccuracies can be calculated - this is the confidence interval - and in the vast majority of cases result in non-significant differences
• "Finally, they provide a relevant view of students' achievements, as they can be correlated with their career path.

Risk of bias and technical limitations

• "These tools can fuel cheating phenomena, which have been highlighted by researchers and which result from the preparation of students, by their teachers, for tests (teaching to the test).
• "The assessment contexts can also influence the cognitive activity and performance of some students by calling on their "attentional" resources, which then become less "mobilized" to carry out the tasks requested. The tests are therefore likely to cause these students to fail, especially when they anticipate this failure because of their school history or their membership of certain social groups. Thus, these assessments do not always provide automatic access to students' real abilities and achievements.
• "Finally, the way in which the tests are taken, particularly the computer-based tests, can destabilize some students or do not allow for the evaluation of complex skills, such as creative writing, group work, oral argumentation or sense of initiative, as is the case with the MCQs focused on writing.

Assessments that must have an educational purpose

In this regard, they make a series of recommendations so that "the pedagogical vocation of national assessments focused on primary and secondary schools is better ensured.

• "Teachers should be involved in both the upstream and downstream phases of these evaluations.
• "To ensure that the purpose of these assessments is understood by all, the knowledge measured and the items proposed should refer explicitly to the knowledge and skills in French and mathematics of the Common Base
• "The assessments should take into account the elementary and middle school cycles, each of which encompasses three levels of education and thus three school years. This 'curricular' conception of schooling should be respected by standardized tests, as it is designed to give students time to achieve the goals set at the end of each cycle.

Contextualize test results

• "The results obtained on standardized assessments do not express an "absolute value" of the performance of the students who take the tests. These are in fact influenced not only by the evaluation context, but also by the social origin, the cultural background and the previous schooling of the children.
• "Student outcomes should be contextualized and not used to assess the value of teachers and schools or to regulate resource allocation.

National assessment at the beginning of sixth grade 2017: contrasting levels of mastery according to academies and student characteristics

• At the beginning of the sixth grade, 85% of students have acquired the knowledge and skills expected in the French program and 73% in mathematics
• Nationally, at the beginning of the 2017-2018 school year, 85.3% of early sixth graders have satisfactory or very good command of French knowledge and skills.
• The results are lower in mathematics, where the elements necessary for the acquisition of knowledge and skills are correctly assimilated by 73.2% of students.
• In French, 88.4% of girls have a satisfactory or very good command of the knowledge and skills assessed. They were 6 points ahead of boys (82.3%). On the other hand, these differences are no longer apparent in mathematics, where the levels of mastery thus measured are 73.7% for boys and 72.7% for girls.
• Students in public schools belonging to a REP+ (about 7% of sixth graders) have particularly significant difficulties.
• Students entering the sixth grade in the private sector have a better command of the knowledge and skills assessed than students entering the public sector outside of priority education (+6 points for French, +8 points for mathematics).
• In the collèges with the most socially advantaged students, the mastery rates, measured by the combination of "satisfactory mastery" and "very good mastery", are close to 90% (93.8% for French, 86.2% for mathematics).
• In the least advantaged schools, the mastery rates are 72.2% and 54.5% for French and mathematics respectively.

Références : 1 2 3 4