Changing Mental Models on Assessment – Start With The Why!
With 32 years of experience in ELT, Isabela Villas Boas holds a master’s degree in Teaching English as a Second Language from Arizona State University and a Doctorate in Education from Universidade de Brasília. She is the Corporate Academic Manager at Casa Thomas Jefferson, Brasília. She has presented in various local, national and international ELT conferences and published articles in journals and book chapters for over twenty years. Her latest book is Teaching EFL Writing – A Practical Approach for Skills-Integrated Contexts.
According to Coombe et al. (2007), tests or tasks administered at the end of the course to determine if students have achieved the objectives set out in the curriculum are called summative assessments. They are often used to decide which students move on to a higher level. Conversely, formative assessments are carried out with the aims of using the results to improve instruction, so they are given during a course and feedback is provided to students. Jang (2014, p. 5), states that “assessment is an activity whose primary purpose is pedagogical, that is, to help teachers plan instruction and guide student learning. If assessment results are not used to inform instruction and guide student learning, they are of little use.” Thus, it seems only logical to focus more on formative than on summative assessment in the language classroom.
In the institution I work for, we have been trying to change our assessment system from a summative to a more formative one for quite a while now. The main aspect of this change lies in offering students more frequent opportunities of assessment, by way of various types of instruments, without accumulating content and with the opportunity to revisit the standards that were not reached and redoing the assessment.
However, we have faced a few challenges in the process. The easiest ones to address are those that involve training teachers to use the new assessment system and helping students and parents understand how it works, that is, the mechanical aspects, the how. The hardest ones are those that involve changing teachers’ and parents’ mental models about assessment, in other words, the why. Why is it so hard to change the way people generally see assessment? What are some myths that persist in people’s minds about assessment that keep them from changing their lenses? According to Senge (2000), our mental models, that is, our theories about how the world works, influence our actions. Unexamined mental models limit our ability to change.
Sinek (2011) explains that it is very common for leaders to try to encourage others to take action by focusing on the what and the how. However, he argues that great leaders are the ones who start with the why. In this article, I will address four reasons why some teachers, managers, and parents hold on to their traditional mental models on assessment and why these mental models need to be reexamined and changed. It is only when we understand why people think the way they do that we can deconstruct their view of the world and lead them to a process of change.
Three mental models that keep us from changing
1.Most of us, administrators, teachers, and parents who grew up in the 80’s or 90’s, did not experience formative assessment in our educational trajectory
We tend to believe that the way we grew up and what we experienced as students is what is right. That is one of the reasons why it is so hard to implement changes in Education. We are always one generation behind. In the case of testing and assessment, many of us experienced a time when a good test was a tough test, preferably with tricky items that only the very bright were able to figure out. The test items were not necessarily aligned with instructional activities. Many times, we encountered a specific type of task for the first time on a test, not in the classes that led to the test. Also, tests were always the most important component of a “decent” assessment system. There was an attempt to add other types of assignments, but they were not taken as seriously, either by the schools or by the students.
In the high school I went to, there were projects for each part that we could work on in groups and that counted towards the final grade, but they were worth very little compared to the test grades, and we carried them out totally independently, so it wasn’t a guided learning process. The teachers provided the instructions and set the date to hand them in, without check points along the process. In fact, I was part of a group in which we organized ourselves so that each one of us would do the project of one of the subjects and included everybody’s name in it. That is, instead of working cooperatively, we divided the work up among ourselves and did it individually. Thus, my experience with grades other than summative tests was only with these types of projects.
Adult students, teachers, managers, and parents who only had this type of experience in their schooling tend to think that giving students shorter and more frequent assessments and allowing them to retake the ones they did not do well in makes the system “weaker” and jeopardizes the learning. They associate effective learning with tough tests: the harder it is to pass the test, the “stronger” the school or the course is. We commonly hear some say that “life does not give us second chances, so why should assessments do so?” It is not really true that life does not give us second chances. In most of our life experiences, we fail once, twice, or three times and try again and again before we reach our objectives.
Our school experiences mold our mental models. We end up believing that what we experienced is what is right. After all, weren’t we all successful in the end? It is difficult for us to see that there is a different way of doing things and still be successful, and, even better, guarantee that everyone is successful, not only the brightest.
2.We confuse standardized, high stakes tests with classroom assessment
Many countries have standardized tests that students take every year or after a certain number of years. Some also have high-stakes university entrance exams that are very competitive. These exams must follow rigorous psychometric principles so that they can be considered valid and reliable. Due to their massive administration, many of these exams only have selected-response types of items. Otherwise, they would be too expensive to administer and to score.
In English-language teaching (ELT), we have the international proficiency exams, such as the Cambridge Exams, the TOEFL, IELTS, and others. They, too, must adhere to very rigorous psychometric parameters because they are high stakes. Some are quite expensive because they balance selected-response and performance-based items. Still, they follow a rigorous format and do not allow for much creativity from the students.
The fact that students have to take these types of exams at one, most of the times “final”, point in their educational cycle leads many people to believe that these exams should always dictate how classroom assessments should be designed. The view of the world that permeates this belief is one that sees classroom tests as tools to prepare students for the standardized tests and that, because of this, they should mirror the standardized tests.
However, classroom tests should reflect the curriculum, and they have to be fully aligned with the learning outcomes and the instructional activities. If most of the classroom time in today’s ELT classroom should be spent with students engaging in authentic interaction among themselves, classroom assessments should reflect this type of instructional activity. If we want students to carry out meaningful projects in class, to use technology to produce artifacts, and to develop their creativity and critical thinking, we should use assessment instruments that are aligned with these principles. In this case, then, it is the projects and the authentic artifacts that should constitute most part of the grade, when a grade is needed, and not a traditional, formal test. Standardized tests have also been highly criticized in education nowadays, as Pastor (2019) portrays. Standardized tests fail to capture what a student actually knows and can do. In the specific case of English Language Learners (ELLs), relying mostly on standardized test scores is even more questionable, for their purpose is to rank-order students against a normed population and make comparison easier. The problem is that these exams are “normed against a population that may not adequately represent language learners” (Jang, 2014, p. 77).
Besides, as shown above, high stakes exams are limited to the types of items that allow them to reach the levels of reliability they need. This is not true of classroom assessments because they are not high stakes. Thus, the concepts of reliability and validity must be seen differently in the case of classroom assessments. The students’ results in this latter type of assessment do not need to be generalizable to the whole population, and neither will their scores be used to compare them to other students or for selection purposes. According to Jang (2014), decisions made to increase reliability often result in decreasing validity, the construct that is the most important in classroom assessment.
Thus, it makes no sense to spend years teaching and testing only to prepare for a proficiency exam. Passing the proficiency exam should be the consequence of good teaching. Of course, perhaps in the last year or semester of the program, the curriculum can concentrate more on preparing students for these exams, but not more than that!
3.We think students should be ranked and that if everyone gets an A, it is because the assessment system is not effective
In a system in which meritocracy prevails, the role of assessment is to rank the students so that the ones at the top can have access to certain exclusive opportunities. Granted, this also helps the school identify the students who need more assistance. Within this view of the world, a good test or assessment system is one that discriminates the high-performing, mid-performing, and low-performing students, and in order to do this, the instruments used have to have items or tasks that are really difficult and that only the high-performing students can succeed in. In other words, if everyone gets an A, there is something wrong with the assessment
Alfie Kohn (2019) has recently questioned this belief that not everybody can get an A. He discusses top-down mandates for school reform that push toward higher standards and more rigor, and he argues that these high standards are those that not every student can achieve. If all students are successful, it should, theoretically, mean that the teacher did a good job, but today it means that the assessment wasn’t rigorous enough. When everyone reaches the standards, they should be raised to ensure that some students will fail. He believes that, in the United States, excellence is regarded as a rare commodity and that ranking creates the misleading impression of permanent failure of some. He also argues that framing excellence in competitive terms doesn’t lead to better performance and that competition many times holds people back.
An approach that challenges this mental model is Mastery Learning, based on the premise that every student can achieve high levels of performance if given enough time. Student skill acquisition is a priority. While in traditional models of assessment students can move on if they achieve 50 or 60% of the standard, in Mastery Learning students move on to the next standard only when they have achieved at least 80%. Mathewson (2018) describes how a fourth-grade teacher adopts this approach in her classroom. She uses frequent assessments to help students understand their strengths and weaknesses and to work to overcome these weaknesses. Students analyze their own data and are provided with individual pathways to improve in the areas they need. The ones who have achieved the standards are involved in enrichment activities, while they ones who haven’t achieved them yet are given special attention. In the end, everyone succeeds, and all students have the chance to get an A. The article explains that adhering to Mastery Learning does not have to be an “all or nothing endeavor”. Teachers can take small steps towards it. They can start by raising students’ awareness of what their testing data means and what they need to work on, becoming self-advocates.
Singapore, one of the leading countries in educational excellence, has banned the ranking of students. The argument is that they want students to focus on learning and not on competition. Marks and grades will be replaced with qualitative descriptors. This leads us to the fourth mental model that needs to be examined.
4.If we don’t see a number on a scale of 1 to 10 or 1 to 100, in other words, a grade, we do not really know how the student is doing
Our attempts to adopt qualitative descriptors based on standards or competencies have been met with criticism by some stakeholders. Though provided with comprehensive descriptions of a student’s level of performance, many parents want to see a grade, a number, a percentage. “Okay, the descriptor says that my child has fully achieved the standard. But what does this mean? Is it a 10, or a 9? How can I compare my child to the rest of the class to see if they are doing well or not if I don’t have a number?
Standards-based assessment is a system in which “teachers evaluate students’ performance on tasks against a set of standards that include distinguishable descriptors of student performance, indicating a range of proficiency levels” Jang (2014, p. 23). Proficiency-level descriptors are often used to evaluate students’ essays, role-plays, oral presentations, or projects. In order to mitigate their use and parents’ expectations, many times we assign numbers to the different levels of performance. However, ideally, we shouldn’t need these numbers, for if the purpose of assessment is to give feedback to students and improve instruction, what really matters is the descriptors of students’ level of performance. These descriptors are what informs students how they are doing and what they need to perform better and reach higher levels.
There is currently a “growing gradeless” movement in education. Its advocates present a number of arguments against grading, as summarized by Blum (2017). According to her, grading requires a uniformity that education is too complex to embrace, grades don’t provide adequate information about performance, they do not motivate students, they lead to the minimax principle, they encourage students to see learning as a game, they lead students to see rules as arbitrary and inconsistent, they teach students to focus on schooling rather than learning, and they do not encourage risk-taking.
As shown above, traditional mental models keep us from being successful in adopting more student-centered and progressive assessment systems due to the lack of buy-in from various stakeholders. In order to change this, we must start by addressing these mental models and explaining why they are outdated, discriminatory, and focus on exclusion rather than inclusion. It is only after we change these mental models that we can focus on how to do it and consider more progressive models, such as mastery-based assessment, standards-based assessment, going gradeless, and assessment as learning, which are part of “school reculturing” initiatives (Jang, 2014).
Blum, S.D. (2017, November 14) Ungrading. Inside Higher Ed. Retrieved from https://www.insidehighered.com/advice/2017/11/14/significant-learning-benefits-getting-rid-grades-essay?utm_content=buffer66e92&utm_medium=social&utm_source=twitter.com&utm_campaign=buffer
Coombe, C., Folse, K., and Hubley, N. (2007) A Practical Guide to Assessing English Language Learners. Ann Arbor, Michigan: The University of Michigan Press.
Jang, E. E. (2014). Focus on assessment. Oxford: Oxford University Press.
Kohn, A. (2019, June 15). Why can’t everyone get A’s? The New York Times. Retrieved from https://www.nytimes.com/2019/06/15/opinion/sunday/schools-testing-ranking.html
Mathewson, T.C. (2018, August 23) What “mastery-based” can look like in the classroom. The Hechniger Report. Retrieved from https://hechingerreport.org/what-mastery-based-can-look-like-in-the-classroom/
Pastor, M. (2019, June 19) Why standardized tests aren’t working for students or teachers. Education Week. Retrieved from https://www.edweek.org/tm/articles/2019/06/19/why-standardized-tests-arent-working-for-teachers.html
Senge, P. M. (2000) Schools that learn: A Fifth Discipline fieldbook for educators, parents, and everyone who cares about education (1st Currency pbk. ed.). New York: Doubleday.
Sinek, S. (2009) Start with why: How great leaders inspire everyone to take action. New York, N.Y.: Portfolio.
Please check the Assessment for the 21st Century English Classroom course at Pilgrims website.
Please check the How to be a Teacher Trainer course at Pilgrims website
Changing Mental Models on Assessment – Start With The Why!
Isabela Villas Boas, Brazil