Balancing the challenges of high-stakes testing, accountability and students’ well-being

In the lead article in this regular issue, Stenlund, Eklöf, and Lyrén (2017) have investigated whether different groups of test-takers vary in their reported test-taking behaviour in a high-stakes test situation. The study involves N = 1129 students (63.2% females, mean age 22) who sat for the Swedish scholastic aptitude test (SweSAT) – a paper and pencil test with multiple-choice items used for selection to higher education in Sweden. SweSAT measures what is seen as general abilities and knowledge. Sweden has a centralised admission system with the regulations that in cases where there are more applicants than places to higher education, the candidate should be admitted based upon their SweSAT scores and upper secondary school GPA. In other words, the stakes are high when students sit for the SweSAT. Since the test was introduced in 2011, between 40,000 and 80,000 test-takers sit for the test twice a year. The authors report that as many as 25% are repeaters, but in their study, they have focused only on students who sat the test for the first time. The participating students responded to an online questionnaire, reporting on their test-taking strategies such as structural organisation, time management and testwiseness strategies. The authors also included items on test anxiety and test motivation.

One of the findings shows that high achieving students skip answers they are not sure of to make sure they do as many items as possible, while low achievers to a greater extent reported it was more important to ensure accuracy in their answers than answering a maximum number of items. The low achievers further reported a higher degree of test anxiety than the high achievers, and less test motivation. Regardless of achievement level, female test-takers reported a higher level of test anxiety, even though females reported using more ‘time management strategies’ than males and they made sure to have time enough to cover all items on the test. The authors suggest that instructions and information on effective test-taking behaviour before, during and after tests should be implemented to support students, not only to avoid biased test scores, but also because such skills might be of importance to be a successful test-taker and future student. As tests are getting increasingly more high-stakes for students, researchers and practitioners need to discuss how to support students not only to be able to reach their academic potential, but also to make sure students’ well-being is at the heart of assessment reforms for all students. How to balance the challenges of high-stakes testing and students’ well-being needs to be taken more into account.

In the area of accountability, Crowe, Rivers, and Bertoli (2017) argue that teachers of students with special education needs often are left behind. In US, the No Child Left Behind act has drawn attention to the need for standards-based accountability systems (Polikoff, McEachin, Wrabel, & Duque, 2014), as teachers are currently experiencing different systems and procedures in US. But as the authors point out, there is currently an absence of observation protocols for special education classrooms, and therefore the authors have reviewed classroom tools presently used by researchers and practitioners. Their study yielded 104 tools, of which nine are presented and discussed in the article (the full list of 104 tools is published online for our journal). Across the tools, the researchers detected the following constructs: organisation instruction and climate of classrooms, one construct unique to special education, paraprofessional involvement, and finally one construct unique to the tools originating from practice, professional behaviour. Of particular interest, is the assessment of students’ well-being, the emotional climate and the growing body of literature indicating that students who feel connected to their teachers and peers are more motivated to learn, perform better and demonstrate stronger social competence. The review is a reminder of the importance of focusing upon all students. Further, the authors warn that ‘one size does not necessarily fit all when it comes to how to hold teachers accountable, particularly among teachers serving students with special needs’. The researchers suggest that observation tools, which are better tailored to the specific classroom, have the potential to improve the quality of the data from which the teachers are held accountable.

Baird, Meadows, Leckie, and Caro (2017) have evaluated ‘rater accuracy’ with rater-monitoring data from high-stakes tests examinations in England, using a cross-classified multilevel modelling. Data included training and monitoring of 576 raters in 110 teams, across 22 examinations. The research team compared two rater quality check systems used in England, one ‘Expert based monitoring system’ where correct scores were generated by the Principal Examiner, and other ‘Supervisor-based monitoring’ system, where correct scores were generated by rating team supervisors. The study involved a within-group comparison, as the raters and supervisors used both monitoring systems. One of the findings from the study is that the face-to-face training coupled with supervisor-based monitoring is likely to underestimate group-training effects and overestimate rater effects. The authors point out that face-to-face training and supervisor-based monitoring is still the norm in many high-stakes examinations for practical reasons, but they further suggest that online training and expert-based monitoring should produce more accurate depictions of rater severity effects. Baird, et al. further remind us that in settings such as Hong Kong, Korea, Australia and US, marking technology now holds the promise of more accurate rating.

In the article Educational assessment in the Republic of Korea: Lights and shadows of high stake exam-based education system, Kwon, Lee, and Shin (2017) present a country profile with particular focus upon the college entrance examination system and its effects upon students’ well-being. Like the SweSAT test in Sweden, students’ futures depend upon the outcome of a high-stakes test, more specifically the Korean SAT, CSAT which is designed to measure the students’ scholastic ability for college education. It is developed and administrated by the Korea Institute for Curriculum and Evaluation and more than 600,000 students take the test annually. Like with the SweSAT, 20% are re-takers. Unlike Sweden, the high-stakes test in Korea has made families spend an enormous amount of money on private education, such as the cram schools (hakown) and private tutoring. In fact Kwon et al. (2017) state that the equivalent amount of money going into education in a major city in Korea is now spent on private education preparing students for high-stakes tests. The authors use different surveys, including PISA 2012 to explain that the high-stakes testing and exam-driven system in Korea now faces challenges regarding students’ well-being, including low intrinsic motivation. In fact, most Korean students’ drive and motivation is now closely linked to instrumental motivation and how to enter a good university to secure future success. The authors suggest that in order to make more positive educational reforms in Korea, the politicians need to decrease their reliance on high-stakes testing, and focus more on assessment that supports students to develop their talents and dreams. Again, the question of students’ well-being is rising, and the question is how to accommodate assessment in the future which can do that.

Appelrouth, Zabrucky, and Moore (2017) have investigated students’ preparation for admission testing (SAT) for College in the US by using archival data from private tutoring centre in Atlanta, Washington and New York. A total number of 1947 students participated in the study. Analysis revealed significant effects of ‘time on task’ and ‘rate of SAT homework completion’ on the students’ SAT scores. Further, official SAT administrators contributed more to higher SAT score improvements than unofficial practice tests. Appelrouth and colleagues remind us that more affluent students would secure more expensive forms of test preparation, a pattern which has also been called ‘shadow education’. In times where teachers are held accountable for students’ results, we need a conversation also regarding the use of ‘out of schools’ support such as cram schools and test preparation activity, especially since these schools are increasingly global phenomena (Elwood, Hopfenbeck, & Baird, 2017).

The final paper in this issue, written by Acquah and Malpass (2017), outlines a proposal for the technical baccalaureate in England, which is intended to deliver the necessary training to enable young people to fill the UK’s skills gap in intermediate-level occupations in STEM and other sectors. In times of changes in 14–19 education in England, the paper offers a timely overview of historic events, with the aim to avoid mistakes from the past and by including previous good practice in the current reforms and changes in England.

This issue also includes a book review written by Deirdre Ni Chróinín, who has reviewed Assessment in Physical Education, a sociocultural perspective, authored by Peter Hay and Dawn Penney. The book review, which highly recommends the book, indicates that it is suitable not only for researchers and practitioners in the field of physical education, but also suggests it would offer insights to researchers interested in assessment from a sociocultural perspective in general.

Therese N. Hopfenbeck
Oxford University Centre for Educational Assessment Department of Education
University of Oxford, Norham, UK
[email protected]