Adaptive assessment

A

18

beings. Certainly, this lack of closure and certitude in taking a dialectical stance will be frustrating to many.

Overall Assessment of Activity Theory in Science Learning What are science educators to make of activity theory? It has been claimed to be able to overcome dichotomies that have plagued education such as individual/collective, body/mind, intra-/ inter-psychological, and so forth. While these goals are still being worked out, at the very minimum it sensitizes us to view learning as an ongoing orchestration of people and cultural artifacts in practices (activity systems) where the past and the present are intertwined. It also inspires us to see the potential for human(e) development when societal contradictions are surfaced, critiqued, and overcome. This is an exciting but extremely difficult endeavor; human learning is multidimensional and complex, which science educators have overwhelmingly theorized at the level of the individual learner. Activity theorists will therefore continue to plod on in their research long after where Vygotsky had left off.

Cross-References ▶ Assessment Framework ▶ Communities of Practice ▶ Dialogic Teaching and Learning ▶ Emotion and the Teaching and Learning of Science ▶ Heterogeneity of Thinking and Speaking ▶ Socio-Cultural Perspectives on Learning Science

References Cole M (1996) Cultural psychology: a once and future discipline. Harvard University Press, Cambridge, MA Engestro¨m Y (1987) Learning by expanding: an activitytheoretical approach to developmental research. Orienta-Konsultit, Helsinki

Adaptive Assessment Langemeyer I, Nissen M (2011) Activity theory. In: Somekh B, Lewin C (eds) Theory and methods in social research. Sage, London, pp 182–189 Roth W-M (2010) Activism: a category for theorizing learning. Can J Sci Math Technol Educ 10:278–291 Roth W-M, Lee Y-J, Hsu P-L (2009) A tool for changing the world: possibilities of cultural-historical activity theory to reinvigorate the field. Stud Sci Educ 45:131–167

Adaptive Assessment Elena Papanastasiou University of Nicosia, Nicosia, Cyprus

Adaptive Testing Adaptive assessment can be defined as any type of assessment that is tailored specifically to each examinee, based on their performance on previous items on the assessment. Most adaptive assessments are based on the theories and advances of Item Response Theory (IRT). More specifically, in IRT the examinee ability estimates, as well as item characteristics such as the item difficulty, are placed on the same continuum. This allows for the administration of items that are matched to the estimated ability level (y), of each examinee, at each point of the assessment. Therefore, adaptive assessments allow for the administration of items that are targeted to the ability level (or trait level) of each examinee, which enables the estimation of more accurate examinee ability estimates. For example, if an examinee responds correctly to item 1, their estimated ability will increase, so the second item that will be administered will be of higher difficulty than the first item. If the examinee responds incorrectly to item 2, the examinee’s estimated ability will drop slightly, so the third item that will be administered will have a level of difficulty in between the difficulty levels of items 1 and 2. By administering more items that are specifically targeted to each examinee’s ability, a more accurate ability estimate is achieved.

Affect in Learning Science

Adaptive assessments come in contrast to linear, nonadaptive assessments where all examinees respond to the same or equivalent forms of a test in a predetermined order. One problem with nonadaptive assessments is that the majority of the items administered are targeted to examinees in the middle of the ability continuum. Therefore, linear tests typically include a large number of items of average difficulty and few items of lower and of higher difficulty. This creates problems for the accurate estimation of examinees at the extremes of the ability continuum, as low ability examinees will find the items at the middle of the ability continuum too difficult, whereas high ability examinees will find such items too easy. Consequently, nonadaptive assessments tend to provide little information for high-achieving and low-achieving examinees, the ability estimates of whom therefore include large amounts of measurement error. Some of the advantages of adaptive assessments are those of increased measurement accuracy for examinees at all ranges of the ability continuum and item efficiency since fewer items are needed to reach the same level of accuracy as with linear tests. Additional advantages of adaptive assessments when they are administered electronically are those of immediate scoring and reporting and more frequent test administrations. Some disadvantages of adaptive assessments include (a) the considerable initial costs of creating and calibrating large item pools that are needed for such assessments, (b) the inability of the examinees to go back and change their answers on most adaptive assessments which can create anxiety and frustration to some examinees, as well as (c) the security issues related to the compromise of the item pool due to the overexposure of some items. Adaptive assessments can take various forms, based on their degree of adaptivity. Fully adaptive assessments are those where every item is matched to the examinee ability estimate with the only goal of increasing the amount of information on each examinee’s ability. Other types of adaptive assessments administer groups of items together, as a testlet, and are called multistage adaptive tests. In other cases, due to various

19

A

content constraints and problems with the overexposure of certain items, the assessments are called Barely Adaptive Assessments. For most types of adaptive assessments, due to the extensive computations that are required, they are typically administered on a computer and are frequently called Computerized Adaptive Tests (CAT).

Cross-References ▶ Assessment: An Overview ▶ Computer-Based Assessment ▶ Test

Advance Organizer ▶ Meaningful Learning

Affect in Learning Science Steve Alsop Faculty of Education, York University, Toronto, ON, Canada

Keywords Affect; Content; Learning; Pedagogy

Introduction In the current era of education where there is so much emphasis on cognitive educational outcomes and accountability, it can be difficult to recognize the importance of affect in learning science. Today, much of the public debate about and rationale for education sees the very basis of that education being best captured by accounts of instructional efficiencies, curriculum statements, lesson plans, and public records of pupils’ performance. This is at best only a partial picture, and in

A