A first application of multilevel latent class analysis (MLCA) to educational large-scale assessment data is demonstrated. This statistical technique addresses several of the challenges that... Show moreA first application of multilevel latent class analysis (MLCA) to educational large-scale assessment data is demonstrated. This statistical technique addresses several of the challenges that assessment data offers. Importantly, MLCA allows modeling of the often ignored teacher effects and of the joint influence of teacher and student variables. Using data from the 2011 assessment of Dutch primary schools’ mathematics, this study explores the relation between the curriculum as reported by 107 teachers and the strategy choices of their 1,619 students, while controlling for student characteristics. Considerable teacher effects are demonstrated, as well as significant relations between the intended as well as enacted curriculum and students’ strategy use. Implications of these results for both more theoretical and practical educational research are discussed, as are several issues in applying MLCA and possibilities for applying MLCA to different types of educational data. Show less
Hickendorff, M.; Mostert, T.M.M.; Van Dijk, C.J.; Jansen, L.L.M.; Van der Zee, L.L.; Fagginger Auer, M.F. 2017
National large-scale assessments have demonstrated a decrease in Dutch primary school students' performance on multidigit multiplication and division problems (e.g., 56x23 and 544:34) in the... Show moreNational large-scale assessments have demonstrated a decrease in Dutch primary school students' performance on multidigit multiplication and division problems (e.g., 56x23 and 544:34) in the past two decades. This performance decrease appears to be related to changes in students' use of mathematical strategies: the use of relatively accurate strategies for which calculations are written down has decreased, while the use of inaccurate mental strategies has increased. To gain more insight into these developments and how to possibly turn them around, factors that affect students' mathematical strategy use and performance were investigated. New applications of latent variable models to national large-scale assessment data were used to relate teachers' reports of their mathematics teaching to students' problem solving. Follow-up experiments were conducted at primary schools for a more controlled investigation of students' strategy choices and of how to influence those choices (and thereby, performance) favorably. It was found that teachers mainly influence students' choices between strategies with written work, and that they have less influence on choices for the inaccurate mental strategies (made more frequently by boys and weaker students). Extra instruction targeted at the latter was found to increase weaker students' choices for writing down calculations, and this improved students' performance. Show less
Fagginger Auer, M.F.; Hickendorff, M.; Putten, C.M. van 2015
Inferring mathematical strategy use from primary school students' written work Verbal reports of mathematical strategy use can be inaccurate and can influence students' spontaneous strategy choices... Show moreInferring mathematical strategy use from primary school students' written work Verbal reports of mathematical strategy use can be inaccurate and can influence students' spontaneous strategy choices, and the effort involved in collecting them limits sample sizes. These disadvantages can be overcome with a straightforward non-verbal method: students' strategy use can often be inferred from the calculation steps that they write down as a natural part of their solution process. In this study, results of verbal and written strategy identification are compared in a recent large-scale mathematics assessment. In addition, the application of written strategy identification is illustrated with two examples. The first application is a study in which students' strategy use on a large-scale assessment was related to their teachers' mathematics instruction. The second application is a study which compared the strategy use of a large group of students on problems presented with and without a realistic context. Finally, the possibilities and limitations of written strategy identification are discussed. Show less
Fagginger Auer, M.F.; Hickendorff, M.; Putten, C.M. van 2018
Welke factoren in het onderwijsleerproces hangen samen met de rekenwiskundeprestaties van leerlingen in het basisonderwijs? Deze vraag stond centraal in een review-studie die op verzoek van de... Show moreWelke factoren in het onderwijsleerproces hangen samen met de rekenwiskundeprestaties van leerlingen in het basisonderwijs? Deze vraag stond centraal in een review-studie die op verzoek van de Inspectie van het Onderwijs en NRO is uitgevoerd. De focus lag op factoren in het onderwijsleerproces die beïnvloedbaar zijn: knoppen waaraan gedraaid zou kunnen worden. Allereerst schetsen we een beeld van de rekenwiskundeprestaties van Nederlandse basisschoolleerlingen aan de hand van de meest recente peilingsonderzoeken in groep 6 en 8 en de eindtoetsen. Uit deze gegevens blijkt ook dat de verschillen in rekenwiskundeprestaties tussen leerlingen maar voor een klein deel, hooguit 10 procent, toe te schrijven zijn aan verschillen in het rekenonderwijs dat zij krijgen op school. Vervolgens zijn de resultaten uit internationaal en Nederlands onderzoek in kaart gebracht om de vraag te beantwoorden welke factoren van het onderwijsleerproces effectief zijn. Dit leverde een grote verzameling op van kenmerken van de les die positief samenhangen met reken-wiskundeprestaties, zoals een verscheidenheid aan interventies met specifieke instructie- en werkvormen, het toepassen van technologische en niet-technologische hulpmiddelen en formatieve toetsing. Het is opvallend dat er geen robuuste onderzoeksresultaten zijn gevonden over de samenhang van leerstofaanbod of rekenmethode met reken-wiskundeprestaties. Verder vonden we een aantal kenmerken van de leerkracht, zoals vakinhoudelijke kennis over rekenen, en van de leerling, zoals rekenangst, die samenhangen met reken-wiskundeprestaties. Over de rol van kenmerken van de klas en van de school, zoals aanwezigheid van een rekencoördinator, zijn weinig onderzoeksresultaten gevonden. Alles bij elkaar genomen lijkt de leerkracht de cruciale rol te vervullen in het vormgeven van goed reken-wiskundeonderwijs. Show less
