Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

In der aktuellen Kompetenzforschung werden diverse komplexe Konstrukte erfasst. Die verwendeten Leistungstests resultieren oft in langwierigen Befragungen von Studierenden, welche so einer hohen Testbelastung ausgesetzt werden. Damit gehen Nachteile, wie beispielsweise eine sinkende Teilnahmebereitschaft, einher. Auffallend ist daher der noch seltene Einsatz adaptiver Testformate. Diese weisen eine höhere Effizienz auf als lineare Formate, womit sie die Belastung senken und die Teilnahmebereitschaft erhöhen könnten. In dieser Arbeit wird die Anpassung eines bereits bestehenden Leistungstests von einem linearen in ein adaptives Format beschrieben. Grundlage des Vorhabens ist der Ko-WADiS-Test zur Erfassung der Kompetenz naturwissenschaftlichen Denkens bei Studierenden der Fächer Biologie, Chemie und Physik. Den Kern der Arbeit bilden die messtheoretische Rahmung adaptiver Testformate und die Konzeption des neuen Testinstruments. Aus den Items des linearen Tests wurden mehrere adaptive Teststrukturen entwickelt und in simulierten Befragungen verglichen. Ein Multistage-Test mit drei Stufen und jeweils zwei Schwierigkeitsniveaus wies die höchste Messeffizienz auf und wurde praktisch implementiert. Zum Abschluss der Testentwicklung wurden mit dem neuen Instrument Lehramtsstudierende befragt und Messgenauigkeit sowie Messdauer in der Stichprobe mit denen des linearen Ko-WADiS-Tests verglichen. Durch das neue Testformat konnte eine signifikante Steigerung der Effizienz um 53% (Messdauer -30%, Messgenauigkeit +8%) erreicht werden.

Education --- Psychology --- Social Science / Statistics --- Social sciences --- Adaptives Testen --- Kompetenzdiagnostik --- Naturwissenschaftliches Denken --- Item Response Theory --- Testkonstruktion

Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

Cet ouvrage aborde les outils principaux de la statistique descriptive pour résoudre des questions très concrètes telles que :
* Comment construire ou interpréter de manière critique des tableaux de données ou des graphiques ?
* Comment travailler avec des proportions (ratios, pourcentages, etc.), avec des mesures d'évolution (nombres-indices), de tendance centrale (moyenne, mode, ..) et de dispersion (variance, espace interquartile, ...) ?
* Comment donner du sens aux chiffres ? S’adressant en priorité à des praticiens occasionnellement ou régulièrement amenés dans leur vie quotidienne à produire ou interpréter des données statistiques, sans avoir aucune expertise particulière préalable dans le domaine, l'ouvrage a été résolument conçu dans une perspective d'auto-apprentissage . Écrit dans un langage clair et accessible, il abonde en exemples concrets et exercices corrigés, principalement dans le champ économique, politique, social ou éducatif.Il intéressera les professionnels dans les domaines des sciences sociales, de l'éducation et de la santé mais aussi les professeurs et les étudiants des 1er et 2e cycles en sciences humaines.

HA 29 Theory and method of social science statistics - General works --- Statistics --- Statistique --- Statistics as Topic --- Sciences humaines --- Methodes statistiques --- Problemes et exercices

Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

"Preface Multilevel modeling has become a mainstream data analysis tool over the past decade, now figuring prominently in a range of social and behavioral science disciplines. Where it originally required specialized software, mainstream statistics packages such as IBM SPSS, SAS, and Stata all have included routines for multilevel modeling in their programs. Although some devotees of these statistical packages have been making good use of the relatively new multilevel modeling functionality, progress has been slower in carefully documenting these routines to facilitate meaningful access to the average user. Two years ago we developed Multilevel and Longitudinal Modeling with IBM SPSS to demonstrate how to use these techniques in IBM SPSS Version 18. Our focus was on developing a set of concepts and programming skills within the IBM SPSS environment that could be used to develop, specify, and test a variety of multilevel models with continuous outcomes, since IBM SPSS is a standard analytic tool used in many graduate programs and organizations globally. Our intent was to help readers gain facility in using the IBM SPSS linear-mixed models routine for continuous outcomes. We offered multiple examples of several different types of multilevel models, focusing on how to set up each model and how to interpret the output. At the time, mixed modeling for categorical outcomes was not available in the IBM SPSS software program. Over the past year or so, however, the generalized linear mixed model (GLMM) has been added to the mixed modeling analytic routine in IBM SPSS starting with Version 19. This addition prompted us to create this companion workbook that would focus on introducing readers to the multilevel approach to modeling with categorical outcomes"--

Quantitative methods in social research --- Psychometrics. --- Psychometrics --- Social sciences --- PSYCHOLOGY / Statistics. --- EDUCATION / Statistics. --- SOCIAL SCIENCE / Statistics. --- #SBIB:303H4 --- #SBIB:303H520 --- Measurement, Mental --- Measurement, Psychological --- Psychological measurement --- Psychological scaling --- Psychological statistics --- Psychology --- Psychometry (Psychophysics) --- Scaling, Psychological --- Psychological tests --- Scaling (Social sciences) --- Computer programs. --- Informatica in de sociale wetenschappen --- Methoden sociale wetenschappen: techniek van de analyse, algemeen --- Measurement --- Scaling --- Methodology --- SPSS (Computer file) --- SPSS for Windows. --- Statistical product and service solutions --- Statistical package for the social sciences --- SPSS (Computer file). --- EDUCATION / Statistics --- PSYCHOLOGY / Statistics --- SOCIAL SCIENCE / Statistics --- Computer programs

Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

Social sciences : Statistics. @@@ Social surveys : Statistical methods.

Mathematical statistics --- Statistics --- Social surveys --- Statistical methods --- HA 29 Theory and method of social science statistics - General works --- Social sciences --- Statistics. --- Statistical methods. --- #SBIB:303H520 --- Community surveys --- Surveys, Social --- Surveys --- Behavioral sciences --- Human sciences --- Sciences, Social --- Social science --- Social studies --- Civilization --- Methoden sociale wetenschappen: techniek van de analyse, algemeen --- Research

Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

The primary data driver behind US drug policy is the National Survey on Drug Use and Health. This insider history traces the evolution of the survey and how the survey has interacted with the political and social climate of the country, from its origins during the Vietnam War to its role in the war on drugs. The book includes first-hand accounts that explain how the data was used and misused by political leaders, why changes were made in the survey design, and what challenges researchers faced in communicating statistical principles to policymakers and leaders. It also makes recommendations for managing survey data collection and reporting in the context of political pressures and technological advances. Survey research students and practitioners will learn practical lessons about questionnaire design, mode effects, sampling, nonresponse, weighting, editing, imputation, statistical significance, and confidentiality. The book also includes common-language explanations of key terms and processes to help data users understand the point of view of survey statisticians.

Health surveys --- Drug control --- SOCIAL SCIENCE / Statistics. --- National Survey on Drug Use and Health (U.S.) --- NSDUH (National Survey on Drug Use and Health (U.S.)) --- N.S.D.U.H. (National Survey on Drug Use and Health (U.S.)) --- National Household Survey on Drug Abuse (U.S.)