Using Propensity Scores in Quasi Experimental Designs

Author: William M. Holmes
Publisher: SAGE Publications
ISBN: 1483310817
Format: PDF, ePub, Mobi
Download Now
Using Propensity Scores in Quasi-Experimental Designs, by William M. Holmes, examines how propensity scores can be used to reduce bias with different kinds of quasi-experimental designs and to fix or improve broken experiments. Requiring minimal use of matrix and vector algebra, the book covers the causal assumptions of propensity score estimates and their many uses, linking these uses with analysis appropriate for different designs. Thorough coverage of bias assessment, propensity score estimation, and estimate improvement is provided, along with graphical and statistical methods for this process. Applications are included for analysis of variance and covariance, maximum likelihood and logistic regression, two-stage least squares, generalized linear regression, and general estimation equations. The examples use public data sets that have policy and programmatic relevance across a variety of disciplines.

Using Propensity Scores in Quasi Experimental Designs to Equate Groups

Author: Forrest C. Lane
Publisher:
ISBN:
Format: PDF, Mobi
Download Now
Education research rarely lends itself to large scale experimental research and true randomization, leaving the researcher to quasi-experimental designs. The problem with quasi-experimental research is that underlying factors may impact group selection and lead to potentially biased results. One way to minimize the impact of non-randomization is through the use of propensity scores. First developed by Rosenbaum & Rubin (1983b), these scores allow researchers to balance non-equivalent groups though matching on a singular scalar variable. The present paper will present the theoretical framework behind propensity scores along with a heuristic data set to demonstrate propensity score calculation and evaluation. Appended is: "PASW (v17.0) Syntax for Propensity Score Matching using Matching within Calipers." (Contains 4 tables.).

Propensity Score Analysis

Author: Shenyang Guo
Publisher: SAGE
ISBN: 1452235007
Format: PDF, ePub
Download Now
Fully updated to reflect the most recent changes in the field, the Second Edition of Propensity Score Analysis provides an accessible, systematic review of the origins, history, and statistical foundations of propensity score analysis, illustrating how it can be used for solving evaluation and causal-inference problems. With a strong focus on practical applications, the authors explore various strategies for employing PSA, discuss the use of PSA with alternative types of data, and delineate the limitations of PSA under a variety of constraints. Unlike existing textbooks on program evaluation and causal inference, this book delves into statistical concepts, formulas, and models within the context of a robust and engaging focus on application.

A Comparison of Propensity Score Estimation and Adjustment Methods on Simulated Data

Author: Jason K. Luellen
Publisher:
ISBN: 9780549020097
Format: PDF, Docs
Download Now
Faced with potential selection bias resulting from nonequivalent groups, researchers employing quasi-experimental designs have become increasingly interested in statistical adjustments to the estimates of treatment effects based upon the propensity score. Propensity score analysis is the process of trying to balance nonequivalent groups by estimating each participant's conditional probability of treatment assignment using observed covariates and then using these probabilities (i.e., propensity scores) for case matching, stratification, covariance adjustment, or weighting of observations. Numerous propensity score methods have been proposed in the literature. This study used simulated data to examine the relative performance of five methods of estimating propensity scores (logistic regression, classification trees, bootstrap aggregation, boosted regression, and random forests) crossed with four types of adjustments that utilize propensity scores (matching, stratification, covariance adjustment, and weighting) at two levels of sample sizes (N = 200 and N = 1,000). One thousand Monte Carlo replicates were used per level of sample size. All combinations of propensity score methods led to at least some average reduction in selection bias, and for most combinations of methods these reductions were statistically significant. However, this seemingly promising finding is tempered by the fact that bias was actually introduced in many replicates, especially when the level of sample size was 200. The traditional approach to estimating propensity scores, logistic regression, worked well at reducing selection bias, on average, at both sample sizes and tended to result in more precise estimates of the treatment effect with less potential for introducing bias. Other combinations of propensity score methods performed better than logistic regression, on average, but with less precision in the estimates and greater potential for introducing bias. These included random forests at N = 200 and boosted regression and random forests at N = 1,000. Of the ensemble methods, boosted regression, in particular, might be a useful alternative to logistic regression for large sample sizes once the default settings have been changed to favor PSA. With regard to methods of adjusting outcomes using propensity scores, weighting tended to perform poorly. Otherwise, matching, stratification, and covariance adjustment were fairly competitive and a clear favorite was not discerned.

Propensity Score Analysis

Author: Wei Pan
Publisher: Guilford Publications
ISBN: 1462519490
Format: PDF, Mobi
Download Now
This book is designed to help researchers better design and analyze observational data from quasi-experimental studies and improve the validity of research on causal claims. It provides clear guidance on the use of different propensity score analysis (PSA) methods, from the fundamentals to complex, cutting-edge techniques. Experts in the field introduce underlying concepts and current issues and review relevant software programs for PSA. The book addresses the steps in propensity score estimation, including the use of generalized boosted models, how to identify which matching methods work best with specific types of data, and the evaluation of balance results on key background covariates after matching. Also covered are applications of PSA with complex data, working with missing data, controlling for unobserved confounding, and the extension of PSA to prognostic score analysis for causal inference. User-friendly features include statistical program codes and application examples. Data and software code for the examples are available at the companion website (www.guilford.com/pan-materials).

Estimating the Impact of the PROMISE Scholarship Using Propensity Score Weighted Frontier Fuzzy Regression Discontinuity Design

Author: Yetty Shobo
Publisher:
ISBN:
Format: PDF, Docs
Download Now
Regression discontinuity (RD), an "as good as randomized," research design is increasingly prominent in education research in recent years; the design gets eligible quasi-experimental designs as close as possible to experimental designs by using a stated threshold on a continuous baseline variable to assign individuals to a "treatment". Fuzzy RD in which the threshold does not perfectly predict treatment receipt is a subset of this increasingly popular design. In higher education, RD design is increasingly prominent in estimating the impact of merit-based aids on enrollment, school completion, grade point average (GPA), and other key outcomes. Merit-based aids are increasingly allocated based on individuals meeting multiple academic proficiency criteria. A careful examination of merit-based scholarship is particularly critical in West Virginia, a state with the lowest percent of adults 25 and older who have a Bachelor's degree (US Census, 2006). Started in 2002, the West Virginia PROMISE grant aimed to "improve high school and post secondary academic achievement through scholarship incentives" and "promote access to higher education by reducing cost to students." Originally, the grant provided full tuition for attending public colleges (equivalent amount was provided for students in in-state not-for-profit private colleges) but it now provides a maximum of $4750 towards tuition. Academic eligibility to receive and continue receiving the four-year maximum grant is increasingly stringent. The present study utilizes propensity scoring technique in a frontier RD design to estimate the effects of West Virginia's (WV) PROMISE on the quantity (4- and 5-year graduate rates, sum of credits earned) and quality (cumulative GPA) of long-term college indicators. The study examines the effects of receiving West Virginia's PROMISE scholarship for in-state students attending four-year public institutions. The present study uses fuzzy and frontier RD designs because PROMISE receipt is not perfect above the threshold and is based on multiple criteria. Administrative data was obtained from West Virginia Higher Education Policy Commission (WVHEPC). This study demonstrates the significant potential that the frontier RD and the propensity score weighting techniques hold for education research. Tables and figures are appended.

The Oxford Handbook of Quantitative Methods in Psychology

Author: Todd D. Little
Publisher: Oxford University Press
ISBN: 0199934878
Format: PDF
Download Now
The Oxford Handbook of Quantitative Methods in Psychology provides an accessible and comprehensive review of the current state-of-the-science and a one-stop source for learning and reviewing current best-practices in a quantitative methods across the social, behavioral, and educational sciences.

Propensity Score Methods and Applications

Author: Haiyan Bai
Publisher: SAGE Publications
ISBN: 1506378064
Format: PDF, ePub, Mobi
Download Now
A concise, introductory text, Propensity Score Methods and Applications describes propensity score methods (PSM) and how they are used to balance the distributions of observed covariates between treatment conditions as a means to reduce selection bias. This new QASS title specifically focuses on the procedures of implementing PSM for research in social sciences, instead of merely demonstrating the effectiveness of the method. Using succinct and approachable language to introduce the basic concepts of PSM, authors Haiyan Bai and M. H. Clark present basic concepts, assumptions, procedures, available software packages, and step-by-step examples for implementing PSM using real-world data, with exercises at the end of each chapter allowing readers to replicate examples on their own.

Schulentwicklung und Schuleffektivit t

Author: Linda Marie Bischof
Publisher: Springer-Verlag
ISBN: 3658146281
Format: PDF, Docs
Download Now
Linda Marie Bischof arbeitet die Diskussion über die Verknüpfung von Schuleffektivität und Schulentwicklung systematisch auf und analysiert bestehende Verknüpfungsansätze. Sie behandelt Herausforderungen, Probleme und Möglichkeiten theoretischer und empirischer Verknüpfung und thematisiert diese als Ausgangspunkt für weitere Ansätze. Basis der empirischen Verknüpfung ist das PISA-Schulpanel, welches das Design der PISA 2009-Studie um die Schulentwicklungsperspektive ergänzt. Die Autorin untersucht in längsschnittlichen Analysen die Wirkung von Reformmaßnahmen (individuelle Förderung, ganztägige Schulorganisation und Evaluation) auf die Entwicklung von Schulen. Als Schuleffektivitätskriterium zieht sie die Lesekompetenz und als Schulentwicklungskriterium das Schulklima heran.

International Development

Author: Bruce Currie-Alder
Publisher: OUP Oxford
ISBN: 0191651699
Format: PDF, Mobi
Download Now
Thinking on development informs and inspires the actions of people, organizations, and states in their continuous effort to invent a better world. This volume examines the ideas behind development: their origins, how they have changed and spread over time, and how they may evolve over the coming decades. It also examines how the real-life experiences of different countries and organizations have been inspired by, and contributed to, thinking on development. The extent to which development 'works' depends in part on particular local, historical, or institutional contexts. General policy prescriptions fail when the necessary conditions that make them work are either absent, ignored, or poorly understood. There is a need to grasp how people understand their own development experience. If the countries of the world are varied in every way, from their initial conditions to the degree of their openness to outside money and influence, and success is not centred in any one group, it stands to reason that there cannot be a single recipe for development. Each chapter provides an analytical survey of thinking about development that highlights debates and takes into account critical perspectives. It includes contributions from scholars and practitioners from the global North and the global South, spanning at least two generations and multiple disciplines. It will be a key reference on the concepts and theories of development - their origins, evolution, and trajectories - and act as a resource for scholars, graduate students, and practitioners.