Evaluating Rater Accuracy in Performance Assessments

A new method for evaluating rater accuracy within the context of performance assessments is described. Accuracy is defined as the match between ratings obtained from operational raters and those obtained from an expert panel on a set of benchmark, exemplar, or anchor performances. An extended Rasch measurement model called the FACETS model is presented for examining rater accuracy. The FACETS model is illustrated with 373 benchmark papers rated by 20 operational raters and an expert panel. The data are from the 1993field test of the High School Graduation Writing Test in Georgia. The data suggest that there are statistically significant differences in rater accuracy; the data also suggest that it is easier to be accurate on some benchmark papers than on others. A small example is presented to illustrate how the accuracy ordering of raters may not be invariant over different subsets of benchmarks used to evaluate accuracy.     

Rater effects: Ego engagement in rater decision-making

The decision-making behaviors of 8 raters when scoring 39 persuasive and 39 narrative essays written by second language learners were examined, first using Rasch analysis and then, through think aloud protocols. Results based on Rasch analysis and think aloud protocols recorded by raters as they were scoring holistically and analytically suggested that rater background may have contributed to rater expectations that might explain individual differences in the application of the performance criteria of the rubrics when rating essays. The results further suggested that rater ego engagement with the text and/or author may have helped mitigate rater severity and that self-monitoring behaviors by raters may have had a similar mitigating effect.     

Pedagogical Considerations for Examining Rater Variability in Rater‐Mediated Assessments: A Three‐Model Framework

Rater‐mediated assessments are a common methodology for measuring persons, investigating rater behavior, and/or defining latent constructs. The purpose of this article is to provide a pedagogical framework for examining rater variability in the context of rater‐mediated assessments using three distinct models. The first model is the observation model, which includes ecological/environmental considerations for the evaluation system. The second model is the measurement model, which includes the transformation of observed, rater response data to linear measures using a measurement model with specific requirements of rater‐invariant measurement in order to examine raters’ construct‐relevant variability stemming from the evaluative system. The third model is the interaction model, which includes an interaction parameter to allow for the investigation into raters’ systematic, construct‐irrelevant variability stemming from the evaluative system. Implications for measurement outcomes and validity are discussed.     

The Measurement of Writing Ability With a Many-Faceted Rasch Model

The purpose of this study is to describe a Many-Faceted Rasch (FACETS) model for the measurement of writing ability. The FACETS model is a multivariate extension of Rasch measurement models that can be used to provide a framework for calibrating both raters and writing tasks within the context of writing assessment. The use of the FACETS model for solving measurement problems encountered in the large-scale assessment of writing ability is presented here. A random sample of 1,000 students from a statewide assessment of writing ability is used to illustrate the FACETS model. The data suggest that there are significant differences in rater severity, even after extensive training. Small, but statistically significant, differences in writing- task difficulty were also found. The FACETS model offers a promising approach for addressing measurement problems encountered in the large- scale assessment of writing ability through written compositions.     

以多面Rasch模型测量写作能力

本研究的目的是描述一个用于测量写作能力的多面Rasch(FACETS)模型。该FACETS模型是Rasch测量模型的多元变量拓展,它可为写作测评中的校标评分员和写作题目提供框架。本文展示了如何应用FACETS模型解决大型写作测评中遇到的测量问题。参加全州写作考试的1000个随机抽取的学生样本被用来显示该FACETS模型。数据表明即使经过强化训练,评分员的严格度有显著区别。同时,本研究还发现,写作题目难度的区分,虽然微小,却具有统计意义上的显著性。该FACETS模型为解决以作文测评写作能力的大型考试遇到的测量问题提供了一个有前景的途径。     

Evaluation of construct-irrelevant variance yielded by machine and human scoring of a science teacher PCK constructed response assessment

Machine learning has been frequently employed to automatically score constructed response assessments. However, there is a lack of evidence of how this predictive scoring approach might be compromised by construct-irrelevant variance (CIV), which is a threat to test validity. In this study, we evaluated machine scores and human scores with regard to potential CIV. We developed two assessment tasks targeting science teacher pedagogical content knowledge (PCK); each task contains three video-based constructed response questions. 187 in-service science teachers watched the videos with each had a given classroom teaching scenario and then responded to the constructed-response items. Three human experts rated the responses and the human-consent scores were used to develop machine learning algorithms to predict ratings of the responses. Including the machine as another independent rater, along with the three human raters, we employed the many-facet Rasch measurement model to examine CIV due to three sources: variability of scenarios, rater severity, and rater sensitivity of the scenarios. Results indicate that variability of scenarios impacts teachers’ performance, but the impact significantly depends on the construct of interest; for each assessment task, the machine is always the most severe rater, compared to the three human raters. However, the machine is less sensitive than the human raters to the task scenarios. This means the machine scoring is more consistent and stable across scenarios within each of the two tasks.     

多面Rasch模型在主观题评分培训中的应用

主观题的评分受到很多因素的影响,如评分者的知识水平、综合能力和个人偏好等。这些评分者偏差不仅会导致不同评分者之间存在主观差异,也会到导致同一评分者在不同的时间也具有主观不稳定性,最终导致主观题评分信度的降低。本研究将多面Rasch模型运用到某国家级考试论述题的评分培训中。通过分析6名有经验评分者对58份试卷的试评数据,鉴别出四种评分者偏差,然后据此对每个评分者进行个别反馈,从而提高评分的客观性和精确性。     

Conceptualizing Rater Judgments and Rating Processes for Rater‐Mediated Assessments

Rater‐mediated assessments exhibit scoring challenges due to the involvement of human raters. The quality of human ratings largely determines the reliability, validity, and fairness of the assessment process. Our research recommends that the evaluation of ratings should be based on two aspects: a theoretical model of human judgment and an appropriate measurement model for evaluating these judgments. In rater‐mediated assessments, the underlying constructs and response processes may require the use of different rater judgment models and the application of different measurement models. We describe the use of Brunswik's lens model as an organizing theme for conceptualizing human judgments in rater‐mediated assessments. The constructs vary depending on which distal variables are identified in the lens models for the underlying rater‐mediated assessment. For example, one lens model can be developed to emphasize the measurement of student proficiency, while another lens model can stress the evaluation of rater accuracy. Next, we describe two measurement models that reflect different response processes (cumulative and unfolding) from raters: Rasch and hyperbolic cosine models. Future directions for the development and evaluation of rater‐mediated assessments are suggested.     

Examining Rater Errors in the Assessment of Written Composition With a Many-Faceted Rasch Model

This study describes several categories of rater errors (rater severity, halo effect, central tendency, and restriction of range). Criteria are presented for evaluating the quality of ratings based on a many-faceted Rasch measurement (FACETS) model for analyzing judgments. A random sample of 264 compositions rated by 15 raters and a validity committee from the 1990 administration of the Eighth Grade Writing Test in Georgia is used to illustrate the model. The data suggest that there are significant differences in rater severity. Evidence of a halo effect is found for two raters who appear to be rating the compositions holistically rather than analytically. Approximately 80% of the ratings are in the two middle categories of the rating scale, indicating that the error of central tendency is present. Restriction of range is evident when the unadjusted raw score distribution is examined, although this rater error is less evident when adjusted estimates of writing competence are used     

Using Multigroup Confirmatory Factor Analysis to Test Measurement Invariance in Raters: A Clinical Skills Examination Application

Psychometric models based on structural equation modeling framework are commonly used in many multiple-choice test settings to assess measurement invariance of test items across examinee subpopulations. The premise of the current article is that they may also be useful in the context of performance assessment tests to test measurement invariance of raters. The modeling approach and how it can be used for performance tests with less than optimal rater designs are illustrated using a data set from a performance test designed to measure medical students’ patient management skills. The results suggest that group-specific rater statistics can help spot differences in rater performance that might be due to rater bias, identify specific weaknesses and strengths of individual raters, and enhance decisions related to future task development, rater training, and test scoring processes.     

Monitoring Rater Performance Over Time: A Framework for Detecting Differential Accuracy and Differential Scale Category Use

In this study, we describe a framework for monitoring rater performance over time. We present several statistical indices to identify raters whose standards drift and explain how to use those indices operationally. To illustrate the use of the framework, we analyzed rating data from the 2002 Advanced Placement English Literature and Composition examination, employing a multifaceted Rasch approach to determine whether raters exhibited evidence of two types of differential rater functioning over time (i.e., changes in levels of accuracy or scale category use). Some raters showed statistically significant changes in their levels of accuracy as the scoring progressed, while other raters displayed evidence of differential scale category use over time.     

Exploring the Effects of Rater Linking Designs and Rater Fit on Achievement Estimates Within the Context of Music Performance Assessments

When good model-data fit is observed, the Many-Facet Rasch (MFR) model acts as a linking and equating model that can be used to estimate student achievement, item difficulties, and rater severity on the same linear continuum. Given sufficient connectivity among the facets, the MFR model provides estimates of student achievement that are equated to control for differences in rater severity. Although several different linking designs are used in practice to establish connectivity, the implications of design differences have not been fully explored. Research is also limited related to the impact of model-data fit on the quality of MFR model-based adjustments for rater severity. This study explores the effects of linking designs and model-data fit for raters on the interpretation of student achievement estimates within the context of performance assessments in music. Results indicate that performances cannot be effectively adjusted for rater effects when inadequate linking or model-data fit is present.     

Investigating rater/prompt interactions in writing assessment: Quantitative and qualitative approaches

This study investigates how experienced and inexperienced raters score essays written by ESL students on two different prompts. The quantitative analysis using multi-faceted Rasch measurement, which provides measurements of rater severity and consistency, showed that the inexperienced raters were more severe than the experienced raters on one prompt but not on the other prompt, and that differences between the two groups of raters were eliminated following rater training. The qualitative analysis, which consisted of analysis of raters' think-aloud protocols while scoring essays, provided insights into reasons for these differences. Differences were related to the ease with which the scoring rubric could be applied to the two prompts and to differences in how the two groups of raters perceived the appropriateness of the prompts.     

Statistically Comparing the Performance of Multiple Automated Raters Across Multiple Items

Automated scoring systems are typically evaluated by comparing the performance of a single automated rater item-by-item to human raters. This presents a challenge when the performance of multiple raters needs to be compared across multiple items. Rankings could depend on specifics of the ranking procedure; observed differences could be due to random sampling of items and/or responses in the validation sets. Any statistical hypothesis test of the differences in rankings needs to be appropriate for use with rater statistics and adjust for multiple comparisons. This study considered different statistical methods to evaluate differences in performance across multiple raters and items. These methods are illustrated leveraging data from the 2012 Automated Scoring Assessment Prize competitions. Using average rankings to test for significant differences in performance between automated and human raters, findings show that most automated raters did not perform statistically significantly different from human-to-human inter-rater agreement for essays but they did perform differently on short-answer items. Differences in average rankings between most automated raters were not statistically significant, even when their observed performance differed substantially.     

Examining rater accuracy and consistency with a special education observation protocol

Research indicates that instructional aspects of teacher performance are the most difficult to reach consensus on, significantly limiting teacher observation as a way to systematically improve instructional practice. Understanding the rationales that raters provide as they evaluate teacher performance with an observation protocol offers one way to better understand the training efforts required to improve rater accuracy. The purpose of this study was to examine the accuracy of raters evaluating special education teachers’ implementation of evidence-based math instruction. A mixed-methods approach was used to investigate: 1) the consistency of the raters’ application of the scoring criteria to evaluate teachers’ lessons, 2) raters’ accuracy on two lessons with those given by expert-raters, and 3) the raters’ understanding and application of the scoring criteria through a think-aloud process. The results show that raters had difficulty understanding some of the high inference items in the rubric and applying them accurately and consistently across the lessons. Implications for rater training are discussed.     

PETS口试评分培训效果的多面Rasch分析

本研究以PETS-1级拟聘口试教师为研究对象,对口试教师评分的培训效果进行了研究。采用多面Rasch分析对比口试教师接受培训前后的评分效果。结果发现：培训后,提升了口试教师与专家评分完全一致的比率,评分偏于严格的口试教师在评分标准上做了恰当的调整,所有口试教师评分拟合值都在可接受范围内,总体上,口试教师评分的培训比较有效,培训后提升了评分的准确性。多面Rasch分析有助于发现评分过于宽松、过于严格、评分拟合差的口试教师以及评分异常情况,为开展有针对性地培训提供了可靠的依据。     

Classical,Generalizability, and Multifaceted Rasch Detection of Interrater Variability in Large,Sparse Data Sets

Classical test theory (CTT), generalizability theory (GT), and multi-faceted Rasch model (MFRM) approaches to detecting and correcting for rater variability were compared. Each of 4,930 students' responses on an English examination was graded on 9 scales by 3 raters drawn from a pool of 70. CTT and MFRM indicated substantial variation among raters; the MFRM analysis identified far more raters as different than the CTT analysis did. In contrast, the GT rater variance component and the Rasch histograms suggested little rater variation. CTT and MFRM correction procedures both produced different scores for more than 50% of the examinees, but 75% of the examinees received identical results after each correction. The demonstrated value of a correction for systems of well-trained multiple graders has implications for all systems in which subjective scoring is used.     

Predicting Operational Rater‐Type Classifications Using Rasch Measurement Theory and Random Forests: A Music Performance Assessment Perspective

The purpose of this study was to build a Random Forest supervised machine learning model in order to predict musical rater‐type classifications based upon a Rasch analysis of raters’ differential severity/leniency related to item use. Raw scores (N = 1,704) from 142 raters across nine high school solo and ensemble festivals (grades 9–12) were collected using a 29‐item Likert‐type rating scale embedded within five domains (tone/intonation, n = 6; balance, n = 5; interpretation, n = 6; rhythm, n = 6; and technical accuracy, n = 6). Data were analyzed using a Many Facets Rasch Partial Credit Model. An a priori k‐means cluster analysis of 29 differential rater functioning indices produced a discrete feature vector that classified raters into one of three distinct rater‐types: (a) syntactical rater‐type, (b) expressive rater‐type, or (c) mental representation rater‐type. Results of the initial Random Forest model resulted in an out‐of‐bag error rate of 5.05%, indicating that approximately 95% of the raters were correctly classified. After tuning a set of three hyperparameters (n_tree, m_try, and node size), the optimized model demonstrated an improved out‐of‐bag error rate of 2.02%. Implications for improvements in assessment, research, and rater training in the field of music education are discussed.