Search

In a world facing climate crisis, a growing divide between rich and poor, racial strife, and a rise of xenophobic populism, social studies educators are obligated to investigate social issues in ways that might lead toward more just, less-destructive futures. This paper theorizes a new materialist social studies curriculum--a curriculum attentive to matter, nonhuman species, and anthropocentric representations of nature and the environment--with the aim of uncovering alternative inquiries and responsibilities that might help us flourish as teachers, students, and citizens. Throughout this paper, we argue that a new materialist lens can reconfigure and enliven social studies curriculum, and our corresponding analysis centers on particular events and concepts in the Michigan K-12 Social Studies Standards, considering how prevalent, normative representations reify damaging ways of being and knowing. This paper contributes to ongoing discourses regarding the boundaries, topics, and events that ought to comprise social studies curriculum, and we conclude by speculating new ethical futures for our field.

This direct replication study compared the use of dichotomized likelihood ratios and interval likelihood ratios, derived using a prior sample of students, for predicting math risk in middle school. Data from the prior year state test and the Measures of Academic Progress were analyzed to evaluate differences in the efficiency and diagnostic accuracy of gated screening decisions. Post-test probabilities were interpreted using a threshold decision-making model to classify student risk during screening. Using interval likelihood ratios led to fewer students requiring additional testing after the first gate. But, when interval likelihood ratios were used, three tests were required to classify 6th- and 7th-grade students as at-risk or not at-risk. Only two tests were needed to classify students as at-risk or not at-risk when dichotomized likelihood ratios were used. Acceptable sensitivity and specificity estimates were obtained, regardless of the type of likelihood ratios used to estimate post-test probabilities. When predicting academic risk, interval likelihood ratios may be best reserved for situations where at least three successive tests are available to be used in a gated screening model.

The current study evaluated whether goal-setting practices that account for seasonal developmental patterns of reading growth decreased the number of weeks data needed to be collected in order to yield accurate response to intervention decisions for a sample of 224 third-grade students. The extent to which more complex decision-making practices improve upon the accuracy of current frameworks across periods of 4 to 8 weeks was also explored. Spring proficiency status on a curriculum-based measurement of reading assessment was used as the reference for classification accuracy. Using a goal line based upon fall to winter grade-level benchmarks tended to produce higher levels of accuracy, kappa, and sensitivity than goal lines based upon fall to spring benchmarks. Collectively, the observed results highlight the potential benefit of establishing seasonal goal lines and suggest that relatively simplistic and accessible approaches to decision-making may result in similar decisions when compared with analytically complex approaches.

This study examined spring literacy scores among at-risk prekindergarten students exposed to supplemental support solely in literacy or a combination of literacy and math. Propensity scores were used to match students receiving combined support (n = 39) with an equivalent number of students receiving only literacy tutoring. Students were matched using fall math scores and fall literacy scores. After confirming baseline equivalence, we used a multilevel model to evaluate the association between support type and spring literacy scores, controlling for fall literacy scores, fall numeracy scores, and the total number of intervention sessions completed. In addition to a significant and positive association between fall and spring scores, students who received support in both literacy and math scored significantly better on the spring literacy assessment. More specifically, participating in both literacy and math support was associated with a 23.81 increase in spring literacy scores relative to participating in literacy support alone, explaining 5% of the student-level variance.

School context can shape relative intervention response in myriad ways due to factors, such as instructional quality, resource allocation, peer effects, and correlations between the school context and characteristics of enrolled students (e.g., higher-poverty students attending higher-poverty schools). In the current study, we used data from 16,000 U.S. Grade 3 students in a community-based supplemental reading intervention program to investigate the degree to which school context factors (percentage eligible for free/reduced-price lunch [FRPL], school-level achievement) relate to the differences in triannual reading fluency growth rates between students actively receiving supplemental intervention (active recipients) and those that formerly received intervention (and therefore only received general class instruction at this time; former recipients). Using Bayesian multilevel modeling, our findings indicate that school-level FRPL eligibility played a more prominent factor in growth rate differences between these two groups than school-level reading achievement. However, school-level reading achievement was much more strongly related to reading fluency differences between active and former intervention recipients at the beginning of the school year (when controlling for FRPL). Implications for investigating school-level heterogeneity in intervention response and sustainability are discussed. [ABSTRACT FROM AUTHOR]

This conceptual article uses new materialism, and its particular focus on material things, as a lens of analysis in social studies education in order to demonstrate alternative ways in which social studies education researchers and teachers might engage in inquiry. Historically, social studies curriculum and teaching have centered human agency and its domination of the material, natural world. However, this article argues that an attendance to things, and to the relational-material entanglements we find ourselves in, might guide us toward a reconsideration of how particular ideals and concepts are (and ought to be) represented in social studies curriculum, teaching, and learning. This article models how social studies analyses of material things in past and current events might take shape, exploring how natural things like Hurricane Maria and COVID-19, as well as human-made things like statues, parks, and textbooks, are both agentic and capable of impacting--diminishing or enhancing--the agency of human beings. We discuss how social studies teachers and teacher educators might enact such a focus in their classrooms, offering examples of how natural and human-made things might be integrated into social studies curriculum and teaching.

Educators and researchers measure student improvement in words read correct per minute (WRCM) to evaluate student response to reading interventions. The utility of evaluating improvement in WRCM via curriculum-based measures to make instructional decisions is based upon the assumption that growth in WRCM is predictive of performance on meaningful distal outcomes (e.g., state achievement tests). This study explored the relative value of measuring growth in WRCM to predict performance on an end of year state achievement test after controlling for baseline performance. We used quantile regression to analyze outcomes from 449 grade three students receiving tier II reading interventions to measure the relationship between growth in WRCM and performance on an end of year test across different levels of performance on that test. Results suggest that growth in WRCM is a meaningful predictor of performance on the end of year test after statistically controlling for baseline performance amongst students that perform better on that test (> 50th percentile). However, the amount of explained variance was consistently small (< 5%). Progress monitoring measures that more closely resemble the criterion of interest may be worth exploring for students that have made the transition from fluency to comprehension, while measures targeting lower order skills may be warranted for students with persistent fluency deficits.

A fundamental assumption of multi-tiered systems of support is the bidirectional movement of students between tiers. In comparison to research on when to intensify instruction, less attention has been paid to the validity of decisions to exit students from supplemental supports. We used data from 554 third-grade students who met the criteria to be exited from a Tier 2 reading fluency intervention to answer two research questions. First, we used latent profile analysis to evaluate whether distinct profiles of student responding would emerge based upon initial skill level and intervention duration. Second, we evaluated whether the proportion of students who passed the end-of-year benchmark differed between profiles. Results favored a three-profile solution with a Below Average Start/Average Response; Average Start/Average Response; and an Above Average Start/Fast Response. The pooled mean proportion of students who did not achieve proficiency on the end-of-year benchmark assessment equaled 0.45, 0.33, and 0.13 across the three profiles, respectively. A series of X[superscript 2] tests indicated the between-profile differences were statistically significant (p < 0.05). The results of this study suggest that there may be distinct profiles amongst students who show positive response and exit Tier 2 reading fluency interventions.

Although it is common for researchers to assess implementation fidelity (IF) within the context of math intervention, IF assessments are often restricted to intervention adherence. Further, the degree to which IF influences observed outcomes is commonly ignored. The current study examined the relationship between three aspects of IF and the math performance of 1,340 grade 4 through 8 students who participated in an evidence-based math intervention. A series of multilevel regression models were fit to the data, including a final model with an indicator of intervention adherence, intervention delivery quality, and intervention engagement. A significant and positive association was observed between students' math performance and intervention engagement; however, a similar relationship was not observed when examining the impact of adherence and the quality of intervention delivery on students' math performance. Results are discussed in the context of implementation research for educational interventions.

Once students meet exit criteria and Tier 2 reading supports are removed, many do not maintain their current progress and fail to meet future performance benchmarks. Although estimates of average growth during intervention exist, typical rates of growth after supports are removed are relatively unknown. In this research brief, we used segmented regression to quantify differences between students' reading fluency progress during intervention and postintervention. Participants were second and third grade students, from 16 schools, who were receiving Tier 2 reading interventions targeting oral reading fluency. Once students met exit criteria, their progress was monitored weekly for the rest of the school year. The average rate of improvement was 3.50 words read correct in 1 minute (WRCM) per week (SD = 1.65) during the intervention and 0.09 WRCM (SD = 1.63) after exiting the intervention. Students who met exit criteria during the fall semester (i.e., fast responders) displayed significantly (p < 0.001) higher growth during the intervention (M = 4.32) compared to students who exited during the spring semester (i.e., slow responders; M = 2.69). Moreover, the average rate of postintervention growth was significantly (p < 0.001) higher for fast responders (M = 0.54) compared to slow responders (M = -0.35 WRCM improvement per week).

This study evaluated the impact of a low-intensity strategy designed to facilitate maintenance of reading skill gains made during Tier 2 intervention. A subsample of kindergarten, second grade, and third grade students who exited from a standard protocol Tier 2 intervention (n = 168) continued to receive brief (2-3 min), weekly opportunities to practice intervention targets with feedback on performance. Propensity scores were used to create an appropriate counterfactual (i.e., students who exited the program but did not participate in postexit practice sessions) using nearest neighbor matching (n = 168). Across all grades, students who engaged in weekly practice opportunities with feedback after exiting the intervention had a higher probability of meeting the end-of-year benchmark relative to students who did not participate in the ongoing practice condition. Results suggest that providing brief opportunities for students to practice key reading skills via typical progress monitoring procedures may be a beneficial maintenance strategy for students exiting Tier 2 support.

Common universal screening methods for determining math risk in middle school grades may not result in optimal diagnostic accuracy. We evaluated current screening practices and several potential modifications for predicting math proficiency on an end-of-year state test in a suburban school district. Previously, the district used the Measures of Academic Progress (MAP) to determine student risk in fall. Creating local cut scores for the preceding-year state test scores and the MAP resulted in the most accurate and efficient methods of assessing risk. Multiskill computation and application curriculum-based measures added little relative value to the combination of MAP and preceding-year state test scores. Results suggest schools could improve their screening practices by first evaluating their current procedures and, if unacceptable, determining whether local cut scores provide the desired improvements to accuracy before considering the adoption of additional measures.

A growing body of research suggests that growth in early literacy skills, including letter sound fluency, is predictive of later reading outcomes. In turn, educators and school psychologists use measures of letter sound fluency to monitor student response to early reading instruction. Limited research has evaluated whether decision-making frameworks that educators apply to early reading time series data to make instructional decisions (e.g., to continue or change the intervention) yield accurate recommendations. Further, it is unclear how long data need to be collected and which type of decision rule (e.g., data point, trend line) will produce the most accurate recommendations. We conducted a series of simulations to investigate the impact of data collection duration (4 to 16 weeks) and decision-rule type (data point, trend line, and median) on the accuracy of data-based decisions using early literacy data. Results suggest that the median and trend-line rules produced recommendations that were sufficiently accurate in identifying students that were not making adequate progress (sensitivity) after about 12 weeks when data were collected once a week. However, the median and trend-line rules did not produce recommendations that were sufficiently accurate to continue an intervention (specificity) across all progress monitoring durations. The opposite pattern was observed for the data-point rule. Outcomes suggest that recommendations developed from other progress monitoring measures (e.g., oral reading fluency) should not be extrapolated to other measures without empirical investigation. More research is needed to identify appropriate decision rules to evaluate early literacy progress monitoring data.

Recent research suggests using multiple screening measures to identify students at risk for academic difficulties may decrease the number of students incorrectly identified as such. Gated frameworks in which students that score below a cut-score on an initial measure are assessed with a follow-up measure have been recommended. Researchers have posited that gated screening practices that use measures that explain unique variance in the outcome of interest will yield optimal results. We assessed the degree to which the correlation between screening measures and an outcome, the correlation among screening measures, and cut-score thresholds influenced diagnostic accuracy outcomes. Screening measures that were highly correlated with one another were less effective at reducing false positive classifications than screeners that were less correlated. Further, using screeners that were highly correlated with one another yielded slight but statistically significant reductions in false positive rates compared to using results from a single screening measure.

Data from a large-scale math intervention program were used to evaluate the degree to which implementation fidelity (IF) and interventionist content knowledge were associated with student outcomes. Data were available for 33 interventionists serving 658 students in Grades 4-6 across one school year. A series of multilevel models were fit to the data to evaluate the impact of procedural IF and interventionists' math content knowledge on students' postintervention achievement, controlling for preintervention achievement and intervention dosage. Higher student posttest scores were observed for interventionists with an average fidelity rating of 95% or greater (ß = 0.15); however, no effects on students' math achievement scores were observed for interventionist content knowledge. Adding IF and a measure of interventionist content knowledge to the model explained a statistically significant amount of variance in growth estimates attributable to interventionists (15%). Results highlight the potential importance of ongoing evaluation and remediation of IF in the context of standardized supplemental intervention in math, while also providing some evidence that higher levels of content knowledge may not translate into greater impact in a standard-protocol intervention setting. Results suggest a need for more research examining characteristics of interventionists and aspects of implementation that may account for variance in student outcomes.

Evidence-based interventions exist for improving multiple fundamental math competencies, but delivering interventions with fidelity and within a data-driven, tiered framework is a practical challenge faced by most schools. The current study evaluated a math intervention program delivered with community-based resources via AmeriCorps. At the beginning of the school year, students in Grades 4-8 (n = 550) were randomly assigned to receive math support via the program or to a waitlist control group. Outcomes were measured with a broad-based assessment of math achievement in the winter. Results from intent-to-treat analyses showed a significant and positive effect (d = 0.17) for the program that increased slightly under optimal dosage conditions (d = 0.24). The observed results extend existing literature on math interventions in schools by illustrating the potential for partnerships between community-based organizations and schools to improve outcomes for at-risk students.

Student response to instruction is a key piece of information that school psychologists use to make instructional decisions. Curriculum-based measures (CBMs) are the most commonly used and researched family of academic progress-monitoring assessments. There are a variety of reading CBMs that differ in the type and specificity of skills they assess. The purpose of this study was to determine the degree to which the CBM of oral reading (CBM-R) progress-monitoring data differed from nonsense-word fluency (NWF) progress-monitoring data in the presence of a common intervention. We used multivariate multilevel modeling to compare growth trajectories from CBM-R and NWF progress-monitoring data from a geographically diverse sample of 3,000 1st-grade students receiving Tier-2 phonics interventions. We also evaluated differences in sensitivity to improvement and reliability of improvement from each measure. Improvement on CBM-R was statistically, but not practically, significantly greater than NWF. Although CBM-R was not as direct a measure of decoding, it still captured student response to phonics instruction similarly to NWF. NWF demonstrated slightly better sensitivity to growth, but CBM-R yielded more reliable growth estimates.

School psychologists regularly use decision rules to interpret student response to intervention in reading. Recent research suggests that the accuracy of those decision rules depends on the duration of progress monitoring, the number of observations available, and the amount of measurement error present. In this study, we extended existing research to evaluate the influence of a student's initial level of performance, goal line type, and decision rule type on the accuracy of interpretations of progress monitoring data. Normative goal lines performed best for students scoring far below benchmark at the beginning of the year, while goal lines based upon a spring benchmark score were appropriate for students performing just below expectations at the beginning of the year. The data point rule performed poorly across all progress monitoring conditions, while comparing the median of the 3 most recent observations to a goal line performed similarly to the trend line rule.

Several states have changed their statewide achievement tests over the past 5 years. These changes may pose difficulties for educators tasked with identifying students in need of additional support. This study evaluated the stability of decision-making accuracy estimates across changes to the statewide achievement test. We analyzed extant data from a large suburban district in Wisconsin in 2014-2015 (N = 2,774) and 2015--2016 (N = 2,882). We estimated the decision-making accuracy of recommendations from the Measures of Academic Progress for predicting risk on a Common Core State Standards aligned test (2014-2015) and a new test based on updated academic standards (2015--2016) in reading and math. Findings suggest that sensitivity and specificity estimates were relatively stable in math. Changes in the criterion measure were associated with decreased sensitivity when predicting performance in reading. These results provide initial support for educators to continue existing screening practices until test vendors or state educational agencies establish cut-scores for predicting risk on the newer test. Using a lower cut-score to establish risk (increasing sensitivity while decreasing specificity) may be prudent in reading. Limitations and directions for future research are discussed.

The accuracy of decision rules for progress monitoring data is influenced by multiple factors. This study examined the accuracy of decision rule recommendations with over 4,500 second-and third-grade students receiving a tier II reading intervention program. The sensitivity and specificity of three decision rule recommendations for predicting year-end spring benchmark targets was evaluated over different data collection durations under good and poor dataset conditions. Across grade level and dataset quality, the sensitivity of decisions made using trend lines and the median of recent data points tended to improve to acceptable levels after 18 weeks. Specificity tended to decrease around that same time but was less pronounced with the median decision rule method. Limitations and directions for future research are discussed.

In the present study, we evaluated the number of attempts required to master specific subskills for working with whole and rational numbers among students at risk for math difficulties. Participants included a subset of students in grades four through eight receiving supplemental math support. Mastery--defined as 85% correct on short tests--was assessed following instruction for each subskill. Using survival analysis, we evaluated the number of attempts required to reach a 0.50 and a 0.90 probability of mastery on each subskill. The number of required attempts varied across subskills, with many subskills requiring more than one attempt to demonstrate mastery. Further, some of the most difficult content was aligned with curricular standards below students' grade level. Thus, among students identified for supplemental support, it may be worthwhile to remediate select subskills that fall outside of the grade-level curriculum before providing additional instruction on grade-level content. Implications for math subskill assessment and remediation are discussed along with limitations and directions for future research.

Relatively little is known regarding post-intervention reading fluency outcomes for English learners (ELs) in comparison with non-EL peers, yet educators must be prepared to sustain growth for all students transitioning to less-intensive tiers of support. The purpose of this study was to investigate whether EL status moderated post-intervention maintenance effects among U.S. second- and third-grade students who transitioned back to Tier 1 instruction only due to successful performance during a Tier 2 reading fluency intervention. Piecewise multi-level models were estimated to address whether EL status uniquely predicted intervention growth patterns and the extent to which these patterns were maintained over a 12- to 13-week post-intervention period. Reading fluency scores were similar between EL and non-EL students prior to the start of and during the intervention, and all students' performance decreased slightly immediately after support ceased. Regardless of grade level or EL status, post-intervention fluency gains generally remained smaller than those observed during intervention meriting attention to individual- and systems-level instructional considerations for ensuring continued growth. [ABSTRACT FROM AUTHOR]

Single-case designs provide an established technology for evaluating the effects of academic interventions. Researchers interested in studying the long-term effects of reading interventions often use curriculum-based measures of reading (CBM-R) as they possess many of the desirable characteristics for use in a time-series design. The reliability of CBM-R scores is often supported by research from group designs, but making idiographic interpretations regarding the change in a student's oral reading rate requires attention to the precision of static scores and growth estimates. The purpose of this paper is twofold. First, we discuss how recent empirical work on the technical adequacy of CBM-R scores has revealed multiple threats to the data-evaluation validity when CBM-R passages are used to measure oral reading rate. Second, we identify pertinent considerations for conducting a visual analysis of intervention effects based on CBM-R data. We conclude with a brief discussion of implications for researchers considering the use of CBM-R within multiple-baseline designs.

The present study examined the relationship between student and observer ratings of the class environment. More specifically, class responses on the Responsive Environmental Assessment for Classroom Teaching (REACT; Theodore J. Christ & Colleagues, 2015) were compared with observer ratings on the Classroom Strategies Assessment System-Observer Form (Reddy, Fabiano, & Dudek, 2013). This study included 38 teachers and 582 students from 5 high-poverty schools. Observational data were reported as discrepancy scores, which reflect the difference between the recommended frequency and observed frequency of specific instructional and behavioral management strategies for classroom teachers. Pearson correlations were used to evaluate the relationship between the 6 subscales included on the REACT and the 9 subscales included on the CSAS-O. Results provide preliminary evidence for the relationship between observer and student ratings of the class environment. More specifically, as discrepancy scores decreased, student ratings of the class environment tended to be more positive. The relationship between the REACT and the CSAS-O differed across subscales; however, in general, subscales that were conceptually similar tended to demonstrate stronger relationships than subscales that were conceptually distinct. Thus, the observed results also provide preliminary evidence that students are capable of discriminating between the quality of different components of the class environment. The potential use of both observer and student ratings of the class environment to provide teachers with a more robust and comprehensive reference for professional development purposes is discussed within the context of a tiered model of support.

We used existing reading (n = 1,498) and math (n = 2,260) data to evaluate state test scores for screening middle school students. In Phase 1, state test data were used to create a research-derived cut score that was optimal for predicting state test performance the following year. In Phase 2, those cut scores were applied with future cohorts. Diagnostic accuracy of the research-derived cut scores was compared with the state's proficiency benchmark from the previous year. Across grades and content areas, research-derived cut scores yielded higher sensitivity and lower specificity values relative to state-defined cut scores. Marked decreases in sensitivity and specificity were not observed in subsequent years. Results provide evidence for procedures in which previous state test data are repurposed for screening decisions.

Computer adaptive tests (CATs) hold promise to monitor student progress within multitiered systems of support. However, the relationship between how long and how often data are collected and the technical adequacy of growth estimates from CATs has not been explored. Given CAT administration times, it is important to identify optimal data collection schedules to minimize missed instructional time. We used simulation methodology to investigate how the duration and frequency of data collection influenced the reliability, validity, and precision of growth estimates from a math CAT. A progress monitoring dataset of 746 Grade 4, 664 Grade 5, and 400 Grade 6 students from 40 schools in the upper Midwest was used to generate model parameters. Across grades, 53% of students were female and 53% were White. Grade level was not as influential as the duration and frequency of data collection on the technical adequacy of growth estimates. Low-stakes decisions were possible after 14-18 weeks when data were collected weekly (420-540 min of assessment), 20-24 weeks when collected every other week (300-360 min of assessment), and 20-28 weeks (150-210 min of assessment) when data were collected once a month, depending on student grade level. The validity and precision of growth estimates improved when the duration and frequency of progress monitoring increased. Given the amount of time required to obtain technically adequate growth estimates in the present study, results highlight the importance of weighing the potential costs of missed instructional time relative to other types of assessments, such as curriculum-based measures. Implications for practice, research, as well as future directions are also discussed.

Educators need recommendations to improve screening practices without limiting students' instructional opportunities. Repurposing previous years' state test scores has shown promise in identifying at-risk students within multitiered systems of support. However, researchers have not directly compared the diagnostic accuracy of previous years' state test scores with data collected during fall screening periods to identify at-risk students. In addition, the benefit of using previous state test scores in conjunction with data from a separate measure to identify at-risk students has not been explored. The diagnostic accuracy of 3 types of screening approaches were tested to predict proficiency on end-of-year high-stakes assessments: state test data obtained during the previous year, data from a different measure administered in the fall, and both measures combined (i.e., a gated model). Extant reading and math data (N = 2,996) from 10 schools in the Midwest were analyzed. When used alone, both measures yielded similar sensitivity and specificity values. The gated model yielded superior specificity values compared with using either measure alone, at the expense of sensitivity. Implications, limitations, and ideas for future research are discussed.

The current study evaluated the degree to which novice visual analysts could discern trends in simulated time-series data across differing levels of variability and extreme values. Forty-five novice visual analysts were trained in general principles of visual analysis. One group received brief training on how to identify and omit extreme values. Participants rated 72 continuous time-series graphs. Inferential analyses were used to estimate the probability of correct responses. Participants who received the additional training were more likely to correctly identify intervention effects across all conditions. Nevertheless, extreme values had a substantial impact on decision accuracy for all participants. The impact of extreme values was exacerbated by increases in overall variability. Results support the notion that automated trend lines are useful but not infallible when interpreting continuous time-series data. Implications for practice and avenues for future research are discussed.

Although extensive research exists on the use of curriculum-based measures for progress monitoring, little is known about using computer adaptive tests (CATs) for progress-monitoring purposes. The purpose of this study was to evaluate the impact of the frequency of data collection on individual and group growth estimates using a CAT. Data were available for 278 fourth- and fifth-grade students. Growth estimates were obtained when five, three, and two data collections were available across 18 weeks. Data were analyzed by grade to evaluate any observed differences in growth. Further, root mean square error values were obtained to evaluate differences in individual student growth estimates across data collection schedules. Group-level estimates of growth did not differ across data collection schedules; however, growth estimates for individual students varied across the different schedules of data collection. Implications for using CATs to monitor student progress at the individual or group level are discussed.

We examined the diagnostic accuracy and efficiency of three approaches to universal screening for reading difficulties using retrospective data from 1,307 students in Grades 3 through 5. School staff collected screening data using the Measures of Academic Progress (MAP), a curriculum-based measure (CBM), and running records (RR). The criterion measure was a high-stakes state accountability test aligned with the Common Core State Standards. We examined the diagnostic accuracy of the tests in isolation, as multivariate batteries, and via a simulated gated-screening approach. CBM and RR data resulted in unacceptable diagnostic accuracy across all three grades. In the fourth grade, the MAP alone resulted in the best balance of sensitivity and specificity. Among third- and fifth-grade students, the multivariate combination of MAP and CBM demonstrated the best balance between diagnostic accuracy and efficiency. Gated-screening increased specificity but lowered sensitivity. Results highlight the need for population-specific considerations in universal screening.

The current study estimated the reliability and agreement of student ratings of the classroom environment obtained using the Responsive Environmental Assessment for Classroom Teaching (REACT; Christ, Nelson, & Demers, 2012; Nelson, Demers, & Christ, 2014). Coefficient alpha, class-level reliability, and class agreement indices were evaluated as each index provides important information for different interpretations and uses of student rating scale data. Data for 84 classes across 29 teachers in a suburban middle school were sampled to derive reliability and agreement indices for the REACT subscales across 4 class sizes: 25, 20, 15, and 10. All participating teachers were White and a larger number of 6th-grade classes were included (42%) relative to 7th- (33%) or 8th- (23%) grade classes. Teachers were responsible for a variety of content areas, including language arts (26%), science (26%), math (20%), social studies (19%), communications (6%), and Spanish (3%). Coefficient alpha estimates were generally high across all subscales and class sizes (a = 0.70-0.95); class-mean estimates were greatly impacted by the number of students sampled from each class, with class-level reliability values generally falling below .70 when class size was reduced from 25 to 20. Further, within-class student agreement varied widely across the REACT subscales (mean agreement = 0.41-0.80). Although coefficient alpha and test-retest reliability are commonly reported in research with student rating scales, class-level reliability and agreement are not. The observed differences across coefficient alpha, class-level reliability, and agreement indices provide evidence for evaluating students' ratings of the class environment according to their intended use (e.g., differentiating between classes, class-level instructional decisions).

The current study explored multiple ways in which middle schools can use and integrate data sources to predict proficiency on future high-stakes state achievement tests. The diagnostic accuracy of (a) prior achievement data, (b) teacher rating scale scores, (c) a composite score combining state test scores and rating scale responses, and (d) two gated screening approaches was compared in a sample of 614 middle school students. Prior state test performance emerged as the strongest single predictor of future state test scores; however, results provide evidence that educators may consider locally derived cut scores or alternative screening procedures that incorporate multiple data sources. Specifically, the combination of prior achievement data and teacher ratings of student competence often resulted in increases in either sensitivity or specificity as a function of how data sources were combined.

The purpose of the current study was to evaluate the importance of growth in math fact skills within the context of overall math proficiency. Data for 1,493 elementary and middle school students were included for analysis. Regression models were fit to examine the relative value of math fact fluency growth, prior state test performance, and a fall administration of a computer adaptive test for predicting performance on a year-end state math test. Across all grades, the largest effect sizes were observed for student performance on the fall computer adaptive assessment (ß = 0.43-0.56). Fact fluency growth accounted for unique variance in each regression model, with the largest effect size observed for students in Grade 4 (ß = 0.20) and the smallest effect sizes observed for students in Grades 7 and 8 (ß = 0.11). Results provide insight into the potential role of math fact fluency across late elementary and middle school. In particular, the observed results indicate that math facts may retain predictive value for math proficiency despite their absence from the formal curriculum in later grades.

The present study used data for 30 classes across 10 middle and high school teachers to evaluate the stability of class-level ratings on the Responsive Environmental Assessment for Classroom Teaching across time. Teachers collected data on 2 occasions and students' ratings (N = 806) were aggregated to the class-level. Classes were arranged into 2 groups on the basis of the latency between repeated administrations (M[subscript groups] = 8.1 and 15.1 weeks). Correlations between class ratings were slightly larger among classes in the first group (range = 0.73 to 0.87) relative to the second group (range = 0.62 to 0.75) which had a larger gap in time between data collections. Results provide evidence for the stability of students' perceptions of the instructional environment when those ratings are interpreted at the level of a class and support ongoing work to establish norms and benchmarks to guide interpretation.

The impact and feasibility of using student perceptions of the classroom teaching environment as an instructional feedback tool were explored. Thirty-one teachers serving 797 middle school students collected data twice across 3 weeks using the Responsive Environmental Assessment for Classroom Teaching (REACT). Researchers randomly assigned half of the teachers to receive student feedback following the first data collection. Student responses in the classrooms of teachers who received feedback were more positive in the second round of data collection compared with the teachers who did not receive feedback. Students' initial REACT scores, gender, and self-reported trouble in class were also significant predictors of REACT scores at the second data collection. Finally, teachers reported the REACT to be feasible for use in practice.

Computational fluency is an important aspect of math proficiency. Despite widely held beliefs about the differential difficulty of single-digit multiplication math facts, little empirical work has examined this issue. The current study analyzed the number of repetitions needed to master multiplication math facts. Data from 15,402 3rd, 4th, and 5th graders were analyzed using a national database. Results suggested that (a) students with lower math skills required significantly (p < 0.001) more repetitions than more skilled students; (b) across all students, single-digit multiplication facts with 4s, 5s, 6s, and 7s required significantly (p < 0.001) more repetition than did 2s and 3s; and (c) the number of practice sessions needed to attain mastery significantly (p < 0.001) decreased with increase in grade level. Implications for instructional planning and implementation are discussed.

Direct Behavior Rating (DBR) is a repeatable and efficient method of behavior assessment that is used to document teacher perceptions of student behavior in the classroom. Time-series data can be graphically plotted and visually analyzed to evaluate patterns of behavior or intervention effects. This study evaluated the decision accuracy of novice raters who were presented with single-phase graphical plots of DBR data. Three behaviors (i.e., academically engaged, disruptive, and respectful) and three graphical trends (i.e., positive, no trend, and negative) were analyzed by 27 graduate and five undergraduate participants who had minimal visual analysis experience. All graphs were unique, with data points arranged to form one of three "true" trends. Raters correctly classified graphs with positive, no, and negative trends an average of 76, 98, and 67% of instances. The generalized linear mixed model was used to handle significance tests for the categorical data. Results indicate that accuracy was influenced by the trend direction, with the most accurate ratings in the no trend condition. Despite the significant effect for trend direction, the current study provides empirical evidence for accuracy of DBR trends and interpretations. Novice raters and visual analysts yielded accurate decisions regarding the trend of plotted data for student behavior.

This study details the initial development of the Responsive Environmental Assessment for Classroom Teachers (REACT). REACT was developed as a questionnaire to evaluate student perceptions of the classroom teaching environment. Researchers engaged in an iterative process to develop, field test, and analyze student responses on 100 rating-scale items. Participants included 1,465 middle school students across 48 classrooms in the Midwest. Item analysis, including exploratory and confirmatory factor analysis, was used to refine a 27-item scale with a second-order factor structure. Results support the interpretation of a single general dimension of the Classroom Teaching Environment with 6 subscale dimensions: Positive Reinforcement, Instructional Presentation, Goal Setting, Differentiated Instruction, Formative Feedback, and Instructional Enjoyment. Applications of REACT in research and practice are discussed along with implications for future research and the development of classroom environment measures.

The authors' purpose was to determine the effects of computer-based practice and conceptual interventions on computational fluency and word-problem solving of fourth- and fifth-grade students with mathematics difficulties. A randomized pretest-posttest control group design found that students assigned to the computer-based practice intervention group outperformed students in the comparison group on the retention measure. Students assigned to the conceptual intervention did not outperform the comparison group on any of the outcome variables. Implications for instruction and interventions are discussed.