Adolescent substance misuse is a persistent concern in the USA and student substance possession on campus presents as an increasing challenge for administrators. This study seeks to understand how students are differentially disciplined for substance possession based on substance type and student race/ethnicity in US schools. Based a national sample of 99,181 students (Grades 6 through 12) from 3285 schools across school years 2012–13 to 2018–19, we used descriptive analysis to explore how students possessing alcohol, tobacco, or drugs on campus on their first referral were disciplined. Using multi-level logistic modeling, we assessed to what extent use of exclusionary discipline (i.e., out-of-school suspension [OSS], in-school suspension [ISS]) varied by year, substance type, and student race/ethnicity (i.e., White/Caucasian, African American/Black, Hispanic/Latino/a/e). Descriptive results indicate that management of substance possession in schools relies heavily on exclusion over other actions. Results from multi-level models for OSS and ISS demonstrated shifts in use of each action over time, significantly different rates by substance type possessed, and a significant student race/ethnicity by substance interaction. Findings suggest that alternative approaches to exclusion for substance possession have not been widely implemented; however, the use of exclusion may be shifting from OSS to the use of ISS for these students. Suggestions for coherent, preventive programming for substance misuse and for planning alternative, supportive responses are reviewed.

Schools and school districts use social media for a variety of reasons, but alongside the benefits of schools' social media use come potential risks to students' privacy. Using a novel dataset of around 18 million Facebook posts by schools and districts in the United States, we explore the extent to which personally identifiable information of students may be revealed. We find that around 4.9 million posts depict one or more students' faces, and approximately 726,000 posts also identify the full name of one or more students. We also examine which Facebook page characteristics and structural factors might be associated with posts that depict or identify students. We find that districts with a higher student poverty rate and districts that share a greater number of posts were more likely to depict students. We discuss these findings and recommendations for educational leaders and researchers through the lens of data ethics.

BACKGROUND: In Oregon in 2019, only 261 students were eligible for special education under the traumatic brain injury (TBI) category. Many students with TBIs are not treated by a medical provider, so the requirement for a medical statement could prevent eligible youth from receiving special education services. OBJECTIVE: This study investigated barriers to using a medical statement to establish special education eligibility for TBI, support for using a guided credible history interview (GCHI), and training needs around GCHI. RESULTS: Among participants, 84% reported difficulty obtaining a medical statement for TBI eligibility determination, and 87% favored the GCHI as an alternative, though they reported a need for training in TBI and GCHI. CONCLUSION: The results support the use of GCHI to establish special education eligibility for TBI and informed Oregon’s addition of GCHI to TBI special education eligibility determination.

Public schools and districts use social media to share announcements and communicate with parents and the community, but alongside such uses run risks to students’ privacy. Using a novel data set of 18 million posts on Facebook by schools and school districts in the United States, we sought to establish how frequently photos of students were shared. Through sequential mixed-methods, we estimated that around 4.9 million posts included identifiable images of students and that approximately 726,000 of these posts also included students’ first and last names and their approximate location. We discuss these findings’ implications from a data ethics perspective.

AbstractSome children are more affected by specific family environments than others, as a function of differences in their genetic make-up. However, longitudinal studies of genetic moderation of parenting effects during early childhood have not been conducted. We examined developmental profiles of child behavior problems between 18 months and age 8 in a longitudinal parent–offspring sample of 361 adopted children. In toddlerhood (18 months), observed structured parenting indexed parental guidance in service of task goals. Biological parent psychopathology served as an index of genetic influences on children’s behavior problems. Four profiles of child behavior problems were identified: low stable (11%), average stable (50%), higher stable (29%), and high increasing (11%). A multinominal logistic regression analysis indicated a genetically moderated effect of structured parenting, such that for children whose biological mother had higher psychopathology, the odds of the child being in the low stable group increased as structured parenting increased. Conversely, for children whose biological mother had lower psychopathology, the odds of being in the low stable group was reduced when structured parenting increased. Results suggest that increasing structured parenting is an effective strategy for children at higher genetic risk for psychopathology, but may be detrimental for those at lower genetic risk.

Schools and districts use social media to share announcements and build vibrant communities, but alongside such uses run risks to students’ privacy and safety. Using a novel data set of more than 17 million posts from 2010-2020 of schools and school districts in the United States on Facebook, we sought to establish how frequently photos of students were shared. Using qualitative and quantitative methods, we estimate that around 4.9 million posts included identifiable images of students, and that around 725 thousand posts included students’ names. We discuss the implications of these findings with a data ethics lens and describe future steps that research and policy may take to make use of our methodologies and findings.

Schools and school districts now use social media for a variety of reasons, but alongside the benefits that accompany these K-12 educational institutions' social media use come questions about privacy and safety. We use a dataset of around 18 million posts by United States schools and districts on Facebook to explore how these posts might risk the privacy and safety of students by depicting (showing a photo of a student's face) or identifying (by depicting a student and naming them with their first and last name) them in public posts. We also examine which Facebook page characteristics and structural factors (i.e., the size of the district) might be associated with posts that depict or identify students. Through our sequential mixed methods design involving qualitative coding of a sample of 400 posts, we found that around 4.9 million posts depict students and that around 725,000 posts identify students. Districts with a higher student poverty rate and districts that post more were more likely to depict students in any one post. We discuss these findings and suggest directions for future research in terms of data ethics.

System-wide educational reforms are difficult to implement in the United States, but despite the difficulties, reforms can be successful, particularly when they are associated with broad public support. This study reports on the nature of the public sentiment expressed about a nationwide science education reform effort, the Next Generation Science Standards (NGSS). Through the use of data science techniques to measure the sentiment of posts on Twitter about the NGSS (N = 565,283), we found that public sentiment about the NGSS is positive, with only 11 negative posts for every 100 positive posts. In contrast to findings from past research and public opinion polling on the Common Core State Standards, sentiment about the NGSS has become more positive over time—and was especially positive for teachers. We discuss what this positive sentiment may indicate about the success of the NGSS in light of opposition to the Common Core State Standards.

Traumatic brain injury (TBI) affects children's ability to succeed at school. Few educators have the necessary training and knowledge needed to adequately monitor and treat students with a TBI, despite schools regularly serving as the long-term service provider. In this article, we describe a return to school model used in Oregon that implements best practices indicated by the extant literature, as well as our research protocol for evaluating this model. We discuss project aims and our planned procedures, including the measures used, our quasi-experimental design using matched controls, statistical power, and impact analyses. This project will provide the evidential base for implementation of a return to school model at scale.

Nobody's Home Daniel Anderson (bio) All winter, these tangled caneswere silver-thorned and vicious to the root.But now, bearing black fruit,they deepen, blush, and thrivein high, unruly lanes,annihilating where they growthe fence posts and corroded chicken wireof some forgotten boundary line.The Cascade air is dry and tossed with pine,but in the pure sunthese thickets smell like wine,a heady, tannin atmosphere.This late, the river mostly runson ghosts of melted mountain snows.The water winks and glisters as it goes.I can understandwhy one would build his cabin here,though now the mossy roof is caving in.Two socket-hollow windows stare.The plywood walls are warped and mushroom-gray.Above what used to be a door,"Nobody's Home," soft, sunken letters say. [End Page 239] This would have been the country once,before the apple orchardsand all the groves of hazelnuts were bought.Nobody's Home. Back then it must have beena fern-blessed, unimpeachable estate.A place where Nobody could be aloneand think the things Nobody thought.I came here almost every daythat last unhappy year.Headful of hurt. Heartful of blame.I came to argue and accuse.Or argue and explain.I argued with myself and always lost.I argued then apologizedto everyone. My wife. The priestwho married us. My in-laws, whom I loved.Our friends. Even the dog,our beautiful and now-dead dogwho, as I argued, scrambled after geese,the grumpy great blue herons, and the squirrels,then scrambled back to me. We sat in silence every night.We mixed our drinks. We watched inane TV.Then one of us, without a word,would rise and leave the other there.We barely spoke. And yetI couldn't stop the yelling in my head.Anger, I learned, was easier than dread.And dread, as it turned out,was better than despair. [End Page 240] I'd never fallen so far out of love.But by the end, we didn't hardly fight.We moved about like strangers in that house—a little shy, not quite at home,and ruthlessly polite. [End Page 241] Daniel Anderson Daniel Anderson has published three books of poems: The Night Guard at the Wilberforce Hotel, Drunk in Sunlight, and January Rain. He teaches on the faculty of the MFA Program at the University of Oregon. Copyright © 2021 The University of the South

For many schools and districts in the United States, Facebook has emerged as an important tool for sharing timely information, building a sense of community, highlighting staff and students, and many other purposes. However, neither researchers nor schools and districts have paid enough attention to how their Facebook use may pose a risk to the privacy of individuals — often students who are minors. Joshua Rosenberg, Macy Burchfield, Conrad Borchers, Benjamin Gibbons, Daniel Anderson, and Christian Fischer describe their recent studies showing that 15-20 million photos of students have been shared on publicly accessible Facebook pages of public schools and districts. They estimate that at least 150,000 of these photos — and perhaps as many as a million or more — depict students who are identifiable by name and school or district. They review some of the risks to students that might result from such social media posts and offer practical steps that schools and districts can take to minimize these risks.

For many schools and districts in the United States, Facebook has emerged as an important communication tool. Facebook has been used for many purposes including sharing timely information, building a sense of community, and highlighting staff and students. However, neither researchers nor, we think, most schools and districts have paid enough attention to how their Facebook use may pose a risk to the privacy of the individuals—often students who are minors. One of our recent social media studies showed that between 15 and 20 million photos of students have been shared on strictly publicly accessible Facebook pages of public schools and districts. We estimated that at least 150,000 of these photos—and perhaps as many as a million or more—depict students who are identifiable by name and school and/or district. In this article, we review some of the risks to students that might result from their identification through social media posts and offer practical steps that schools and districts can take to minimize these risks.

There is no doubt that public education has suffered as a result of the COVID-19 pandemic. Researchers are comparing the COVID slide to summer learning loss, noting this loss could be much worse for those already underserved by U.S. schools (Kuhfeld & Tarasawa, 2020). In order to understand who this loss impacts and how, we need data on school, district, and statewide responses to the pandemic as it unfolded. Luckily, these data are available, as school districts across the country updated their communities about plans for spring 2020. To capture these updates, our team uses multiple approaches for collecting COVID-19-related information via school district websites and social media to create a new, nationwide dataset of district responses.We also analyze how these responses relate to contextual characteristics of districts and their surrounding communities, which will provide a picture of how district characteristics may drive disparities in access to and quality of schooling during the pandemic. Identifying these associations are critical for understanding and disrupting the reproduction and deepening of educational inequality caused by the COVID-19 crisis. The resulting dataset will provide researchers with the information necessary to understand how education during the pandemic may impact students for years to come.

We used data from the 2014–2015 easyCBM assessment system to explore the applied reading intervention characteristics in a sample of 3,074 Grade 1 students (and 5,145 interventions) in school districts applying a multitiered systems of support (MTSS) framework. We describe the number of interventions, number of assessments, the intervention start dates, curricula, instructional strategies, tier, group size, frequency, dosage, total time, and quantitative intensity. We found variance across all instructional variables, with 156 curricula and 59 instructional strategies applied. Based on our data, a “typical” intervention was a Tier 2 intervention that began before October, was delivered for 30 minutes/day for 5 days/week in a group with three to five students, was changed once if at all, and student progress was most likely monitored with word reading fluency measures.

Achievement gaps are well documented by income and race/ethnicity. Comparatively little research, however, has investigated between-school differences in these gaps. Using publicly available data from California, Oregon, and Washington from the 2014-15 to 2017-18 school years, we estimate school-level racial and economic achievement gaps by grade and content area. We find the majority of the variance in these estimates lies between schools (within district). We then investigate the extent to which these estimates depend upon geography, with visual displays (maps) providing clear evidence of spatial clustering. Finally, we build a computational model using the physical location of the school (longitude and latitude) to predict the achievement gap. We find approximately 18-61% of the total variability accounted for by location, depending on the specific model. We further use the residualized estimates to identify schools with unexpected achievement gaps. Implications for future research and educational systems-level reform are discussed.

The Center for Open Science (COS) will create an ECR Data Resource Hub to facilitate rigorous and reproducible research practices such as data sharing and study registration. The Hub will integrate training materials, infrastructure, community engagement, and innovation in research to advance rigorous research skills and behavior across the STEM education research community. The Hub will foster innovation in open and reproducible research practices for the breadth of research activities in education including experimental, observational, longitudinal, and qualitative methods. Finally, the Hub will connect the STEM education research community with neighboring communities to leverage shared insights and knowledge building.

New data sources and analytic techniques have enabled educational researchers to ask new questions and work to address enduring problems. Yet, there are challenges to those learning and applying these methods. In this chapter, we provide an overview of a nascent area of both scholarship and teaching, educational data science. We define educational data science as the combination of capabilities related to quantitative methods in educational research, computer science and programming capabilities, and teaching, learning, and educational systems. We demonstrate that there are two distinct—but complementary—perspectives on educational data science in terms of being both in education (as a research methodology) and for education (as a teaching and learning content). We describe both of these areas in light of foundational and recent research. Lastly, we highlight three future directions for educational data science, emphasizing the synergies between these two perspectives concerning designing tools that can be used by both learners and professionals, foregrounding representation, inclusivity, and access as first-order concerns for those involved in the growing community, and using data science methodologies to study teaching and learning about data science. We highlight the potential for the growth of educational data science within learning design and technology as situated with the broader data science domain and in education more broadly.