Tuesday, November 21, 2017

The Collaborative Problem Solving Skills of U.S. Students and their International Peers

U.S. students outperformed their peers around the world, on average, on an international assessment of collaborative problem solving skills, with 14 percent of students scoring at the highest level of proficiency.

The Program for International Student Assessment (PISA) collaborative problem solving assessment was given in 2015 and measured students’ ability to solve a problem by sharing the understanding and effort required to come to a solution, and pooling their knowledge, skills, and effort to reach that solution.

PISA is coordinated by the Organization for Economic Cooperation and Development (OECD) and conducted in the U.S. by the National Center for Education Statistics (NCES). The U.S. was one of 51 education systems participating in the PISA collaborative problem solving assessment, along with two U.S. states—Massachusetts and North Carolina.

NCES released a Data Point report today (Nov. 21) on the results of the assessment, which found:

• The U.S. average score of 520 was higher than the international/OECD average score of 500;

• The U.S average score was statistically higher than 35 other education systems and statistically lower than 10 education systems. The U.S. score was about the same as five other systems. Singapore (561) and Japan (552) had the highest average scores, while Tunisia (382) had the lowest score; and

• About 14 percent of U.S. students scored in the highest range of proficiency (Level 4) and 30 percent scored in the second-highest range (Level 3), compared to 8 and 28 percent, respectively, for international/OECD students. About 24 percent of U.S. students scored in the lowest range of proficiency (Level 1 and below), compared to 28 percent internationally.

Prentice Hall/Pearson Literature© : no discernible effects on general literacy achievement and comprehension for adolescent readers

The What Works Clearinghouse (WWC) recently reviewed the research on Prentice Hall Literature© (1989–2005) and Prentice Hall/Pearson Literature© (2007–2015) to determine their impacts on the literacy achievement of students in grades 6–12. The results are summarized in two intervention reports released today, November 21, by the Institute of Education Sciences.

Prentice Hall Literature© (1989–2005) and Prentice Hall/Pearson Literature© (2007–2015) are English language arts curricula designed for students in grades 6–12 that focus on building reading, vocabulary, literary analysis, and writing skills. These curricula are available in multiple editions.

These intervention reports refer to earlier editions of the curricula (Prentice Hall Literature© and Prentice Hall Literature: Timeless Voices, Timeless Themes©) as Prentice Hall Literature© (1989–2005) and later editions (Prentice Hall Literature: Penguin Edition©, Prentice Hall Literature: Language and Literacy©, Prentice Hall Literature: Common Core Edition©, and Pearson Literature©) as Prentice Hall/Pearson Literature© (2007–2015).

After reviewing the research on Prentice Hall Literature© (1989–2005), the WWC found no studies that meet WWC standards. Therefore, more research is needed to determine the effectiveness of the curricula for adolescent readers.
Complete report

After reviewing the research on Prentice Hall/Pearson Literature© (2007–2015), the WWC found no discernible effects on general literacy achievement and comprehension for adolescent readers. There were no studies of Prentice Hall/Pearson Literature© (2007–2015) that meet WWC group design standards in the areas of alphabetics or reading fluency.

Report provides little useful information about school choice in Indiana or about public support for school choice programs

Why Indiana Parents Choose: A Cross-Sector Survey of Parents’ Views in a Robust School Choice Environment, authored by Andrew D. Catt and Evan Rhinesmith and published by EdChoice, presents results from a survey of K-12 parents within and across the public and private sectors.
Jeanne Powers, Professor at Arizona State University, reviewed the report and found significant problems with both the methods used to conduct the survey and the data analysis.
The report concludes that parents are highly satisfied with voucher and tax credit scholarship programs. The authors suggest that the findings support the expansion of school choice programs. However, even setting aside problems with the study’s data and methods, parent-satisfaction survey data almost always yield strongly positive findings. In fact, one underemphasized finding in this new survey is that substantial proportions of public school families—the largest constituency of K-12 parents in Indiana—did not participate in private school choice programs because they are happy with their current schools and want to support public schools.
But more importantly, Professor Powers finds that the survey and analysis fall short in four ways. First, three incompatible data collection methods were used to collect small samples of non-representative groups of Indiana parents. Second, the statistical analyses are too weak to draw clear conclusions. Third, while organized like a conventional research study, the report appears to be designed to advance an agenda rather than provide substantive answers to important policy questions. Finally, the report provides little new information about parents’ experiences with their children’s schools.
Thus, the report provides little useful information about school choice in Indiana or about public support for school choice programs.
Find the review, by Jeanne M. Powers, at:

Predicting Postsecondary STEM Success for Hispanic Students

A new study of Texas public high school graduates suggests that academic experiences and achievement in high school math and science are critical to Hispanic students’ postsecondary success in science, technology, engineering, and math (STEM) fields.

Nationwide, Hispanic students are underrepresented among students who complete a four-year degree in STEM areas and among workers in STEM fields. Concern is particularly acute in Texas, where Hispanic students account for 51 percent of the K–12 student population. This study, from Regional Educational Laboratory Southwest, sought to identify indicators that predict postsecondary STEM success among Hispanic students in Texas. (STEM success is defined as enrolling in, persisting in, or completing a postsecondary STEM major or degree.)

Key indicators that predict postsecondary STEM success for all student groups were the highest math course and the highest science course taken in high school. Taking Precalculus or Physics was associated with postsecondary STEM success among students who enrolled in a two-year college, and taking an Advanced Placement calculus course or Advanced Placement Physics was associated with postsecondary STEM success among students who enrolled in a four-year college. Other important indicators included the number of math courses taken, the number of science courses taken, and math and science scores on state assessments.

Hispanic students benefited from taking high school math and science courses just as much as non-Hispanic White students did. Furthermore, among students with similar academic preparation in high school, Hispanic students did not differ from non-Hispanic White students in postsecondary STEM success. This suggests that academic preparation in math and science in high school is critical to overcoming the historical pattern of lower participation in STEM majors among Hispanic students.

Friday, November 17, 2017

Spanking linked to increase in children's behavior problems

Children who have been spanked by their parents by age 5 show an increase in behavior problems at age 6 and age 8 relative to children who have never been spanked, according to new findings in Psychological Science, a journal of the Association for Psychological Science.

The study, which uses a statistical technique to approximate random assignment, indicates that this increase in behavior problems cannot be attributed to various characteristics of the child, the parents, or the home environment - rather, it seems to be the specific result of spanking.

"Our findings suggest that spanking is not an effective technique and actually makes children's behavior worse not better," says psychological scientist Elizabeth T. Gershoff (University of Texas at Austin), lead author on the study.

Historically, trying to determine whether parents' use of spanking actually causes children to develop behavior problems has been difficult, because researchers cannot ethically conduct experiments that randomly assign parents to spank or not.

"Parents spank for many reasons, such as their educational or cultural background or how difficult their children's behavior is. These same reasons, which we call selection factors, can also predict children's behavior problems, making it difficult to determine whether spanking is in fact the cause of behavior problems," Gershoff explained. "We realized that the statistical method of propensity score matching could help us get as close to an experiment as possible."

Gershoff and coauthors Kierra M. P. Sattler (University of Texas at Austin) and Arya Ansari (University of Virginia) examined data from 12,112 children who participated in the nationally representative Early Childhood Longitudinal Study. When the children were 5 years old, their parents reported how many times they had spanked their child in the past week (if any). The researchers classified any child whose parent provided a number other than zero as having been spanked.

The researchers then matched children who had been spanked with those who hadn't according to 38 child- and family-related characteristics, including: the child's age, gender, overall health, and behavior problems at age 5; the parent's education, age, and marital status; the family socioeconomic status and household size; and factors related to parenting quality and conflict in the home.

Pairing the children in this way yielded two groups of children whose main difference was whether their parents had spanked them, effectively accounting for other factors that could plausibly influence the behavior of both parent and child. This approach allowed the researchers to approximate the random assignment of participants to groups, a hallmark of experimental design.

To gauge children's behavior problems over time, Gershoff, Sattler, and Ansari examined teachers' ratings when the children were 5, 6, and 8 years old. Children's teachers reported the frequency with which the children argued, fought, got angry, acted impulsively, and disturbed ongoing activities.

The results were clear: Children who had been spanked at age 5 showed greater increases in behavior problems by age 6 and also by age 8 when compared with children who had never been spanked.
Gershoff and colleagues conducted a similar analysis with only those children who had been spanked by their parents, comparing children who had been spanked in the week before the study (which suggests frequent spanking) and those who had not. Children spanked in the past week at age 5 also experienced greater increases in problem behavior at age 6 and 8 compared with children not spanked as frequently.

"The fact that knowing whether a child had ever been spanked was enough to predict their levels of behavior problems years later was a bit surprising," says Gershoff. "It suggests that spanking at any frequency is potentially harmful to children."

"Although dozens of studies have linked early spanking with later child behavior problems, this is the first to do so with a statistical method that approximates an experiment," she concluded.

Early childhood education significantly reduces special education placement and grade retention and increases high school graduation rates

Despite calls to expand early childhood education (ECE) in the United States, questions remain regarding its medium- and long-term impacts on educational outcomes.

This study uses meta-analysis of 22 high-quality experimental and quasi-experimental studies conducted between 1960 and 2016 to find that on average, participation in ECE leads to statistically significant reductions in special education placement (d = 0.33 SD, 8.1 percentage points) and grade retention (d = 0.26 SD, 8.3 percentage points) and increases in high school graduation rates (d = 0.24 SD, 11.4 percentage points). These results support ECE’s utility for reducing education-related expenditures and promoting child well-being.

Free-play (non-academic) activities may be much more helpful in developing bilingual preschoolers’ English-language skills than teacher-structured (academic) activities.

This research compared the relative impact of different preschool activities on the development of bilingual students’ English-language skills. The study investigated whether bilingual preschool children would engage more, and use more of their second language (English), during free-play (non-academic) versus teacher-structured (academic) activities.

The researcher utilized both quantitative and qualitative research approaches; data sources included 285 preschool observations made in three classrooms in Northern California. Data analysis consisted of descriptive statistics (e.g. frequencies/percentages, mean values, and standard deviations). In addition, children’s observed scores were also analyzed by normative scales using standardized z-scores.

The findings of this study indicated that bilingual children engaged and interacted significantly more during free-play (non-academic) preschool classroom activities than during teacher-structured (academic) activities. Specific free-play activities enabling optimal engagement and second language acquisition were pretend play, free play, and monkey bars.

The study’s major implication is that free-play (non-academic) activities may be much more helpful in developing bilingual preschoolers’ English-language skills than teacher-structured (academic) activities. Free-play activities are an affordance for making language available, which helps with building academic skills and cultural capital.

This study proves that free-play activities are an affordance for language learning because bilingual children have shown dramatically greater engagement in non-academic activities (vs academic activities). The importance of free-play activities may extend beyond preschool classrooms (e.g. greater English-language development in early preschool may subsequently positively impact student performance in kindergarten). Thus, unstructured, social-based activities should be implemented for bilingual students in K-12 classrooms. Free-play (i.e. non-academic) activities should be implemented in preschool.