Publications by authors named "J W H Leer"

This study examined how living in a gentrifying neighborhood may impact adolescents' reading and math achievement via educational aspirations and psychological distress and asked whether these pathways differ according to socioeconomic status and race. A framework combining theories of adolescent development and neighborhood effects was empirically tested using a racially diverse sample of adolescents living in urban neighborhoods in North Carolina matched to administrative school records and census data ( = 1,045, = 12, 8% American Indian, 4% Asian, 32% Black, 62% White, 15% multiracial, 16% Latinx, categories not mutually exclusive). At the population level, structural equation models found no relation between the extent of gentrification occurring in youths' neighborhood of residence and reading and math achievement, educational aspirations, or psychological distress.

View Article and Find Full Text PDF

This study examined the relation between schools' color-evasive versus multicultural diversity ideologies, school characteristics, and adolescent development. Across two datasets linking individual-level survey data (N = 1692) and administrative records (N = 300,063; M = 12.4, 52% female, 48% male), schools' stated support for diversity (via a pro-diversity mission statement) was related to adolescent mental health and academic achievement, but in nuanced ways depending on individual racial/ethnic backgrounds, the racial/ethnic diversity of the student body and teachers, and the extent of racial disparities in discipline and gifted education.

View Article and Find Full Text PDF

Children's socioeconomic status (SES) is linked to disparate access to resources and affects social behaviors such as inclusion and resource allocations. Yet it is unclear whether children's essentialized view of SES (i.e.

View Article and Find Full Text PDF

Political violence affects more than 25% of children globally, yet little is known about how to support positive adaptation among conflict-affected children. Using a sample of 3797 Nicaraguan child-caregiver dyads (M  = 1.5 years, M  = 5.

View Article and Find Full Text PDF
Article Synopsis
  • Deep learning models for auto-segmentation in radiotherapy were evaluated for their effectiveness in segmenting cancerous areas, focusing on both quantitative and qualitative measures.
  • The study involved training separate models for left- and right-sided breast cancer, measuring the time taken for automatic and manual segmentation, and comparing them using several scoring techniques.
  • Results showed significant time savings with auto-segmentation—averaging about 42% to 58% reduction in time—while maintaining high accuracy, as 92% of automatically generated contours were deemed clinically acceptable.
View Article and Find Full Text PDF