Portfolios, Accountability, and an Interpretive Approach to Validity

How can the results of classroom-based portfolio assessment be communicated outside the classroom? How might a portfolio-based assessment system be designed and implemented? How can we evaluate the merits of portfolio-based assessments?     

Maternal Representations of Attachment during Pregnancy Predict the Organization of Infant-Mother Attachment at One Year of Age

While strong retrospective and concurrent associations between maternal and infant patterns of attachment have been noted, this is one of the first reports of a prospective investigation of such associations. The Adult Attachment Interview was administered to 100 mothers expecting their first child, and, at 1-year follow-up, 96 of these were seen with their infants at 12 months in the Strange Situation. Maternal representations of attachment (autonomous vs. dismissing or preoccupied) predicted subsequent infant-mother attachment patterns (secure vs. insecure) 75% of the time. These observed concordances, as well as the discordances, are discussed in terms of the uniquely powerful contribution the Adult Attachment Interview makes to the study of representational and intergenerational influences on the development of the infant-mother attachment.     

Prior degree and student assessment performance: how can evidence guide decisions on assessment policy?

This study aimed to provide evidence to guide decisions on the type and timing of assessments in a graduate medical programme, by identifying whether students from particular degree backgrounds face greater difficulty in satisfying the current assessment requirements. We examined the performance rank of students in three types of assessments and also calculated the relative risk of failure in the summative assessments for four cohorts over all four years of a four‐year graduate medical programme. Our findings suggest that early summative assessment may disadvantage the very students who create the diversity that the faculty values, particularly if methods such as Single Best Answer questions are used, unless students are provided with additional support in bridging their knowledge gap. Use of assessment methods such as Modified Essay Questions and Objective Structured Clinical Examinations appears to be more justifiable on educational grounds.     

Student Veterans’ Outcomes by Higher Education Sector: Evidence from Three Cohorts of the Baccalaureate and Beyond

We use nationally representative survey data from three cohorts of bachelor’s degree recipients to examine military veterans’ student loan debt at graduation and their employment and earnings 1, 4, and 10 years after graduation, relative to observably similar graduates. We also examine whether the effects of veteran status depend on college sector. Using propensity score weights with covariate adjustment, we find employment and earnings premiums for veterans 1 and 4 years after graduation, and an employment penalty but earnings premium (conditional on employment) 10 years after graduation. We find no clear evidence of a for-profit employment or earnings penalty among bachelor’s degree recipients, including veterans. We do find that veterans are as likely as similar non-veterans to graduate with debt, and that graduating from a for-profit institution yields markedly higher debt amounts, especially for veterans, despite the availability of Montgomery GI Bill benefits for these cohorts.