Search

A frequently asked question in quantitative research is how to compare two samples that include some combination of paired observations and unpaired observations. In our publications and R package, we refer to the scenario as "partially overlapping samples". Most frequently the desired comparison is that of central location. Depending on the context, the research question could be a comparison of means, distributions, proportions or variances. In the 20th century, traditional approaches that discard either the paired observations or the independent observations were customary. In the 21st century approaches that make use of all available data are becoming more prominent. Traditional and modern approaches for the analyses for each of these research questions are reviewed. We conclude that tests that report a directly measurable difference between the two groups provide the best solutions.

Use of herbal medications and supplements has experienced immense growth over the last two decades, with retail sales in the USA exceeding $13 billion in 2021. Since the Dietary Supplement Health and Education Act (DSHEA) of 1994 reduced FDA oversight, these products have become less regulated. Data from 2012 shows 18% of U.S. adults used non-vitamin, non-mineral natural products. Prevalence varies regionally, with higher use in Western states. Among preoperative patients, the most commonly used herbal medications included garlic, ginseng, ginkgo, St. John's wort, and echinacea. However, 50-70% of surgical patients fail to disclose their use of herbal medications to their physicians, and most fail to discontinue them preoperatively. Since herbal medications can interact with anesthetic medications administered during surgery, the American Society of Anesthesiologists (ASA) and the American Association of Nurse Anesthetists (AANA) recommend stopping herbal medications 1-2 weeks before elective surgical procedures. Potential adverse drug effects related to preoperative use of herbal medications involve the coagulation system (e.g., increasing the risk of perioperative bleeding), the cardiovascular system (e.g., arrhythmias, hypotension, hypertension), the central nervous system (e.g., sedation, confusion, seizures), pulmonary (e.g., coughing, bronchospasm), renal (e.g., diuresis) and endocrine-metabolic (e.g., hepatic dysfunction, altered metabolism of anesthetic drugs). During the preoperative evaluation, anesthesiologists should inquire about the use of herbal medications to anticipate potential adverse drug interactions during the perioperative period.

Standard approaches for comparing the means of two samples, comprising both paired observations and independent observations, involve the discarding of valuable information. An alternative test which uses all of the available data, is the partially overlapping samples t-test. Two variations of the test are available, one assuming equal variances, and one assuming separate variances. Issues with standard procedures, and considerations for choosing appropriate tests in the partially overlapping scenario are discussed. An example with details of how to apply the partially overlapping samples t-test is given along with an R package that implement these new tests.

Background: Understanding the natural history of developmental speech and language impairments can support the selection of children whose difficulties are persistent rather than transitory. It can also provide information against which the effectiveness of intervention can be evaluated. However, natural history data are difficult to collect ethically. Furthermore, as soon as an impairment is identified, the behaviour of those around changes, thus creating some level of intervention. Longitudinal cohort studies, where intervention is minimal, or the control arm of randomized trials have provided the best evidence. However, occasional opportunities arise where service waiting lists can provide data about the progress of children who have not received intervention. This natural history study arose within an ethnically diverse, community paediatric speech and language therapy service in the UK where levels of social disadvantage are high. Aims: To identify (1) the characteristics of the children who attended initial assessment and were selected for treatment; (2) the differences between children who did and did not attend reassessment; and (3) the factors associated with outcomes. Methods & Procedures: A cohort of 545 children were referred and assessed as in need of therapy. Due to resource constraints, intervention was not available for an average of 12 months. Children were invited to attend for a reassessment of need. Initial and follow-up assessments were conducted by experienced clinicians using service guidelines and the Therapy Outcomes Measures Impairment Scale (TOM-I). Descriptive and multivariate regression analyses examined child outcomes for changes in communication impairment, demographic factors and length of wait. Outcomes & Results: At initial assessment, 55% of children presented with severe and profound communication impairments. Children offered appointments at clinics in areas of high social disadvantage were less likely to attend reassessment. By reassessment, 54% of children showed spontaneous improvement (mean TOM-I rating change = 0.58). However, 83% were still judged to require therapy. Approximately 20% of children changed their diagnostic category. Age and impairment severity at initial assessment were the best predictors of continuing requirement for input. Conclusions & Implications: Although children do make spontaneous progress post-assessment and without intervention, it is likely that the majority will continue to be assigned case status by a Speech and Language Therapist. However, when evaluating the effectiveness of interventions, clinicians need to factor in the progress that a proportion of the caseload will make spontaneously. Services should be mindful that a lengthy wait may disproportionately impact children who already face health and educational inequalities.

Problem solving is a fundamental skill that chemistry graduates should possess, yet many students have difficulties solving problems in chemistry. These difficulties may be either student- or instructor-driven. Instructor-related difficulties could stem from some teaching practices, such as expecting students to apply procedures without requiring them to show their reasoning or solely focusing on worked examples. Such practices could inhibit the development of problem-solving skills. To address these challenges, our group developed a metacognitive scaffold (Goldilocks Help) to support both students and instructors through structured problem solving. This scaffold breaks down the problem-solving process into phases and places emphasis on reasoning required throughout that process. This study explored how teaching associates (TAs) used the problem-solving scaffold and how this practice affected their teaching and perceptions of student learning. Seven TAs based at a large research-intensive Australian university were interviewed, and the data were analysed using the framework approach. Teaching with the problem-solving scaffold was found to be beneficial, albeit with initial student resistance. The scaffold provided a common thinking structure between the TAs and students, enabling TAs to easily identify mistakes and address specific areas of concern. However, TAs also experienced students' attention shift from content to the scaffold. Initially, many students unproductively viewed the process as requiring two separate actions of solving the problem and being explicit about the problem-solving process they used, as opposed to an integrated activity. Through constant reinforcement and prompting by TAs during and prior to solving the problem, students continued to grasp how to effectively internalise the scaffold to assist their problem solving. Understanding how TAs use problem-solving scaffolds with students will add to the field of education research to inform innovations in supporting the development of students' problem-solving skills.