Webinars

TSHS Fall 2024 Webinar: “Clarifying practical relevance of p-values and confidence intervals. Clear choice of words and images, that foster conceptual understanding and steer away from misconceptions.” The ASA Section on Teaching of Statistics in the Health Sciences (TSHS) is excited to present our Fall 2024 webinar. Our speaker will be Hilary Watt of the Imperial College London. She will discuss the practical relevance of p-values and confidence intervals. The webinar is FREE and open to all. Details and registration information are below. TITLE: Clarifying practical relevance of p-values and confidence intervals. Clear choice of words and images, that foster conceptual understanding and steer away from misconceptions. PRESENTER: Hilary Watt, Imperial College London DATE/TIME: Tuesday, October 29, 2024, 3:00pm Eastern Time (US & Canada) VENUE: Online webinar hosted using the Zoom platform ABSTRACT: Informal p-value and CI interpretations are often used in health science teaching yet may be too crude to help much in developing understanding. Not all students deduce CIs and p-values practical relevance from their definitions. This talk suggests transparent explanatory interpretations, that promote understanding of their practical relevance. They are compact enough for routine repetition whilst teaching. Referring to the sample (not blood sample!) as participants provides a clear starting point. Repeatedly clarifying the relationship of participants to the population, including the random sampling assumption, addresses the issue that ‘population’ is poorly understood. Implications of random sampling being impractical can be discussed. Images of the distribution of sample statistics are useful for understanding calculation methods yet can readily be misinterpreted. Strategies are suggested to offset the risk of common misinterpretations. This includes use of images designed to foster conceptual understanding, for both CIs and for p-values (for p-values, illustrating their continuous nature). The intention is to make statistics easier and more intuitive to understand. The inspiration is to steer towards higher standards of data interpretation within health sciences. REGISTRATION: To register for the webinar, please complete this form: https://uniofqueensland.syd1.qualtrics.com/jfe/form/SV_4I9TBkB3xKNS7AO We will send the Zoom link for the webinar to your email address the day before the webinar.

The challenges of teaching introductory data analysis in an online environment are well known. These challenges can increase when the primary audience for the course are students pursuing non-quantitative degrees. In this talk, we will discuss the development of a fully online synchronous course designed for such a student audience, specifically Master of Public Health (MPH) students. Both problem-based learning and experiential learning theory methodologies informed course design. Students in the class worked individually and as team scientists to complete a data analysis project. They were exposed to data analysis elements from project initiation to dissemination while simultaneously learning methodologic concepts. Although the course was designed for MPH students, an instructor could modify the course for any cohort of students in an introductory statistics course where the focus is application and communication. Both course development and design will be discussed, and evaluations from both students and the instructor will be provided.

Null hypothesis significance testing (NHST) with a critical level of significance of 5% (P<0.05) has become the cornerstone of research in the health sciences, underpinning decision making. However, considerable debate exists about its value with claims it is misused and misunderstood. It has been suggested it is because NHST and P-values are too difficult to teach, and encourage dichotomous thinking in students. Consequently, as part of statistics reform it has been proposed NHST should no longer be taught in introductory courses. However, this presentation will consider if the misuse of NHST principally results from it being taught in a mechanistic way, along with claims to knowledge in teaching and erosion of good practice. Whilst hypothesis testing has shortcomings, it is advocated it is an essential component of the undergraduate curriculum. Students’ understanding can be enhanced by providing philosophical perspectives to statistics, supplemented by overviews of Fisher’s and Neyman-Pearson’s theories. This helps the appreciation of the underlying principles of statistics based on uncertainty and probability, plus the contrast of statistical with contextual significance. Moreover, students need to appreciate when to use NHST rather than being taught it as the definitive approach of drawing inferences from data.

The idea of developing a rubric for assessments or flipping lectures in an Applied Biostatistics (or even Applied Statistics) classroom can be overwhelming, but it does not have to be. I will lead a discussion introducing several ideas for building a rubric for statistics assignments and exams, and flipping parts of a lecture to combine traditional lecture with interactive components to fully engage students to enhance their learning in the classroom or live synchronous sessions (like teaching through Webex or Zoom) using polling software like PollEverywhere. The polling software strategy I introduce will also provide instructors real-time feedback regarding students’ current comprehension of the material. One of the techniques can also be modified to increase engagement for an online only format (pre-recorded lectures). Attendees who consider themselves beginners with respect to rubrics or flipped classrooms as well as those who consider themselves more experienced are welcome to this webinar.

A new form of pedagogical approach was thrust upon us by the pandemic – online classrooms; a concept that was never experienced or experimented to the extent that was witnessed in the past few months. The challenges involved in online teaching are many, especially for a discipline like statistics which is essential for students undergoing courses in science, health science as well as social sciences. The challenges can be anything, including difficulty in comprehending foundational concepts through virtual classrooms, lack of availability of technical tools such as electronic gadgets or internet coverage, lack of online teaching tools and resources, absence of appropriate technical knowhow which can hinder the ease of communication between the facilitators and students, and lack of appropriate evaluation of student performances. The extent of these challenges may vary from region to region and would depend upon socio-economic profiles of the places. However, the global academic fraternity in the statistics community is committed to effective dissemination of statistics content and knowledge to students from multiple disciplines amid these changed circumstances. In such a scenario, it will be both important as well as informative to have a platform for experience sharing of experts from around the world. This will not only help us exchange information regarding multiple academic approaches and evaluation aids successfully implemented by the statistics fraternity, but also provide significant insights into the availability of shared resources and identify what worked well in different geographical regions.

Many areas of clinical research, such as comparative effectiveness research and patient-centered outcomes research, strongly depend on making causal inferences from observational data. Further, these topic areas also utilize pragmatic trials and quasi-experimental designs, where consistent estimation of causal effects is also more challenging than traditional randomized controlled trials, and/or involves distinct approaches for intention-to-treat versus as-treated or per-protocol effects. While substantial literature exists on associated designs and analysis strategy, the corresponding methods are complex and not always taught in formal training, even within graduate statistics or biostatistics programs. Therefore, a critical need exists for accessible educational resources and the expansion of relevant courses and training programs. Regarding that goal, however, significant debate exists on whether these advanced methods should even be taught at all to non-statisticians, and/or researchers with more limited statistical training (e.g. a fundamental course and some background in regression). This talk proposes some possible perspectives to effectively address these concerns, while still avoiding the result of "knowing enough to be dangerous". The presenter has some related links at www.landsittellab.pitt.edu. This work was supported by AHRQ grant R25HS023185, PCORI contract R-IMC-1306-03827, and supplemental funding from the NIH/NLM grant 5 T15 LM007059-32.

The TSHS Resources Portal (www.causeweb.org/tshs) contains datasets from 13 real health sciences research studies. Each dataset is accompanied by a study description and a data dictionary. Most are linked to a published paper as well. These datasets, plus some extra teaching tools, are peer reviewed and ready for use with your learners. In this webinar, Amy and Carol will walk through what is available and how to get the most out of this resource.

Over the past 10 years we have adopted a variety of new teaching methods to make both our in-person and our online introductory biostatistics courses more active, relevant and effective. These include the flipped classroom approach, active learning, collaborative answer keys, and group projects using “The Islands”. The virtual world of The Islands, created by Michael Bulmer at the University of Queensland, allows students to actually do research (and statistics) from start to finish by designing, executing, analyzing and reporting the results of a “real” study on virtual people. We have collaborated with Dr. Bulmer to add features to The Islands (such as clinics and hospitals) to facilitate health-related research studies, both experimental and observational. Carrying out an Island study provides students with sometimes painful but nevertheless invaluable experience in many aspects of research, including study design, data collection, teamwork, data analysis, and communicating research results to others. This webinar will describe The Islands and how we use them for student projects and will discuss the benefits and challenges of these projects, both for students and for instructors. Webinar Recap: https://tshsblog.wixsite.com/main/single-post/2020/05/01/TSHS-Spring-Webinar-Recap-Explore-The-Islands

The idea of a "flipped classroom" has been integrated for two years into the introductory biostatistics course required of all Masters of Public Health (MPH) students at the University of Michigan. The course was divided into eight modules, with each module consisting of one or more video lectures and three modes of assessment: a quiz and two in-class projects. The in-class projects consisted of (1) data analysis of contemporary public health data sets using Excel and (2) review of statistical methods and results in manuscripts published recently in the American Journal of Public Health. This talk will review my experiences with the development of the course, with the implementation of the course, and student input received from anonymous end-of-semester evaluations. Registration: Please use the following form to register: https://redcap.hfhs.org/redcap/surveys/?s=4WH8JJ9KYH. The webinar link will be sent to you ahead of the session, and a link to the webinar recording will be sent to you about a week after the session.

Flipped classrooms have appeared in all levels of education. One of the major benefits is that the passive learning (lecture) is completed at home and the active learning (activities and problem solving) are done in class with the instructor. However, the issues with flipped classrooms are the cost to make high quality video content and the time. Due to the cost and time many classes are created and then not updated. This talk will discuss common ways for creating and updating flipped classrooms, considering a case study of PHP 2560: Statistical Programming in R at Brown University. We will discuss the first flipped version of this course, in terms of content and creation time. Then we will discuss how subsequent iterations have been adapted and updated to maintain relevance.