STATISTICS EDUCATION RESEARCH JOURNAL

Editorial

JENNIFER KAPLAN, SUE PETERS — Thu, 30 Nov 2023 00:00:00 +1300

USING A MIDTERM WARNING SYSTEM TO IMPROVE STUDENT PERFORMANCE AND ENGAGEMENT IN AN INTRODUCTORY STATISTICS COURSE: A RANDOMIZED CONTROLLED TRIAL

Thu, 30 Nov 2023 00:00:00 +1300

This article reports on an evaluation the effectiveness of e-mailed grade “nudges” on students’ performance and engagement in an introductory statistics course for undergraduate health science students. In 2020–2021, 358 students were randomized to an e-mail (n = 178) or no e-mail (n = 180) group. The intervention e-mail contained information on each student’s predicted final grade (grade nudge). Using two-sample t-tests, the statistical analysis of final grades in the course showed a higher compatibility with a model of no mean difference for students in the e-mail (73.5%) vs. no e-mail (72.1%) group. Comparison of the distributions of final grades between the two groups, however, suggested the e-mailed nudges may be related to slight improvements in final grades. Specifically, the median final grade was higher in the e-mail group (74.6 vs. 72.4); the Q1 value in the e-mail group was also higher, and the interquartile range was similar: no e-mail group (15.8) vs. e-mail group (14.2). Students also completed the Scale of Student Engagement in Statistics (SSE-S). Total engagement, affective and cognitive subscale scores of the SSE-S were higher in the e-mail group, resulting in low compatibility with a model of no difference in engagement scores. Overall, the results showed there is potential for our midterm warning system to be used to improve outcomes, particularly given that it is simple to implement, cost-effective, and easily scalable.

MODEL-ELICITING ACTIVITY WITH CIVIL ENGINEERING STUDENTS TO SOLVE A PROBLEM INVOLVING BINOMIAL DISTRIBUTION

MARTHA ELENA AGUIAR BARRERA, HUMBERTO GUTIÉRREZ PULIDO, VERÓNICA VARGAS ALEJO — Thu, 30 Nov 2023 00:00:00 +1300

This research presents the results of the implementation of a model-eliciting activity called Brickyards, designed to promote the learning of the binomial distribution. The theoretical framework used was the Models and Modeling Perspective, and the participants were undergraduate students enrolled in a probability and statistics course of the Bachelor Civil Engineering Program at the University of Guadalajara, Mexico. The activity was refined during three semesters, and here we report the models generated by the students in the fourth implementation. In the first stage of the activity of this implementation, students proposed wrong solutions, which were based on ideas of proportionality and linear thinking. The activity was designed to inhibit these types of solutions and to encourage students to realize when they are dealing with a random phenomenon, and that they need a probability distribution to solve the activity. The students used RStudio software to calculate probabilities.

STATISTICS ATTITUDES AFTER USING GUIDED PROJECT-BASED LEARNING AS AN ANDRAGOGICAL STRATEGY IN A GRADUATE STATISTICS COURSE

ADAM ELDER — Fri, 01 Dec 2023 00:00:00 +1300

This study builds on previous studies that have examined guided project-based learning in undergraduate statistics courses to examine students’ attitudes toward statistics after participating in a graduate-level statistics course that used guided project-based learning as an andragogical technique. This phenomenological qualitative case study utilized multiple student interviews and reflections over a semester-long statistics course in a doctoral education degree program. The results showed that guided project-based learning immersed students in the quantitative inquiry process and emboldened them to read and use statistics in their academic and professional lives. It also revealed several elements of guided project-based learning that are important for instructors looking to implement this approach in their own courses.

BRAZILIAN RESEARCH IN STATISTICS, PROBABILITY AND COMBINATORICS EDUCATION: A LOOK AT THESES

LORI VIALI, MAGNUS CESAR ODY, CLARISSA CORAGEM BALLEJO, ELISABETE RAMBO BRAGA — Tue, 19 Dec 2023 00:00:00 +1300

In this article, we first analyze the situation of statistics education in Brazil, within an international context of interest for this discipline. Secondly, we aim to answer the following research question: How has doctoral research in statistics, probability, and combinatorics education evolved in Brazil? For this purpose, we analyzed the Brazilian doctoral research production from 1994 to 2021. We identified 102 completed dissertations in doctoral programs over the last 28 years. Within this research context, we analyzed the graduate programs, advisors, research themes, grade levels and participants studied, theoretical frameworks, and the data analysis methodologies employed. Even though doctoral research has developed slowly and irregularly, there has been an increase in the number of studies, driven by the growth of graduate degrees offered in the last few years.

OPPORTUNITIES TO LEARN MEAN, MEDIAN, AND MODE AFFORDED BY TEXTBOOK TASKS

KARIN LANDTBLOM — Mon, 04 Dec 2023 00:00:00 +1300

This research paper examines tasks related to mean, median, and mode in seven Swedish textbook series for students aged between 10–13 years. The tasks were analysed based on context, mathematical properties, input and output objects, and transformations. These categories allowed for a thorough analysis of the opportunities afforded to students to understand these measures. The analysis revealed that most tasks focus on the mean and on procedural transformations with quantitative values. The findings suggested that the textbooks do not afford enough explicit context for students to develop a deep understanding of the mathematical properties of different measures of central tendency. By analysing various textbooks, a broader understanding of the learning opportunities afforded to students was gained. The discussion includes the implications of these results for task design.

ANALYSING COSTA RICAN AND SPANISH STUDENTS’ COMPARISONS OF PROBABILITIES AND RATIOS

CARMEN BATANERO, LUIS A. HERNÁNDEZ-SOLÍS, MARÍA M. GEA — Thu, 30 Nov 2023 00:00:00 +1300

We present an exploratory study of Costa Rican and Spanish students’ (11–16-year-olds) competence to compare probabilities in urns and compare ratios in mixture problems. A sample of 704 students in Grades 6 through to Grade 10, 292 from Costa Rica and 412 from Spain, were given one of two forms of a questionnaire with three probability comparison and three ratio comparison problems each. The full questionnaire consisting of both forms covers six different proportional reasoning levels for each type of problem. We analysed the percentages of correct responses to the items and the strategies used by students in each grade and in each country. The results suggest students had the highest difficulty in comparing probabilities for items at the same proportional reasoning level. The results also point to the probabilistic biases in students’ responses, although these biases were less frequent than in previous research.

RESOURCES AND TENSIONS IN STUDENT THINKING ABOUT STATISTICAL DESIGN

KELLY FINDLEY, BREIN MOSELY, AARON LUDKOWSKI — Thu, 14 Dec 2023 00:00:00 +1300

Reform efforts in statistics education emphasize the need for students to develop statistical thinking. Critical to this goal is a solid understanding of design in the process of collecting data, evaluating evidence, and drawing conclusions. We collected survey responses from over 700 college students at the start of an introductory statistics course to determine how they evaluated the validity of different designs. Despite preferring different designs, students offered a variety of productive arguments supporting their choices. For example, some students viewed intervention as a weakness that disrupted the ability to generalize results, whereas others viewed intervention as critical for identifying causality. Our results highlight that instruction should frame design as the balancing of different priorities: namely causality, generalizability, and power.

THE IMPACT OF WORKING IN A STATISTICAL LABORATORY ON THE STATISTICS STUDENT COLLABORATORS

Fri, 01 Dec 2023 00:00:00 +1300

Graduate level statistics education curricula often emphasize technical instruction in theory and methodology but can fail to provide adequate practical training in applications and collaboration skills. We argue that a statistical collaboration center (“stat lab”) structured in the style of the University of Colorado Boulder’s Laboratory for Interdisciplinary Statistical Analysis (LISA) is an effective mechanism for providing graduate students with necessary training in technical, non-technical, and job-related skills. We summarize the operating structure of LISA, and then provide evidence of its positive impact on students via analyses of a survey completed by 123 collaborators who worked in LISA between 2008–15 while it was housed at Virginia Tech. Students described their work in LISA as having had a positive impact on acquiring technical (94%) and non-technical (95%) statistics skills. Five-sixths (83%) of the students reported that these skills will or have helped them advance in their careers. We call for the integration of stat labs into statistics and data science programs as part of a comprehensive and modern statistics education, and for further research on students’ experience in these labs and the impact on student outcomes.