Developing Statistical Reasoning in Educational Citizen Science Projects

• In Event: 30.034 - The Ecology of Citizen Science: An Opportunity for Learning

Abstract

Objectives
The growing number of CS projects brings to the forefront the value in statistical reasoning. People are not always aware of the benefits they can gain from vast amounts of available data. In societies that become more and more technological, individuals should be able to make educated decisions related to scientific issues that affect their personal lives (e.g. causes and effects of air pollution and related policy). Making such decisions requires statistical understanding of the scientific process. Most of the statistical analysis in CS projects is done by scientists, however if engaging citizens in the statistical investigative phase provides an opportunity to develop participant statistical reasoning. In this poster we try to build bridges between two areas of research: CS and statistics education. Building these bridges might be an important step in understanding the statistical skills needed for participating in CS projects.
Statistical education research suggests that engaging learners in carefully designed exploratory data analysis can develop their inquiry skills (Ben-Zvi, 2006; Makar, Bakker, & Ben-Zvi, 2011) and model-based learning may assist learners in quantifying uncertainties in the process of making inferences from data (Manor & Ben-Zvi, 2015; Pratt, 2000). CS projects typically involve big data (Gould, Johnson, Moncada-Machado, & Molyneux, 2015) characterized with large amounts of diverse data that is rarely collected through random sampling thus may involve biases and uncertainties. These uncertainties challenges current pedagogical approaches, which will have to be refined in order to assist and develop citizens’ inferential reasoning in the context of explorations of CS data sets. Therefore, citizens’ interactions with data in the context of CS projects should be designed and studied.

Approach and data sources
In this poster we present results of a study of young students’ reasoning with statistical uncertainty in the context of informal statistical inference (ISI). The study is part of the long-term design and research Connections Project (2005-2016), a technology-enhanced-inquiry-based learning environment that aims to develop and study statistical reasoning of children in grades 4–9. Presented results originate from a five-week intervention conducted in a private school in northern Israel with two seventh grade students. Data was analyzed using interpretive microanalysis (a microgenetic method) and advanced answering the research question: How can students’ articulations of uncertainty emerge while they informally explore sampling distributions in the context of informal statistical inference?

Results
Participants succeeded in articulation of the uncertainties in the context of making ISIs. These young students understood that exploration of sampling distribution can help them decide on minimal sample size for reliable conclusions about the population.

Four instructional key ideas enabled us developing students’ reasoning with uncertainty:
• Investigate an authentic and meaningful phenomenon
• Use Inquiry-based learning activities
• Focus on key statistical ideas
• Integrate the use of appropriate technological-tools (Garfield & Ben-Zvi, 2009).

Significance
Identifying pedagogical key ideas that supported the development of students’ uncertainty reasoning while taking into account the unique characteristics of data involved in CS projects will lead to the development of fruitful ways for developing statistical reasoning in educational CS projects.

Authors

• Hana Manor, University of Haifa

• Dani Ben-Zvi, University of Haifa

• Ornit Sagy, University of Haifa