Open Data Exploration Presentations in Statistics

Work on integrating the Guidelines for Assessment and Instruction in Statistics Education (GAISE) of the American Statistical Association and the movement towards authentic assessment have led to ongoing evolution of the curriculum in MS 150 Statistics, an introductory statistics course.

Fall 2008 the course implemented term long statistical projects. The projects included the submission of multiple drafts during the term and a final report. The students, however, had little experience with project based learning. A typical "project" was "How many betel nuts I chewed this term." The projects did not result in the students gaining any insight into analyzing data beyond the specific type of data they had collected. In a curriculum which is dominated by statistical content and a course with up to 90 students, there was not the time available to develop the projects and provide the individualized advising that the approach would demand.

In an attempt to provide an opportunity for the students to gain insights into analyzing different types of data, in 2011 the course curriculum implemented multiple mini-projects. These projects were single draft submissions. There were roughly four of these in a term. The mini-projects revealed that when given raw data and asked to determine the most appropriate analysis, the students had difficulty determining the appropriate statistical analysis.

As noted in the spring of 2012, the students in statistics were able to calculate specific statistical results when told what statistics to calculate. When given raw data and questions about that data which did not include information on what the appropriate analysis would be, the students were stumped.

In the fall of 2012 the curriculum was overhauled with content removed from the course to provide time to explore data. These exercises were termed "open data exploration" and were driven by the American Statistical Association recommendation to "give students plenty of practice with choosing appropriate questions and techniques, rather than telling them which technique to use and merely having them implement it."

Since 2012 the course text has been expanded to provide support for choosing appropriate questions and techniques. The students are now introduced to a statistical questions to ask approach, a variables analysis approach, and a statistical tools considerations approach.

Parallel and separately from the above, tracking of student performance was shifted from spreadsheets to the Engrade on line grade book in spring 2013 after a disappointing attempt to use Jupiter Grades in 2012. In the fall of 2014 the course moved to using Schoology learning management system including the submission of assignments via Schoology. Spring 2015 the course implemented the use of on line assessment, quizzes and tests, using Schoology.

By the end of the spring 2015 term the students were well versed in the use of Schoology including familiarity with assignment locking. Assignments in statistics are usually due at the next class, with late turn-in permitted for a week. Spring 2015 the course implemented the use of locking dates to prevent the submitting of homework well after the late turn-in period expired.

During the final three weeks of the spring 2015 term the students engaged in open data exploration exercises. The third week was wrapped up by assigning an open data exploration assignment on a Wednesday and requiring the students to present their analysis and findings to the class 48 hours later on Friday.

Locking set for 10:00 class start in Schoology (click to enlarge)

An assignment lock was set to invoke at the start of each statistics class on that Friday, and then those students who had submitted their work presented their work using their submission in Schoology. Students worked either in pairs or solo on the data. For those who worked in pairs, both submitted assignments.

Brittny Rose and Loryann present their findings on a SMART board

Only those students who had submitted their work by the start of class were permitted to present, the use of Schoology's assignment locking at a specific future time made this possible. This put pressure on the student's to be prepared ahead of class and to not attempt to use the class time ahead of their presentation to work on their presentation. The assignment mimicked having 48 hours to prepare a data presentation for a meeting in a business, corporate, or other institutional setting.

June Joy presents her analysis

The students gained the opportunity to present in front of a group and to explain their analysis, adding yet another element of authenticity to the assessment of their ability to select the appropriate statistical analysis and to communicate their findings.

This analyze and present exercise occurred in the last week of the course, yet there is scant room in the curriculum to do otherwise. Bearing in mind that some students start the course not knowing the concepts of sample size, mode, median, and mean; let alone standard deviations, regressions, and confidence intervals; the first twelve weeks are necessarily content driven. The course starts with levels of measurement and by week twelve to thirteen reaches a two sample t-test for a difference of sample means. The next couple of weeks are spent on the confusing and difficult open data exploration material. When faced with data and questions, what are the appropriate choices of statistical techniques? And what do the results of those calculations mean? The students barely have the confidence to stand up and present their results even in the third week of working with data - and that is usually the final week of the course.

The open data exploration presentation adds yet another element of authentic assessment to the ever evolving curriculum of the statistics course.


Popular posts from this blog

Box and whisker plots in Google Sheets

Areca catechu leaf sheaf petiole plates

Setting up a boxplot chart in Google Sheets with multiple boxplots on a single chart