Year 13 students (age 17-18 and mostly males) who were studying for the Business and Technology Education Council (BTEC) qualifications undertook 9 hours of a project based learning, aimed at developing IT students’ statistical literacy.
The students have studied basic statistical concepts, such as appropriate graphical representation involving discrete, continuous and grouped data; and appropriate measures of central tendency (mean, mode, median) and spread (range, consideration of outliers), scatterplots, etc. They engaged in a data analytics process using statistical measures and data visualisation tools, such as mean, median, mode, range, inter quartile range, standard deviation, histogram, box plot, stem and leaf and so on with a variety of IT tools, in particular CODAP and Excel.
The students investigated airborne pollutants, such as particulate matter (PM10) and Nitrogen Dioxide (NO2) and weather data provided by the UK Met Office. These pollutants have been identified as being dangerous to health and are generated by industrial and consumer activities, specifically diesel vehicle emissions.
Extracted from larger data (approx. 28,000 records) containing anomalies (missing data, incorrect data etc.) provided by Exeter City Council, one of the datasets given to students included 720 cases with hourly PM10 data from November 2015, 2016 and 2017 and time of the day (morning, afternoon, evening, night) in two different locations in the city (Alphington and Royal Albert Memorial Museum (RAMM) in Exeter, UK).
After relatively free explorations with the data, the students were asked to write a report on questions such as ‘Have PM10 levels reached a dangerous level?’, ‘Are PM10 levels rising over time?’, and so on.
The students used both tables showing different sets of raw data for different locations and times and histograms for their data analysis. Looking at the raw data also allowed the students to acknowledge some anomalies in the PM10 values. The various graphs such as histograms helped them interpret the range of values where most of the data were clustered to compare the PM10 levels in both locations.
Click to enlarge
A post attitude survey suggests that the students have more positive attitudes towards learning statistical concepts, are less stressed during the learning process, and make better use of technological tools.