by Ann Holmes, University of Tennessee, Chattanooga
Sheila Roberts, Bowling Green State University
Eleanour Snow, University of South Florida/University of Texas at Austin
Ignacio Pujana, University of Texas at Dallas
authored as part of the 2010 workshop, Teaching Geoscience Online - A Workshop for Digital Faculty
This page addresses the challenges of using real data in lessons, especially as they affect on-line teaching. It offers:
- Real data are often "messy."
- Students are not familiar with data analysis and plotting tools like Excel.
- Students are easily frustrated by technology, even when instructors think that the directions are clear and complete.
- There is a tremendous amount of data out there; how do we find good data sets that work?
- Students often don't think about what the data mean, or what the graph means, so they don't always recognize when they have an answer that makes no sense.
- On-line students may be working in isolation and therefore have no one to ask when things go wrong.
- Create an introductory (within the first week) assignment that serves to test computer or mathematical base-level skills necessary for successful completion of your course. A "can you do it?" exercise rather than a "how well can you do it?" exercise. If students do not possess the skills, they find out early enough to drop the class.
- Start small. Incorporate simple activities early in the semester, building to more complex ones later on. This builds confidence and independence.
- Ask them pointed questions about graphs and data in general before asking about a particular graph or data set. This helps you see their basic level of math understanding, and fix any obvious problems. If they don't know that the slope on a graph is showing the rate of change, they won't be able to interpret what a change in slope means.
- Clean-up really messy data sets. Sometimes real data are too hard for beginning students to interpret. Don't make it so clean that it is obviously fake -- all the data points should not be exactly on the line -- but do make it clean enough so that the students can make reasonable interpretations.
- Try it first. If you find something you want to use, or create something you want to assign, try it out. Better yet, have a friend try it out. Nothing is more frustrating to student to get partway through a problem and find it is not working because you're dividing by zero, or a step was left out.
- Use short video clips (e.g., Jing-generated or Flip film clips) to illustrate methods and procedures.
One of the hardest things about using data in an online in environment is making the instructions clear enough for the students to follow. One solution for that is to make a short video showing them how to do things. The free-ware Jing is simple to learn, simple to use, and can make short demonstration clips. Download it for MAC or PC at jingproject.com
A more robust screen-capture video editing system is SnapZPro which is available for a small cost, and creates Quicktime movies for MAC.
The demonstrations below were made in less than 10 minutes on a MacBook Pro, using Jing and the built-in microphone. One tip: write out the steps you want to demonstrate before you start recording, and use the pause to change the screen if you need to.
- How to make a simple calculation in Excel: Basic Excel Calculations-MAC
- How to use formulas in Excel: Formulas in Excel-MAC
- How to make an X-Y plot in Excel: X-Y Plots in Excel-MAC
- How to download data from the USGS River Site Getting River Data from USGS upload data ( 13.2MB Jun29 10) using a PC
- Demo, Mercury in Fish Mercury Calculation Demo-MAC
- Demo, Vostok Ice Core Task 1-MAC Task 2-MAC Task 7-MAC
Exemplary Online Lessons
- Analyzing Plate Motion Using EarthScope GPS Data was created by S. Olds, S. Ericsson, D Munson & L. Dahlman. In this chapter, users access GPS data from the Plate Boundary Observatory; they analyze the data in a spreadsheet to measure the motion of GPS stations in the Pacific Northwest. From their analyses, users generate and map annual velocity vectors of GPS stations. Finally, they explore patterns in the direction and length of velocity vectors on the map to understand tectonic motion and surface deformation associated with the subduction of the Juan de Fuca plate under the North American plate.
- Plotting Earthquakes using near real-time data exercise- from SERC visualization website
- Exploring Seafloor Topography is an activity created by Sandra Swensen as part of the Earth Explotation Toolbook website. It uses GeoMapApp.
- Detecting El Niño in Sea Surface Temperature Data was written by D. Smith, and B. Youngman. This module introduces users to normal seasonal sea surface temperature variation as well as extreme variation, as in the case of El Niño and La Niña events. Students learn how to download seasonal data, analyze it using GIS, then look for the tell-tale SST signature of El Niño and La Niña events. At the conclusion of the module, users are given the opportunity to analyze a season of their own choosing to determine if an El Niño and La Niña pattern emerged in that year's data.
- Vostok Ice Core Lab - from Stephanie Pfirman at Barnard College, is an excellent on-line, data rich lab.
Collections of Online Data-rich Lessons
- GeoMapApp is an earth science exploration and visualization application from the Lamont-Doherty Earth Observatory of Columbia University. This free Java-based tool which works on any type of computer, allows a broad range of geoscience data sets (digital elevations, earthquake data, geochemistry, plate motion information, research cruise tracks and profiles, sediment thickness grids, hydrothermal vent images, amongst many others) to be explored and viewed in their correct geographical context. Lessons and resources are posted at GeoMapApp-Resources and a page about GeoMapApp in Geoscience Education links to 14 activities using GeoMapApp.
- MARGINS has been awarded an NSF Course, Curriculum, and Laboratory Improvement (CCLI) grant to develop web-based undergraduate classroom teaching modules, in cooperation with SERC at Carleton College. As a part of this collaboration, a series of workshops for researchers and educators has been conducted. These workshops have facilitated the creation of a body of teaching activities called Mini-lessons , which are learning materials that repurpose data and resources developed through MARGINS for use in examining earth processes in undergraduate classrooms from a multidisciplinary perspective.
- The Earth Exploration Toolbook is a collection of computer-based Earth science activities. Each activity, or chapter, introduces one or more data sets and an analysis tool that enables users to explore some aspect of the Earth system. Step-by-step instructions in each chapter walk users through an example — a case study in which they access data and use analysis tools to explore issues or concepts in Earth system science. In the course of completing a chapter, users produce and analyze maps, graphs, images, or other data products. The ultimate goal of each activity is to build user's skills and confidence so they can use data to conduct their own investigations of the Earth system. NOTE: Most of these activities are designed for high schol students, but some are for undergraduates or could be easily adapted for introductory courses. See the teaching notes on each exercise for guidance.
- DLESE is the Digital Library for Earth System Education, a free resource that supports teaching and learning about the Earth system.
- Teaching with Data, Models, and Simulations - this collection contains over 80 activities that use data to teach various geoscience topics.
- Teaching Quantitative Skills- more than 200 activities that take a quantitative approach to teaching geoscience.
- Calculating the Level of Mercury in Fish: by Eleanour Snow. This is a fairly straight-forward use of data to determine whether various fish species are safe to eat. The calculations can be done fairly easily in Excel. Students are often confused when some species register >100% of the EPA recommended safe mercury level; apparently there is a misconception among some that 100% is the highest percent one can calculate. mercury (Acrobat (PDF) 250kB Jun29 10)
- A Lesson in Research Reporting using Real Data: by Ignacio Pujana. The purpose of this "research paper‟ is to familiarize the student with research involving a significant tectonic event of his or her choice that is happening right now. There is an amazing amount of real-time data being captured and posted on the WWW. The student first decides on a topic - a recent earthquake or volcanic eruption. They must choose one significant enough to be reported on the web and elsewhere. They must then find both real-time data and journal articles about the event and the tectonic setting.
The report should be very short, a maximum of 2 or 3 pages in length, and include a concise analysis of the subject. There is an example and a template provided as a model for the student's report. The report is graded using a rubric which provides a quick, uniform way to do it.
Short Research Report (Acrobat (PDF) 203kB Jun29 10) : [link file 23108 'Example Paper'] : [link file 23109 'Grading Rubric']
There are many great resources available on the internet, and these can used as-is or adapted for your class. The adaptation may be a simple as putting your class name and assignment number at the top, or it may involve changing elements of the material to better suit your needs. In either case, be sure always to attribute the original author in all materials you use. Here is an example:
Creating a Flood Map: Eleanour Snow. The original, created by Benjamin Laabs is entitled Floods on the Minnesota River - Planning for the 100 Year Flood in St. Peter Minnesota. It uses data from the USGS Real Time Water Data Site calculations in Excel, and topographic maps to calculate and map a 100 year flood. This can be adapted by selecting river close to home.
Adapting the Lab: First, I needed to change some things in the lab overall. It was laid out to be done partly in groups, and over several class periods. I reframed it to be done by individuals. Then, I searched through the USGS river data for a river closer to home that had data available, that went through a town, and had topo maps available on-line. The West Fork of the Trinity River was a good choice on most fronts, but it only had Gage Height data for 6 years, a problem on many local rivers. So I downloaded the data, and made up some gage height data that made sense, and compiled it into an excel file. Then I went to the USGS Topographic Maps site and located the topo maps for the town, turns out there were two (Eluess and Irving quads). Note that printing the maps in black and white on an 8.5x11 inch piece of paper results in a map that is difficult to read. You can see the map better when printed in color, but students should be encouraged to look at the file on their computer, where they can zoom into the area of interest. Students can draw the 100-year flood plain online in a bright color (say red) and take a photo or screen capture to send to you for grading.
Here are the files I used:
Extensions: Sheila Roberts. To make the exercise more challenging, you can have the students download the data themselves; you might even let them choose the river. Having students download data, draw and interpret those data, and draw a floodplain is challenging in a face-to-face class. It is even more challenging in an online environment. The following file explains how to download data from the USGS, put it into an Excel file, and make the appropriate plots.
Instructions for downloading and manipulating river data (Microsoft Word 2007 (.docx) 14kB Jun25 10)
To illustrate the challenges in using real data, the two data files below were created for the Trinity River from the raw data available online. The first contains the last 7 days of data, and the second contains peak discharge data from 1925 to 2009.
There are several questions you can ask about this data set. First, I would ask the students to plot discharge and stage for the last 7 days. These plots are shown on the USGS website when you download the data, so students can check if their plots look right. They can also make a rating curve (discharge vs stage). You can ask them what is the discharge when the gage height is 3 feet and what the gage height is when the discharge is 300 cfs. These numbers may need to be modified if there is abnormally high or low discharge the week you do the exercise.
You can then have students analyze the peak discharge data. A plot of peak discharge vs time shows that flooding events are not predictable. For example, in the period from 1942-1957 there are 4 years when the peak flood exceeded 20,000 cfs. In the time period between 1958-1988, there was only 1 flood that (barely) exceeded 20,000 cfs. The final task is to plot the flood-frequency curve. The jing examples above show how to write equations in Excel.
Note that data on the flood-frequency curve are not nice and simple. You may want to modify this exercise by decreasing the discharge on the 5 largest floods so the data look nicer and are easier for the students to use. Alternatively, this is a good example of what real data look like. Based on the flood-frequency curve, students can determine the discharge associated with the 100-year flood (say 65,000cfs) and use the rating curve to determine the gage height of that flood (say 32 feet).
Regional Earthquake Databases
Some Online How-To Tutorials