Using Data to Improve Learning and Improve Teaching

Using Data to Improve Learning and Improve Teaching Steven A Ackerman

Outline n n A bit about me A bit about you (well maybe not) The Literacies What I do… and why… • … the value of data… n Summary

A bit about who I am n n Faculty in Department of Atmospheric and Oceanic Sciences (Univ. Wisc. -Madison) Director of the Cooperative Institute for Meteorological Satellite Studies Research focuses on satellite observations of weather and climate. Teaching focuses on weather and remote sensing courses • • All levels Also outreach

Outline Again… n Information Literacy • K-12 • College level intro for non-science majors n Data Literacy • College level; meteorology majors n Statistical Literacy • College level intro for non-science majors

3 -literacies n Information literacy requires that an individual (student or teacher) recognize that information is needed and be able to locate and evaluate the needed information. They also have to effectively and efficiently access the information.

3 -literacies n To be data literate, students must be familiar with a variety of tools for accessing, manipulating and visualizing data.

3 -literacies n Statistical literacy requires that a student knows how to use statistics to develop evidence in an argument. They need to demonstrate an ability to assemble data into a summary form, and know what comparisons to make between data sets.

Statistical Literacy Data Literacy Information Literacy A Conceptual Model… Science Literacy

Information Literacy Data Literacy Statistical Literacy A Conceptual Model… Science Literacy

A Conceptual Model… Science Literacy Data Literacy Statistical Literacy Information Literacy

A Conceptual Model… Science Literacy Statistical Literacy Information Literacy Data Literacy

Information Literacy n How do we enable students to know they need data, and what type of data? • One way is through stories

Wisconsin Weather Stories Classroom-tested lessons that: • integrate earth science, social studies, language arts and math; • are adaptable for different grade levels, and • focus on Wisconsin weather, culture and narrative arts

Wisconsin Weather Stories AC Hi. I'm AC. I'm interviewing my mom, DC. So tell me a little bit about yourself DC Well, I was born and raised in Columbus, Ohio. I grew up in a family of eight people. Pretty much played music most of my life with my parents. Right now currently in school, going to be a pharmacist some day, working at the library school, plus I work at a Pharmacy and I sing in the band.

AC So have you ever been in a major weather storm? DC Actually I have. Back in 1995. I was performing on the Carnival Cruise Line in Florida playing on the M. S. Fantasy and we were about five weeks into the gig when we got caught in Hurricane Aaron. AC How bad was the weather? DC Pretty bad. The boat was pretty rocky. 144 -foot waves. Hit the side of the ship, broke a stabilizer. People were sick. We were confined to our rooms, weren't able to go aboard the ship. They didn't want us wandering the halls. Pretty much for about three days we weren't able to land, to dock, do anything. We circled around the eye of the hurricane 180 miles out and it was not a fun ride.

AC Did you get sick at all? DC I was very sick. I ended up having to go to the ship doctor. Got a shot. Knocked me out for two days. AC Where did this take place? Where exactly in the ocean were you?

A satellite view of Hurricane Aaron on August 3, 1995

Where were you on December 31, 1967? You probably wanted to be at Lambeau Field in Green Bay, Wisconsin. This was the site of the National Football League’s (NFL) Championship game between the Green Bay Packers and the Dallas Cowboys. Anyone alive will remember this game known as the historic “Ice Bowl”. Daily Maximum (Red) and Minimum (Blue) Temperature - December 1967 - Green Bay, WI

Applied to undergrad nonmajor n n n Write/Tell a personal weather story Grouped students by common story Developed questions • How much…, how many…, when does… • Why does… n n n Collected data Analyzed Presented data

Approaching a teaching/learning problem How do we learn?

Activity

Activity

Learning Diagram (Conceptual Model) n Input: n Reception: n Processing: n n New stimulus. Can be self generated or obtained otherwise Student’s filter to an input. Motivation, social and physical environment. Re-working the input by integrating reception with accumulated experiences and knowledge Practice: Use of re-worked input Feedback: Evaluation of re-worked input

Learning Diagram Bloom’s Taxonomy

So, how does learning diagram help? Bloom’s Taxonomy

Interactive web pages – contouring http: //profhorn. aos. wisc. edu/wxwise/Ackerman. Knox/

Formative Evaluation n n Students were surveyed regarding the applets using fieldtested questions from “The Flashlight Project” (Ehrmann 1997). 82% “believed” or “somewhat believed” that the applets helped them to master skills and concepts more quickly than with traditional methods. 75% “believed” or “somewhat believed” that the applets helped to develop their skills and under-standing better than through traditional reading and pencil-and-paper exercises. 71% “believed” or “somewhat believed” that the applets allowed them to learn more on their own, with less assistance from the TA.

Assessment of Learning Gains

What we learn on applets? Immediate Recall • TOTAL understanding F = 4. 380, p <. 038 • RH-temp relationship F = 49. 160, p <. 001 • RH as moisture indicator F = 7. 898, p <. 005 Delayed Recall • TOTAL understanding F = 3. 562, p <. 06 • RH as moisture indicator F = 5. 062, p <. 025

A teaching conceptual model…

Teaching as Research Learning Community Learning Through Diversity A teaching conceptual model… Good Teaching

A teaching conceptual model… Good Teaching Learning Community Teaching as Research Learning through Diversity

A teaching conceptual model… Good Teaching Learning Community Learning through Diversity Teaching as Research

Why these differences between classes… n n n The AOS 101 meets an extra time during the week and while students are exposed to new material during this 50 minute meeting, these extra meetings may help identify material in the lecture class that is not fully understood by the student Small group work projects in the discussion section may benefit the targeted group. Self selection, there may be a certain population that self selects out of the AOS 101 class.

How do we fix this? n The AOS 101 discussion section focuses more on weather analysis than in the lecture class. So, going to a hybrid class structure, part lecture, part on-line group work, will enable us to include more real-time weather map discussions and interpretation. Addresses: …AOS 101 meets an extra time during the week

How do we fix this? n Including some sort of social networking tools in the lecture class may help reduce the adverse outcomes of the targeted minority. Addresses: … Small group work projects in the discussion section

How do we fix this? n We are not sure that technologies will impact this explanation of the difference. Addresses: … Self selection

Why use technology?

Why use the Web specifically… n n n The Web can make teaching more effective. The Web can augment traditional resources. The Web can save time for students and teachers.

Why use the Web specifically… n n n The Web can make teaching less effective. The Web can distract traditional resources. The Web can cost time for students and teachers.

Why use the Web specifically… n Program with funding…

Why use the Web specifically… n Program with funding…

The issue: Can we simulate lab based activities/learning in a large lecture for the Introduction to Weather and Climate course. The course: Two major groups of students AOS 100 No lab section, three credits: n=319 AOS 101 Lab section, four credits: n=125 Mostly non-science majors Plurality are sophomores The solution? Will on-line activities work? Photo: Classroom Management

Forecast Contest

Forecast Contest

Forecast Contest

Forecast Contest

Forecast Contest

Forecast Contest

Multi-pronged plan to assess impact of activities on learning: 1. Regurgitation questions. Use on homework, then ask again on test Students retain relatively well. “What layer of atmosphere is closest to the earth's surface? ” 99. 2% of students got this right on homework. 81. 5% got this right on test 2. Application-based questions Make students think about what they know, get them to apply it to new situations. Students are not as good at these.

Testing Services provides an data on each student's response to every single multiple choice question on the exams. We can also have data on which students visited the optional content activities on the course website. So. . . we can write code to parse these two data files to compare performance on exams to completion of and performance on applets.

Example Activity: Precipitation Type http: //profhorn. meteor. wisc. edu/wxwise/precip-wb. html

Which of the following pairs contain two forms of precipitation that both require an inversion near the ground? a. b. c. d. e. snow and sleet snow and rain snow and freezing rain and sleet and snow 64. 9% of students looked at the applet at some point during the semester. Of those, 85. 3% got the answer right on 1 st exam 90. 6% got the answer right on final. Of the students who did not look at the applet: 75. 9% got the answer right on 1 st exam 84. 8% got the answer right on final. Sample size was 319. Difference significant at the 1% level

Build your own web-based learning activities with Reusable Content Objects (RCOs) + Chosen Tool (RCO) Data + XML + HTML Learning Activity

Tools for instructors to develop customized lessons Satellite Observations in Science Education (SOSE), with NASA support, has developed tools to demonstrate the quantitative information in satellite imagery.

Build your own web-based learning activities with Reusable Content Objects (RCOs) + Chosen Tool (RCO) Data + XML + HTML Learning Activity Reusable Evaluation Objects (REOs)

In summary n n n That’s a glimpse of what I do, why I do it and the frameworks I use. Data important for learning and improving teaching Data is good, but no data is better than bad data.