Using Real Data to Test Hypotheses by Graphing in Excel

Using Real Data to Test Hypotheses
by Graphing in Excel
Part of:
Inquiry Science with Dartmouth
Developed by: Samantha Roberts and Christian Durgin
Overview
This module provides data about a number of different climate measurements from around the globe.
Students will develop hypotheses about the data that they will test by graphing different variables (or
combinations of variables for more advanced students/classes) in Microsoft Excel.
Science Standards
Science and Engineering Practice 4:
-Analyzing and Interpreting Data. Represent data in tables and various graphical displays to
reveal patterns that indicate relationships.
Disciplinary Core Ideas:
-ESS2.D: Weather and Climate
-ESS3.D: Global Climate Change
Crosscutting Concepts:
-Graphs, charts, and images can be used to identify patterns in data.
-Relationships can be classified as causal or correlational, and correlation does not necessarily
imply causation.
-Cause and effect relationships may be used to predict phenomena in natural or designed systems.
-Scientific knowledge assumed an order and consistency in natural systems
-Science addresses questions about the natural and material world.
Focus Question
How can we use real data from around the world to ask questions about climate change, and how can
graphs help us understand complex data?
Objectives
Through this lesson, students will:
-Develop hypotheses that can be tested using the provided dataset.
-Create graphs that are appropriate to test these hypotheses and communicate conclusions about real data.
Background
Students should have a basic understanding of climate and the concept of climate change. Data set
includes temperature, rainfall, CO2 levels, and other climate related variables. They will use these data to
form hypotheses about changes in these variables over time that graphs can help them test.
Previous experience looking at spreadsheets and creating graphs is helpful, but the first part of the
exercise could be extended to an entire day and the teacher could walk them through these processes.
Materials
Computers with Microsoft Excel or other spreadsheet and graphing software.
Preparation
It is recommended that the teacher familiarize him/herself with the data set first. Many variables do not
have data points for every year included. This is intentional; many “real life” data sets scientists work
with have missing information. Microsoft Excel often fails at interpreting how you want a graph to look
when you select a data range and tell it to create a certain type of graph. Be prepared to demonstrate to
students how to specify X range and Y range data in the event that Excel automatically chooses
incorrectly. There is a provided Powerpoint presentation for instructors demonstrating how this can be
done in Microsoft Excel 2011 for Mac, the process is similar in other versions.
Procedure
1. Introduction: There is now an entire field of science called “Data mining,” in which researchers
create code to search through publicly available data. Just as students may post pictures and
information on Facebook or other social media sites, scientists post their data for other scientists to
see and use in their own studies. A powerpoint presentation is included with some slides that can
serve as an introduction. They include information about using publicly available data, why graphs
are important, and why we need to be careful about the graphs we produce as well as the ones other
people show us.
In addition, we are constantly presented with data in our daily lives, and often graphs are created to
intentionally mislead data consumers. There’s a great quote, “Statistics don’t lie, statisticians do.”
How do we interpret data we see presented in the media and the world around us? Scientists use
visualizations such as graphs to present complicated information, how do they choose what types of
graphs are best to communicate a concept?
2. Background: In the provided powerpoint presentation there are slides to discuss why visualization of
data is important, why students should both think critically about graphs that others are showing them
and how to present their own data so that others can understand the point they are trying to make.
Additionally, the variables in the dataset include temperature, rainfall, snow cover, solar flux,
atmospheric CO2, and sea level for a number of years. Not all variables have values for all dates, as is
typical of real datasets. A basic understanding of what each of these variables means and how it could
be affected by climate change would be helpful.
The first activity involves the entire class graphing the same variables in different ways to discuss
why certain charts are more appropriate than others for communicating information about the
variables. This can either be a guided activity by the teacher to demonstrate how to select data and
create graphs in Excel, or if the students already have experience with the program they can do it on
their own as a review.
3. Experiment: The worksheet walks students through examining the spreadsheet provided, then
graphing two variables (Year and Average Temperature in Burlington, VT) in several ways. They
then are asked which graph or chart is the most appropriate for communicating data about the
variables. After this introduction or review of using spreadsheets and creating graphs, students will
develop a hypothesis about one or more variables provided in the worksheet. They will then create a
graph that will help them test this hypothesis, discuss whether the data support or refute their
hypothesis, and explain why the type of graph they chose was the most appropriate.
4. Extend the Experiment: There are many different variables in the spreadsheet, if students complete
the worksheet early they can experiment with graphing different variables. Additionally the project
can be extended for students to do more research about the variables they chose and write up and/or
present an explanation of their findings.
5. Wrap-up: Have students discuss their graphs and how their hypotheses held up. Also possibly discuss
the challenges of using “real” data. Perhaps ask them what variables they would have liked to have in
the set and/or how they could collect relevant data on their own. Emphasize there are Terabytes of
publicly available data that are available to researchers studying any number of topics.
Assessment
Worksheet provided. Students will walk through one example with specific variables and explain which
graphs are relevant for communicating data. They will then develop a hypothesis they can test using other
variables in the dataset provided and graph these variables using an appropriate method. They are
required to explain their hypothesis, whether the data support or refute that hypothesis, and why they
chose a specific graph type to communicate their conclusion.
Extensions
Development of a hypothesis may be done in pairs or assigned as homework. If done in pairs, it is
recommended that each student still be required to produce a graph and independently interpret the
results. If students are experienced with Microsoft excel, the entire section on developing a hypothesis
and creating an appropriate graph may be assigned as homework, and this could be incorporated into a
larger project involving a more extensive written or presented description of the topic, hypothesis, and
conclusions.