(Also available in Pyret)
Students come up with a research question and design a survey to gather data to answer it. They exchange surveys to get some hands-on practice with clean and dirty data and incorporate what they learn to polish their surveys. This project supports the learning goals of our lesson on Collecting Data.
Lesson Goals |
Students will be able to…
|
Student-facing Lesson Goals |
|
Materials |
|
Key Points for the Facilitator |
|
Preparation |
|
🔗Design a Survey flexible
Overview
It is exciting to collect data to answer the questions we are most curious about! But collecting high quality data is a tricky endeavor; many potential problems lurk beneath the surface. Good Data Scientists know how to minimize those issues.
Launch
In this project, your team will design a survey that yields "cleaner" data by using advanced survey tools to avoid many of the challenges faced by Data Scientists.
Complete Survey Brainstorming by choosing a topic and brainstorming:
-
a list of questions that would help you gather data about the project
-
the data types you would expect as responses for each of your questions
-
the displays you would be interested in seeing as part of your analysis
-
grouped samples that you’d want to explore separately (e.g. just 8th graders, just cyclists, etc).
Investigate
Now that you know what questions your team wants to ask, it’s time to actually create the survey!
With your team, create a first draft of the survey by opening a new google form and entering:
-
The title of the survey
-
A description of what the survey is for and how responses will be used
-
At least 8 good questions, copied from your Survey Brainstorming.
It’s time to read through and polish our work, just as we would after writing a draft of anything! In this case, our goal is to make improvements to the survey so that it yields the cleanest results possible.
With your team, focus on each of your survey questions, one by one:
-
Should the question be required?
-
Is the wording as clear as possible?
-
Would your question yield cleaner results if you specified the data type of the answer?
-
Would your questions yield cleaner results if you used multiple choice options?
-
If you’re using multiple choice, have you included all of the options that people would need to be able to reply to your question?
Now, let’s get some feedback. It’s time to hack each other’s surveys!
-
Exchange survey links with another group and test the limits of their survey by trying to enter as much "dirty" data as you can!
-
Turn to Survey Hacking and, for each dirty data example, write down what you submitted and what suggestion you would have to help the other group guard against it.
-
When you’re done, trade your Survey Hacking papers with the other group and review the data they entered in your form.
-
Revise your survey so that it’s ready to collect "cleaner" data from the general public! Be sure to incorporate what you learned from reflecting on the other team’s survey as well as the written feedback they gave you and address all instances of bad data entered in the form.
Synthesize
Review the Design a Survey Rubric and confirm that you’ve done a thorough job of addressing all of the project goals before submitting your finalized survey.
Some teachers have students share their finalized surveys with the community, import the resulting spreadsheets into CODAP, and analyze the data using the skills and concepts they have already learned.
These materials were developed partly through support of the National Science Foundation, (awards 1042210, 1535276, 1648684, 1738598, 2031479, and 1501927). Bootstrap by the Bootstrap Community is licensed under a Creative Commons 4.0 Unported License. This license does not grant permission to run training or professional development. Offering training or professional development with materials substantially derived from Bootstrap must be approved in writing by a Bootstrap Director. Permissions beyond the scope of this license, such as to run training, may be available by contacting contact@BootstrapWorld.org.