A Little About This Tutorial

This tutorial is aimed to teach you how to design and create your own experiment. We will be using general information about basic experimental design and how to use CrowdExp(eriment), software that has been created by the Human Language Processing (HLP) lab in order to run experiments over the web.

The navigation bar at the top of this page is set up so that you can reference back to any section that you need. While this tutorial is meant to introduce the most important general concepts required to conduct experiments on language processing, the primary goal of the tutorial is to help you prepare a self-paced reading experiment (SPR) via CrowdExp. To help you with this, the tutorial goes through a specific example of an SPR experiment. The buttons on the bottom on the screen will indicate when you are moving on the a general or SPR section of the tutorial by blue and yellow buttons respectively. The SPR specific and general section pages will also be colored the respective yellow and blue colors.

After each section of the tutorial, there is an optional quiz. We highly recommend you take these quizzes. They will test your understanding of what you read and will guide you through the materials you do not understand.

For any items that are highlighted in red, you can hover over them and text will appear. If you click on the link it will take you to the term on the glossary page. Go ahead, try it. There are several other links. Purple superscript links will link to a paper [Paper]. A link that is [grey] will link to an external page.

After you have completed this tutorial, you will be ready to use what you have learned here and in class to then create your own experiment. This could be an idea you just want to test or a replication study based on readings you have done.

One final note before we continue. This tutorial, at times, goes into a lot of detail (for example, when we discuss possible problems with experiments). This is all meant to help you create the best experiment possible. But we also recognize that the first step is the hardest. So, when it all gets too much, feel free to simplify your experiment. Keeping it simple is often a good idea anyway =).