Design

Overview

---

This page guides you through the Design tab of the Experiment Builder. Check out the topics in the menu for walkthroughs of specific features.

The screenshot and description below shows New Experiment Builder. If you are still using Classic Experiment Builder, open the expandable section.

The Design Tab in Classic Experiment Builder

1. Experiment Name. Change the name and description of your experiment by clicking the pencil icon.
2. More Actions menu. Use this to view previous versions of your experiment.
   • When in Edit mode, use this menu to cancel your changes and revert to a previous version.
   • When not in Edit mode, use this menu to clone your experiment or send it to another user.
3. Commit/Edit button.
   • When in Edit mode, click Commit to commit changes to your experiment.
   • When not in Edit mode, click Edit to start making changes to your experiment.
4. Preview button. Click this to see how your experiment looks for a participant.
5. Update Nodes button. Click this to update nodes in your experiment to their latest versions.
6. Check For Errors button. Use this button to check for errors in your experiment tree.
7. Add Node button. Click this to add new content to your experiment. You can also use the blue plus buttons to add a node in a specific location in the experiment tree.
8. Design Space. Use the scroll wheel on your mouse to zoom in and out.

---

Adding Experiment Nodes

---

To find this guide useful, you should have added your Task and Questionnaire Nodes to an experiment tree. If you haven't done this yet, read more about how to build an experiment.

Once you have added your tasks and questionnaires to your experiment, you may want to consider adding Control Nodes to your experiment. These nodes allow you to manipulate the path of participants through your experiment. For example, you may want to randomise participants to two different conditions of your experiment - this can be achieved by adding a Randomiser Node.

Checkpoint Nodes are also incredibly useful nodes to add to your experiment. These allow you to keep track of Participant progress throughout your experiment. Checkpoint Nodes can be helpful in tracking participants returning to your study in longitudinal or multi-part studies, or helping judge when to manually include participants in your dataset.

You can find more information about the available Experiment Nodes in our Experiment Tooling Reference Guide.

---

Manipulations and Multiple Spreadsheets

---

If you are using multiple spreadsheets or manipulations within your tasks, these will all need to be set within the Task Node at the experiment level.

---

Multiple Spreadsheets

You can add as many spreadsheets to your task as you wish. You may want to use multiple spreadsheets to show participants a different set of stimuli, for example. You can find some more information about how to add multiple spreadsheets in your task in our How-To guide.

If you would like to set a specific spreadsheet to a specific instance of the task, you will need to click on the Task Node within the experiment tree, and choose the spreadsheet from the Spreadsheet dropdown.

Once you have selected which spreadsheet you would like to use for more than one node, you will be able to see the spreadsheet associated with each task at the experiment tree level.

If you do not set up the spreadsheet at the task node level in this way, the first spreadsheet in the list will be used by default.

If you have a larger number of spreadsheets, setting the spreadsheet for each task individually can lead to a very large experiment tree which can be difficult to manage. In this case, we would recommend using a Counterbalance Node to assign participants to different spreadsheets within the same task node.

---

Task Manipulations

If you have set up manipulations within your task, you will need to configure these at the experiment level before participants start your experiment.

To configure the manipulation at the experiment, click on the task node. When you click on the task node, you will be able to see the Task Manipulation in the window. This will include the name and description you have set for the Manipulation. Then, from the dropdown or in the free text field, you can set the value you would like to use for each task.

Once you have set Manipulations for more than one task node, you will be able to see which manipulation has been set for each task, as shown below.

If you do not set the manipulations in this way, the default value of the manipulation will be used for every participant.

---

Checking Experiment Flow

---

Before you get ready to launch your experiment, there are a few things we recommend doing to check your experiment works properly.

---

Preview your experiment

Previewing your experiment is the best way to see what your experiment will look like to your participants. You can preview your experiment as many times as you want to!

The screenshot below shows New Experiment Builder. If you are still using Classic Experiment Builder, open the expandable section.

The Preview button in Classic Experiment Builder

In New Experiment Builder, click the white and blue Preview button at the top right:

At the end of the preview, you will be given the opportunity to download the data from that preview. This is a great way to check your data, and make sure all the relevant data that you need for your analysis are being recorded.

---

Check your nodes are up-to-date

If your experiment preview doesn't look how you expect, a good first step is to check if any of the nodes in your experiment tree are out-of-date.

In New Experiment Builder, new task and questionnaire nodes have Auto-update enabled by default. This means that the node will automatically reflect the latest changes, including any changes that have not yet been committed.

If any of your nodes are set to 'Fixed version', you will need to either switch them to 'Auto-update' or take further steps to make sure they reflect your latest changes. For more guidance on this, see Updating Your Nodes.

---

Commit your experiment

When you are happy with how your experiment works, be sure to commit this version of the experiment. When you recruit participants, they will enter the latest committed version of the experiment.

The screenshot below shows New Experiment Builder. If you are still using Classic Experiment Builder, open the expandable section.

The Commit button in Classic Experiment Builder

In New Experiment Builder, click the green Commit button at the top right:

---