The central purpose of PCAT is to create datasets for easy coding. Datasets are different than archives. A dataset is all or part of an archive that has been assembled for coding. You can create numerous datasets that are sub-parts of an overall archive coding project. A dataset is created by first defining a bucket. See the bucket entry for the process of creating and defining buckets. Once a bucket has been created with the files chosen to be included in the dataset, click 'Create Dataset from Bucket' as shown here.



A pop-up window will then appear, prompting “New Dataset Name” which in the example below is Investment.



At this point, you will have a series of options depending on the coding style you find best suited for the particular stage of the project or the aims underlying the coding effort. See the Hint below for a method to upload an entire code list with definitions and associated keystrokes). A few high level points are in order. First, you should try all the coding types and styles while you are learning the system so that you understand what is possible. Second, a feature like “user defined codes” is linked to early stage concept or code discovery, whereas mutually exclusive coding can be useful for data triage.



The check boxes that appear:

• Disable verification for user-defined and multiple coding: Determines whether or not a verification prompt will appear each time a coder codes a file in the dataset
• Allow user-defined codes: Coders are able to enter their own codes going through the data
• Allow coder to select multiple codes: Coders are able to use multiple codes for each file in the dataset
  

Once a name and these options have been specified, click 'Create Dataset' to be brought to the next screen. At this point you will be prompted to define the codes you want to be used in the dataset. Once you have added your codes, click 'Finished' to create the dataset. Note: Even if you have allowed user-defined coding, you are given the option of defining additional codes that will be available. The codes defined here are what will appear to the coders as they code the dataset.

HINT: The easiest way to create a code list is to prepare it as a plain text file and upload the file at the time a dataset is uploaded or created from a bucket. The file should contain three elements for each code and needs to be formatted as follows:

<code name1>|<definition1>|<keystroke1>
<code name2>|<definition2>|<keystroke2>
<code name3>|<definition3>|<keystroke3>
etc… etc…

This is a sample 4-code list below. In case you are wondering, the delimiter that looks like a straight up and down line is called a “Pipe” and you create a pipe by holding shift and pressing the backslash key just above the “Enter” key.

Not Useful|Comment is redundant or otherwise not useful|1
Borderline|Potentially a useful public comment|2
Important|This comment must be addressed|3
Attachment(s) Only|Comment only points to attachments|4

Why would I use this system? | Where do I get FDMS bulk downloads? | Does PCAT identify duplicates? | What is QDAP?

© 2009 - 2010 Qualitative Data Analysis Program (QDAP), in the University Center for Social and Urban Research, at the University of Pittsburgh, and QDAP-UMass, in the College of Social and Behavioral Sciences, at the University of Massachusetts Amherst. As of 2010, PCAT and this PCAT Help Wiki are maintained and improved by personnel from Texifter, LLC, which is a software start-up located in North Amherst & Springfield, MA and online at http://texifter.com.

Content on this website was made possible with the following grants from the National Science Foundation: III-0705566 “Collaborative Research III-COR: From a Pile of Documents to a Collection of Information: A Framework for Multi-Dimensional Text Analysis” and IIS-0429293 “Collaborative Research: Language Processing Technology for Electronic Rulemaking.” We are also grateful for financial support from the U.S. Environmental Protection Agency and the U.S. Fish & Wildlife Service. Any opinions, findings and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect those of the National Science Foundation.

Home | Accounts | Credentials | Peers | Projects | Upload | De-duplicate | Cluster | View | Browse | Search | Buckets | Datasets | Assign | Notifications | Toolbox | Code | Bookmarks | Validate | Report | FAQ | Contact