5/4/2015 While the Dedoose Descriptor functionality can be very useful and a valuable component of your project, it can also be confusing…hence the number of blogs we post on the topic. Today, we decided to provide a short list of the top 3 best-practices that you should consider to help make the most of your descriptors.

1. Have an ID Descriptor field: Having a unique identifier for each of your research participants (or settings, dyads …whatever your level of analysis) is an important part of keeping your project data organized. In addition to having a way to reference a descriptor in particular, an ID field is an excellent counter measure to any possible problems you might experience with your descriptors. That is, we always suggest some clear identification of descriptors and media titles, so it is a straightforward matter to match them up. As an example, for descriptors about your interviewees, you may have a number of fields including gender, age, and ethnicity. What happens when you end up with 2 people who are male, age 30, and Asian? That’s where the ID field comes in handy. We cannot stress enough that this field needs to be UNIQUE for each descriptor. Also keep in mind that this ID field does not have to be a number. It could also be their first name, a pseudonym, etc. Just remember, UNIQUE, UNIQUE, UNIQUE!

2. Have your descriptor fields be option list types as often as possible: Out of all the field types in Dedoose, we strongly encourage using the ‘option list’ because it is the field type that best lends itself to analysis in the Dedoose charts and tables.

Text: Often, you may have some text data, but there will be a limited number for each unique value. So, rather than using a text type field for fields like state, city, institution, we’d recommend you use an option list type. This would set things up so that when you are analyzing your data and you hit any chart that has descriptors included, these options will show up and can be analyzed independently.

Numbers: Without changing anything, a continuous numerical type field will show up as a set of ranges in many Dedoose charts. So, if you are trying to analyze your responses by the groupings that are created with the algorithm we have dialed in, then you are set. However, if you do want to control how the ranges are defined it would be best to have the field be defined as an option list with your own pre-defined groupings. Further, if you want to independently analyze the responses of each of those unique numbers (say if they are ‘ages’) you would need to made an option list field with all the individual ages. Replacing those numerical responses with option lists will allow you to independently analyze each of the ‘options’ with full control over how things break out. Your decisions about how to group ranges of continuous data when charting in Dedoose can be determined by your research questions and are sure to be much smarter than the arbitrary algorithm’s attempt to assist.

3. Using Dynamic Descriptors: Dynamic Descriptors are a cool feature designed to help track change over time within your coding activity. An example of this would be if you are interviewing the same people each year for 5 years total. Of course you want to be able to show that it is the same person just at different times in their life and it doesn’t make sense to create a new descriptor for each time point if it is the same person. Here, you would want to create an option list dynamic field that has values for the time: Ex. Time 1, Time 2, etc. When you create the descriptor, don’t worry if you do not see this field initially. When you actually attach that descriptor to a media file, you will be prompted to choose the period of time that particular media file is from. Dynamic Descriptors should only be used on data that will actually change over time, so be careful to not overuse it. Setting, for example, the gender field as a dynamic field usually doesn’t make sense unless it is possible in your population to change gender during the course of your study…. generally not a characteristic changing over time. We hope that these little tips will help you dig further into your data in new ways.

So dive in, mess around, and have fun Dedoosing!