5 Benefits of using nets when coding

Imagine you’re launching a new online clothes retailer. Suppose you decided to display all of your products in one giant list on the homepage. Your customers (and you!) are going to struggle with your site.

The human brain is limited in the number of thoughts or concepts it can hold in the conscious mind at once. So, to ease the mental load, we love to organize things into groups, categories and hierarchies. Think football leagues, musical genres, geographies, biological taxonomies...etc....

Unwieldy Codeframes

The same is true for the process of coding verbatim data. As you start coding, you’ll begin building up a list of “codes” (themes) which you use to code (tag) your verbatims. Before long, the list will grow and become unwieldy and hard to interpret.

In the world of coding, “nets” provide a way to tackle this problem. Because nets are so useful, a dedicated coding platform, like codeit, should provide baked-in support for them. This then provides a set of benefits, some of which are subtle and not immediately obvious.

The Benefits of Nets

1. Keep things organized
   

Firstly, nets give you the means to organize your codes into a manageable structure.

Think about the way you might organize files into folders and subfolders on your computer.

It’s not uncommon for a codeframe in codeit to contain hundreds or thousands of codes. You can prevent this from becoming overwhelming by organising the codes into nets and subnets - leading to a zen-like sense of orderliness!

2. Manage granularity
   

When analyzing coded survey data, granularity and nuance are very important.

For example, imagine you conduct a product test on a new variant of lemonade. If people say they don’t like it, you ask them why they don’t like it.

You need the data generated to be as actionable as possible. So, it’s not very useful if the data tells you, say, “55% of people mentioned the flavor”. You need to know specifically, what is it about the flavor they dislike. Is it too sweet, not sweet enough, too lemony, too sharp, too strong and so on...

So, when coding verbatim data, it’s incumbent on the coder to generate codes at a very granular level. This leads to the problem of a large and unwieldy codeframe.

Clearly, nets are the answer here. You can create an overall net for “Flavor” and sub-nets within that for “Sweetness”, “Sharpness”, “Strength” and so on…

At the final level, you can have the individual codes themselves.

So, from a granularity point of view, nets give you the best of both worlds – allowing you to code your data at a very granular, actionable level but in a way that is also manageable and well organized.



3. Represent a hierarchy
   

Nets and subnets offer a hierarchical structure within a codeframe. Sometimes this is exactly what you need if your codes do fall into a natural hierarchy. For example, you might be coding data about geographical areas, so you may want a structure that organizes the data into Continents | Countries | Regions | Cities.

Or, you may be coding something like brands and brand variants which also fit neatly into a hierarchical structure. In these cases, the hierarchical order of nets and subnets is perfect.

4. Fluid thought process
   

The process of coding is quite a fluid activity, often involving as much art as science.

So, it’s important that your coding tool, allows you to work fluidly with your data as you build up and refine the shape of your codeframe.

Nets are invaluable here, because they lighten the mental load during this creative process – allowing you to see commonalities and make refinements without becoming overwhelmed.

5. Getting the right counts
   

At the end of the coding process, nets are also as much about analysis as they are about the coding process itself. Once coding is complete, it is often invaluable to be able to analyze the results at all levels of the codeframe hierarchy. For example, it is useful to know how many people thought the lemonade was “Too sweet” but it is also useful to know how many people mentioned the flavor generally compared to the color or packaging. So, it is important that a coding tool can provide these hierarchical aggregations in the way that researchers require.

And now, could you automate that please?

As you can see, nets and codeframe design is a subtle and nuanced activity. This is something that people are very good at, but increasingly people expect to be able to automate with AI.

So, a coding tool like codeit also needs to offer tools that can automatically generate codeframes that include net structures as appropriate.

The themeit tool, built into codeit, provides the means to automatically extract themes from your data and organize them into nets and subnets if required.

Additionally, if you have a set of high-level areas you’re interested in, you can specify these up front and the themeit process will ensure that these are weaved in as nets into the autogenerated codeframe.

Of course, at the end of the process, codeit makes it easy for real people to adjust the autogenerated results and apply that all-important human judgment to refine and finesse the final output.

If you're looking for a verbatim coding tool, book a demo of codeit to discover how it can help you code your verbatims with precision using human-led AI.