Introduction

Before you start the themeit theme extraction process, there are a number of settings you can adjust to influence the behaviour of the process. These settings influence the themes that are extracted, the shape of codeframe that is created and the autocoding that is applied.


Settings

Within the "Settings" tab you can adjust the following options:


OptionsDescription
Create NetsIf "Yes" is selected, the theme extraction process will attempt to group extracted themes into nets.
Max Themes
Set a number (between 10 and 300) to limit the maximum number of themes the process should extract. The actual number of themes returned may be lower than this maximum depending on the content of your data.
If this setting is left blank, themeit will use a default of 100 themes as a maximum.
Autocoding Mode
Controls whether the autocoding applied by the AI should Minimise Gaps or Maximise Coding.
See here for an explanation of these two coding modes.



Advanced

Within the "Advanced" tab you can adjust the following options:


OptionsDescription
Create Codeframe Only
If toggled, themeit will only generate the codeframe and will not attempt to autocode the data.
By default, this setting is off. A codeframe is automatically generated and the data autocoded by themeit.
Theme Language
Specify the language for the autogenerated themes. By default, the language of the project is selected.
For multilingual data, the data can still be autocoded in-language even if the auto-generated codeframe is in a different language. Learn more about multilingual AI here.
Question TextEnter the survey question used to collect the verbatims being analysed.
Adding the survey question provides themeit with additional contextual information that it can use to generate a more appropriate and accurate codeframe.
Required NetsEnter a list of nets (theme categories) that you're interested in or expecting to find within your data. This provides the system with key information about your analysis objectives which it wouldn't otherwise have. As a result it can help generate a more appropriate and accurate codeframe.
The resulting codeframe will include all of the specified required nets, even if there are no themes discovered that fall into these nets.
Required ThemesEnter a list of themes that you're interested in or expecting to find within your data. This provides the system with key information about your analysis objectives which it wouldn't otherwise have. As a result it can help generate a more appropriate and accurate codeframe.
The resulting codeframe will include all of the specified required themes, even if there are no verbatims found to contain these themes.