Friday, September 24, 2021

NarraFirma 1.5.0 is out

All righty then. I've been working on the latest pulse of NarraFirma development for about two months, and I've got some things to show you. I'll just show you a few of my favorite changes here. For more detail (than you could ever want), visit the NarraFirma blog.

Better surveys

The surveying part of NarraFirma has always been its weakest part. It's a little stronger now. Here are a few of the biggest changes.

You can now create surveys that accept write-in answers for every type of survey question. You can use these for "other" answers below a fixed list, or you can use them for follow-up questions (which are especially useful in pilot projects). Write-in answers are treated in the same way as answers to free-text questions (they have a special "graph type" in catalysis, and you can write observations based on them).

New write-in field example

You can now create multi-lingual surveys. Participants can choose a language at the start of the survey, and their choice is saved for graphing. You can enter and edit translations within NarraFirma or using CSV import.

New multi-language surveys

There are also several smaller surveying improvements, which you can read about in the blog post.

Better qualitative analysis

Another weak spot in NarraFirma has been its annotation system. I meant it to support qualitative analysis, but I didn't have time to finish it properly. So this time I went back and improved it. You can now create annotation questions whose answers emerge as you read through your stories. 

Improved annotation

This should improve the qualitative half of NarraFirma's mixed-methods support tremendously.

Better data lumping

This version of NarraFirma adds display lumping: treating similar answers to choice questions as if they were the same. Like story filtering, display lumping massages your data on its way to being graphed and tested.

New display lumping feature

You could always do data lumping in NarraFirma by exporting and re-importing your data, but that took time, and doing it over and over while you worked out your lumping schemes could be annoying. Now you can try out a lumping scheme in seconds by filling in a field. This improvement means that the integrity-checking and data-massaging step of the catalysis process should be over and done with much more quickly.


I also made many little usability improvements to parts of the application that were ugly or difficult to use. For example, here's a better way to choose and arrange questions on a story form:

New questions chooser

Please send bug reports!

If you are using NarraFirma and you find a bug, please visit the GitHub issues list and tell me what happened. You can also send me an email at

What's next

I have one more to-do item on my list for this pulse of work. It is to export catalysis reports in ODT (Open Document Text) format, which can be read by most word processors. This is another longstanding limitation of NarraFirma that I think I can remove. I hope to have that done within the next month or so.

If you would like to suggest any new features you think would make NarraFirma work better for you, please send me a note. I would love to hear what you would like to see.

No comments: