Early release of 3.5
This an early release of 3.5 to have all current updates and fixes in place, as there is potentially a major change coming in the next version, with introduction of artificial intelligence and GPT4 identification of coded text. (Still in development thanks to Kai Droge).
A binary (double-click to run) for Windows is available at the end of the release page for:
QualCoder-3.5.exe md5 checksum: 3faa37a813bf3944b2409c424077ee19
On first use of the exe, Windows will ask you to allow to run QualCoder. This is because it is from an unknown publisher. It costs a lot of money to get a trusted publisher certificate - so that will not be possible for the foreseeable future.
A binary (double click to run) is available for Ubuntu 22.04. The md5 checksum is: 4d754d388a3e921162fbfd56b3c84edc
For install from source code, download the zip file below and use the instructions on the main GitHub QualCoder page, or the wiki to install on your operating system.
A PDF manual of the 3.5 release can be purchased from:
https://www.buymeacoffee.com/ccbogelb/extras
Wiki: https://github.com/ccbogel/QualCoder/wiki
Report codes
Removed [VIEW] in the header of all the code segments in results. It has been replaced with character positions for coded text segments. e.g. [150-188]
Added context menu option - Change code to another code.
Replaced cogs button with Play button triangle. Repositioned and made larger button. Replace word 'search' with 'Run report'.
NEW Report - Code text exact matches
Added a new report that reports on exacted text matches in each file, where multiple codes have been applied to the exact same text. This report is perhaps particularly useful for those people who use codes as variables to perform analyses.
Code text
Export html now includes tooltips when hovering over coded sections. Some character set display issues for some characters, e.g. Euro symbol.
Export text is a new option. this exports a text file with coded segment marked with text in parentheses like this:
Regular text {{code name{{the tagged text portion}}code name}} regular text.
Stating character position: click in the displayed text and press ! to show the numeric position in the text. Press $ to shift all coded segments after the clicked position by up to to +/- 500 characters.
Charts - Word cloud
The word cloud has been changed with 'home grown' code. That is, I made my own word cloud code in python. This was mainly to prevent errors with packaging the PyPi Wordcloud into pyinstaller.
Error fixes
-
Zero division error in compare coders. Occurs under a specific circumstance - a code has been created but never assigned to data. Fixed.
-
Creating a Case Attribute from the Cases screen. An error occurs. A work around is to create the Attribute from the Manage Attributes screen. Fixed.
-
Wordcloud function does not use coded text within selected category. Fixed.
Problems in this release
The Code Audio/Video Segment right-click context menu does not work when using the Windows exe. (a brief flash then disappears). Download and run the source code instead, if this is a feature you need.
Issue with the python vlc module with Fedora 39. Unable to resolve this. So, for audio/video (loading files or coding) either use a different OS or another qualitative software.
Issue with installing pyqt6 on ChromeOS. Unable to resolve this.
Coding report - when selecting cases via attribute parameters, and multiple cases are assigned to parts of the same text document (e.g. from a survey import), the entire text document is used in the report.
Coding report - when Important only is checked, this is ignored in cases search report. Fixed in 3.6.