You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
* Coding comparison - for text files (compares 2 coders)
9
-
* Coding comparison by file - for text files (compares 2 coders)
10
-
* Code counts by file/case
11
-
* Code frequencies (totals of applied codes)
12
-
* File summary (details of a file and its coded segments)
13
-
* Code summary (details of a code and associated files)
14
-
* Code relations - for text files (shows proximity, exact match, overlaps, distance between selected codes)
5
+
# Analysis menu options
6
+
7
+
* Coding retrieval (the main reporting function)
15
8
* Code Co-occurrence - for text files (Shows counts where a code abuts/overlaps/exact-matches another code)
9
+
* Code relations - for text files (shows proximity, exact match, overlaps, distance between selected codes)
16
10
* Code text exact matches - for text files (Shows counts of exact matches between codes)
17
11
* Codes by text segments (Shows table of file names (rows), coded text, and codes assigned - denoted by 1).
18
-
* View graph (Create or show a graph (map) of objects and their linkages)
19
-
* Charts (Display statistical and other charts in a web browser)
20
-
* Database Queries (Database querying, including default queries)
12
+
* View graph (Create or show a graph (map) of objects and their linkages) See separate page on this.
13
+
* AI assisted analysis - by topic, by text, by code.
21
14
22
-
# Coding Reports
15
+
# Code Retrieval
23
16
24
17
This report gives a list, or matrix, of coded data based on your selections. One or more codes need to be selected. If a category is selected, all codes in that category are also selected. Multiple categories can be selected using the mouse and the ‘ctrl’ button.
25
18
@@ -91,41 +84,18 @@ Reports can be exported as html, odt, txt, csv, xlsx, (in 3.8+) iramuteq.
This option shows the overall similarities and differences between two coders across all text files. Select two coders and click the run comparisons button. The statistics created here are only for coded text, not coded image nor media files.
97
-
98
-
For each code:
99
-
100
-
Agreement % shows agreement for a combination of coded and non-coded text characters.
101
-
102
-
A and B % shows agreement for only the coded text characters divided by the total characters in the text.
103
-
104
-
Not A and Not B % shows the total of the uncoded text divided by the total characters in the text.
105
-
106
-
Disagree % shows the percentage of all the coded and non-coded text that did not match between coders. It is the same as 100 – the Agree %.
107
-
108
-
Agree coded only % shows the percentage of all the dual-coded text divided by single plus dual-coded text.
109
-
110
-
Cohen’s Kappa is calculated based on the information in [Wikipédia](https://en.wikipedia.org/wiki/Cohen%27s_kappa).
Buttons to run the comparisons, to clear the coders selected, to export to Excel, and to open a statistics help window.
115
-
116
-
# Coding Comparison by file Report
117
-
118
-
This option shows the overall similarities and differences between two coders in ONE selected file. Select one file, one code, and two coders. Then click the run comparisons button. The statistics and comparisons for audio/video coded segments are in development.
119
-
Overall agreement, disagreement, and Cohen's Kappa are calculated for text and images. An image file is displayed in a separate window, showing the two coders' selections in yellow or blue-boxed areas. Green text indicates where both coders coded the same section of text. Mouse hover provides additional information.
This report shows a table of code names along the x and y axes. The table cells show the count of coding co-occurrences. This is where two codes overlap or are directly abutting. Click on the cell to show the text coding details.
122
90
91
+
Options include: selecting specific text files, selecting files by attributes. Selecting specific codes, or selecting categories. There is an export-to-Excel option.
You can right-click on a cell containing co-occurrence coded text to make a new merged code that combines the two overlapping codes.
129
99
130
100
# Code relations Report
131
101
@@ -134,7 +104,6 @@ Select two or more codes, and press the Calculate button.
134
104
his report may be useful to show proximity or overlaps between codes. Note, other reports such as Code Co-occurrence and Code Text Exact Matches also do this. This report gives a lot of fine details, such as the coding character positions for each code.
135
105
A csv file can be exported.
136
106
137
-
138
107
The relations are:
139
108
140
109
* Proximity – Two codes do not overlap. The distance in number of characters is shown.
@@ -150,36 +119,6 @@ Summary statistics and box plots can be produced.
This report shows a table of code names along the x and y axes. The table cells show the count of coding co-occurrences. This is where two codes overlap or are directly abutting. Click on the cell to show the text coding details.
157
-
158
-
Options include: selecting specific text files, selecting files by attributes. Selecting specific codes, or selecting categories. There is an export-to-Excel option.
In version 3.8 plus, you can right-click on a cell containing co-occurrence coded text to make a new merged code that combines the two overlapping codes.
166
-
167
-
# File Summary Report
168
-
169
-
Click on a file to get a summary of that file. It describes media metadata, characters and word frequency. Attributes. It summarises the count of codes for that file. The text codings summarize the average text length. For image codings it summarises the average pixel area. For audio/video codings it summarises the average segment length.
170
-
171
-
The top 100 words have stopwords applied - so common filler words such as 'a ', 'the', etc. are excluded. See the list in the _Examples folder_. You can override this list by providing your own stopwords text file and placing this in the _.qualcoder folder_.
Click on a code to get a summary of that code. The coders who have used the code. The number of codings in text, image and audio/video media. The text codings summarize the average text length and the most common words. For image codings it summarises the average pixel area. For audio/video codings it summarises the average segment length.
178
-
179
-
The top 100 words have stopwords applied - so common filler words such as 'a ', 'the', etc. are excluded. See the list in the _Examples folder_. You can override this list by providing your own stopwords text file and placing this in the _.qualcoder folder_.
This report details all the exact matching coded text segments, where different codes are applied to the exact same text. Select text documents and two or more codes.
@@ -195,83 +134,14 @@ An Excel file can be exported.
Various chart types and filter options are available.
202
-
203
-
Pie, Bar, Sunburst and Treemap charts can be created.
204
-
205
-
Charts, except for the WordCloud, are displayed in the default web browser. If the browser is not displaying your chart, first change your default web browser to another recent browser, e.g. Chrome or Edge or Safari. If you have javascript switched off the charts will not display. Checking the _Export html file_ box will make all charts be exported to the Default export folder as an html file.
206
-
207
-
Heatmaps of files or cases against codes can be created. These are limited to 40 rows and columns for quicker rendering and better visibility.
A word cloud is available with different options for background colour, foreground colour or colour range, rectangle size, number of words (n-grams) from 1 to 4, and maximum number of words to include. The maximum words is determined by the most frequent to the least frequent and changing this option will change how crowded or sparse the word cloud appears.
There are a list of stop words that are used to exclude common words from the word cloud display (e.g. an, a, the). This can be overridden by creating your own stopwords.txt file with one word on each line. This file should be put inside the .qualcoder folder. e.g. on Windows:
241
-
242
-
"C:\Users\yourname\.qualcoder"
243
-
244
-
The stopwords currently used are in stopwords.txt in the Examples folder. Alternatively, in 3.8+, you can use the button in the Wordcloud section to temporarily link to another stopwords file of your choosing.
This window contains three panes. The top pane is where SQL statements are entered and the bottom pane contains the results of queries. The left pane contains tables and field names. Double-clicking on a field name adds it to the SQL statement. Results can be exported to a delimited file. If you are not familiar with SQL take care as you will be able to update and delete the data as well as select data. Note: Some Unicode symbols are not converted to plain text and are ignored.
253
-
254
-
There are several prepared statements listed:
255
-
256
-
* Case text
257
-
* Coded text using a case attribute selection (v3.6+)
258
-
* Codes fileid and coded text
259
-
* Get coding table - an implementation of an RQDA function that returns the code name, table, text length and text start and end positions.
260
-
* Coded text with each case
261
-
262
-
Most table fields are text. The following fields are integer: anid, avid, attrid, caseid,catid, cid, fid, id, imid, jid, pos0,pos1, x1, y1, width, height.
263
-
264
-
Right-click in the results table for filtering options.
265
-
266
-
Right-clicking in the SQL window gives you various options such as select all, copy, and paste.
267
-
268
-
At the bottom of the left-hand side list, you can see: stored sql and Default queries
139
+
See page ...
269
140
270
-
Stored sql are your own created and stored sql statements. To store a sql statement, right-click on the sql and choose 'save query'.
141
+
# AI assisted analysis
271
142
272
-
Default queries, such as the one shown in the image, are ready-made SQL queries to perform additional extracts from your data. They are commented '-- a comment' to describe what each statement does.
0 commit comments