[resolved] 6.3 beta 1: Counting bugs in the lemma section of Concordance Tool

Mark Barnes

I was looking at hapax legomena in the NT, using the Concordance tool with the frequency set to 1. You can see from the first screenshot that Bible Text has 1,931 words, but the second screenshot (with Bible text added to the filter), says 1,934. I know that the sidebar shows the number of unique words, whilst the count at the top shows the total number of words, but if the frequency is set to 1, the two numbers should be identical.

Some suggested improvements to this tool:

• Ability to exclude parts of speech, not just include them.
• More drill down for parts of speech (to case, mood, etc.)
• Export to Word List.

Keep Smiling 4 Jesus :)

Mark Barnes said:

Some suggested improvements to this tool:

• Ability to exclude parts of speech, not just include them.
• More drill down for parts of speech (to case, mood, etc.)
• Export to Word List.

+1 [Y], especially like idea of Morph Code field filtering plus drill down.  Also would appreciate Louw-Nida field.

Keep Smiling [:)]

Jack Caviness

Mark Barnes said:

Some suggested improvements to this tool:

• Ability to exclude parts of speech, not just include them.
• More drill down for parts of speech (to case, mood, etc.)
• Export to Word List.

All good suggestions Mark.

Jacob Carpenter (Faithlife)

Great observation, Mark.

If you look at Words of Christ (instead of Bible Text) the difference is even more pronounced: 228 vs. 658 (*after* clicking Words of Christ). Here is an explanation of what the Frequency filter is doing and why you are seeing these confusing looking counts:

When you start out by looking at all of the Lemmas, and filter the Frequency to 1, you'll see 228 next to Words of Christ. This indicates that 228 of the headings in the included result set (right-side of Concordance) occur within the Words of Christ field.

Once you click Words of Christ, the entire result set from the resource is filtered down to only textual occurrences within the Words of Christ field. This throws out lots of occurrences of lemmas in the rest of the text. It turns out, throwing out those lemmas that occur outside of the Words of Christ field cause a lot more of the headings in the result set (right-side of Concordance) to have only 1 occurrence. So, in addition to the 228 that were counted before, we add 430 more lemmas. These are lemmas that are single-occurrence lemmas *within* the Words of Christ (but they aren't single-occurence, if you look at all of the lemmas).

The Frequency refinement is always being applied to the current set of results, not the book as a whole. Does that make sense?

Mark Barnes

https://community.logos.com/discussion/comment/717746#Comment_717746

Jacob Carpenter (Faithlife) said:

The Frequency refinement is always being applied to the current set of results, not the book as a whole. Does that make sense?

That does make sense, and I can see how my logic was faulty. Thanks.