Major Discrepancies in Search Results

Page 1 of 1 (18 items)
This post has 17 Replies | 1 Follower

Posts 63
Apexite | Forum Activity | Posted: Wed, Mar 27 2013 1:11 PM

I just received the recommendation from Tech Support to post this issue here, with the rep saying most of the "power users" are on here and able to help.  Here is the best summary I can come up with:

1)  In Gen 2:16, looking at the Hebrew word for “commanded”: (H6680)....

          In ESL, entry for H6680 says 494 occurrences in AV

          BUT using Bible Search, with AV, a search for <H6680> says 499 occurrences

2)  In studying the Greek verb, merimnao: (G3309).....

          In ESL, entry for G3309 says 19 occurrences in AV

          BUT using Bible Search, with AV, searching <G3309> says 24 results in 17 verses

          Using Bible Search, with NKJV,  searching <G3309> says 19 results in 17 verses

          Using Bible Search, with NASB95, searching <G3309> says 39 results in 17 verses

               BUT manually counting the occurrences in NASB95 search results list shows 19, not 39

4)  In studying the related Greek word, merizo (G3307)....

          In ESL, entry for G3307 says 14 occurrences in AV

          BUT using Bible Search, with AV, searching <G3307> says 16 results in 13 verses

          Using Bible Search, with NKJV, searching <G3307> says 14 results in 13 verses

          AND using Bible Search, with NASB, searching <G3307> says 28 results in 13 verses

I am getting similar discrepancies in just about every search for an original language word using Strong's #s.  Either there is something I'm not understanding about the Bible Search results concept, or somewhere in the method of counting the occurrences of a given original language word Logos is having "math issues".  Any help you "power users" can provide is appreciated.   

Posts 27681
Forum MVP
JT (alabama24) | Forum Activity | Replied: Wed, Mar 27 2013 1:26 PM

My first thought is that some of the "discrepancies" may be textual issues... The KJV doesn't use the same "original" sources as many of the newer translations. Since there are <minor> differences between the texts the translators used, there will be differences in counts here and there.

EDIT: I see now that this isn't likely the issue... I thought you were comparing the English Standard Version & NASB with the KJV... I think I remember there being an issue with the NASB "doubling" the results for some reason... Hopefully someone else will be able to shed light.

OSX & iOS | Logs |  Install

Posts 26529
Forum MVP
MJ. Smith | Forum Activity | Replied: Wed, Mar 27 2013 10:56 PM

The NASB95 appears to have double counting when you look at the detail. The AV / NKJV version differ due to translation quirks - see Matthew 6:34 as an example.

Remember that you are not counting the occurrences of the Greek word; you are counting the number of times the Greek word appears as a tag to the English.

Orthodox Bishop Hilarion Alfeyev: "To be a theologian means to have experience of a personal encounter with God through prayer and worship."

Posts 24930
Forum MVP
Dave Hooton | Forum Activity | Replied: Thu, Mar 28 2013 5:13 AM

Apexite:

1)  In Gen 2:16, looking at the Hebrew word for “commanded”: (H6680)....

          In ESL, entry for H6680 says 494 occurrences in AV

          BUT using Bible Search, with AV, a search for <H6680> says 499 occurrences

How can you trust ESL when it says it is translated as "command" 514 times!?

H6680 occurs 499 in Logos' 1769 AV, different to the version Strong was using.

It occurs 493 x in Logos' KJV 1900 version (the interlinear) and 493 in AV 1873 (Cambridge Paragraph Bible)

Apexite:

2)  In studying the Greek verb, merimnao: (G3309).....

          In ESL, entry for G3309 says 19 occurrences in AV

          BUT using Bible Search, with AV, searching <G3309> says 24 results in 17 verses

          Using Bible Search, with NKJV,  searching <G3309> says 19 results in 17 verses

          Using Bible Search, with NASB95, searching <G3309> says 39 results in 17 verses

               BUT manually counting the occurrences in NASB95 search results list shows 19, not 39

Here, the 19 in ESL is consistent with the rest of the entry.

The NASB always gives an exaggerated count of around 2x. The manual count is 21, though.

AV 1873 has 19, KJV 1900 has 19

Apexite:

4)  In studying the related Greek word, merizo (G3307)....

          In ESL, entry for G3307 says 14 occurrences in AV

          BUT using Bible Search, with AV, searching <G3307> says 16 results in 13 verses

          Using Bible Search, with NKJV, searching <G3307> says 14 results in 13 verses

          AND using Bible Search, with NASB, searching <G3307> says 28 results in 13 verses

NASB says 28. Manual count is 14.

AV 1873 has 14, KJV 1900 has 14.

Apexite:
Either there is something I'm not understanding about the Bible Search results concept, or somewhere in the method of counting the occurrences of a given original language word Logos is having "math issues".  

Different bible versions will give different results, apparently even between KJV editions.

But the NASB has a bug inherited from Libronix (Logos 3) because it carries its own Strong's numbers in addition to the Strong's numbers from the interlinear data. Hence the double counting. You can see this with the AV (1769) as it has NO interlinear data, but clearly has Strong's numbers.

However, if you look closely you see that consecutive words are highlighted in the AV and if you right click each word you see it is assigned the SAME Strong's number e.g. Lk 12:13 "that he divide" where each word has G3307. yet it is only counted as 1 result.

Now look at 1 Cor 1:13 & 1 Cor 7:34 where both count as 2 in the AV and only 1 in AV 1873, KJV 1900. So this is why AV has 16 results vs 14. This also accounts for the discrepancy with G3309 (look at the results in Aligned view). This will also account for the 499 vs 493 results for H6680.

So we must blame the legacy versions of NASB and AV for discrepancies with Strong's number searches.

Dave
===

Windows & Android

Posts 63
Apexite | Forum Activity | Replied: Thu, Mar 28 2013 7:09 AM

Thank you, Mj/Dave, for the explanation.  

At this point, this is my take-away:  in moving forward, when searching for occurrences of original language words, I'm not going to consider the reported "search results" from NASB as trustworthy, in light of the inherited double-counting bug dating back to v3.  I'll also not consider the reported "search results" in any TR-based translation as trustworthy, since multiple variations of TR are used in Logos (depending on the dating of a translation), and all of them are going to have the count thrown off a tad any time an original word is tied to multiple English words, regardless of whether they are non-consecutive words or even consecutive.  I'll also not consider the ESL as trustworthy when pointing out the number of times a Greek/Hebrew is translated a certain way.

So if the legacy versions of NASB and AV are not to be trusted with Strong's number searches, and the ESL is untrustworthy, what e-tool can I trust?

Previously, I've always thought that the TR was consistent as the underlying text for all KJV's, AV, and NKJV - and a count of G/H words in any of those translations would yield the same accurate count - but now that appears to not be the case.

Even when, earlier this morning, I looked at the RI for both the KJV and the NKJV, when looking at Luke 12:25, the Strong's # that underlies the English "taking thought" in the KJV and "worrying" in the NKJV shows BOTH G3009 AND G3008!  One is a noun, the other a verb.  Is that true - that in the TR both the noun form and the verb form are present in that verse?  Or is that just coincidentally a double-entry of related Greek words in how Logos tags the original language, and it happened so close to the words I was trying to get an accurate count of in search results?

I understand that any time a human puts together a tool, a human error or two can be expected.  But overall, it appears I should never use the "search results" from a search window as a trustworthy count for actual occurrences of an original language word.  What should I use?  What e-tool can I trust?  

Posts 1838
David Thomas | Forum Activity | Replied: Thu, Mar 28 2013 9:45 AM

Apexite:

What should I use?  What e-tool can I trust?  

    Is there any reason why you are searching by Strong's # rather than just selecting a lemma and search this resource to get a count?

Making Disciples!  Logos Ecosystem = Logos8 on Microsoft Surface Pro 4 (Win10), Android app on tablet, FSB on iPhone, [deprecated] Windows App, Proclaim, Faithlife.com, FaithlifeTV via Connect subscription.

Posts 63
Apexite | Forum Activity | Replied: Thu, Mar 28 2013 10:31 AM

Two ways to get at the same info.  The method you described does a search for the Strong's #, only through a different path.  I like to keep multiple Search panels open - already set up for 1 and 2 Greek words, as well as 1 and 2 Hebrew word - then all I have to do is go to the appropriate search panel and input the Strong's #(s) - and with that number inserted between the <...>, it's easy to change the range and/or translation per my current interest, if needed.  It's not so much the method of how I get at the information - it's more about the actual information I get being accurate, or not.

Posts 21920
Forum MVP
Graham Criddle | Forum Activity | Replied: Thu, Mar 28 2013 10:55 AM

Apexite:
The method you described does a search for the Strong's #, only through a different path.

Searching for a lemma doesn't use Strong's numbers. Rather it uses the morphological tagging within the resources themselves.

Posts 63
Apexite | Forum Activity | Replied: Thu, Mar 28 2013 11:27 AM

When I use the method suggested, a Search panel opens up and in the Search criteria is this:  <GreekStrongs = Strong’s Greek #3309>.  Doesn't that indicate that Logos completed a search for Strong's # G3309 in the Greek text underlying that particular translation?  That's the same Search criteria I put in, only through a different method.

[EDIT:  I just went back and re-examined the Lemma search, and with a little more attention to detail the search criteria window indeed indicates a non-Strong's string:  <Lemma = lbs/el/μεριμνάω>.  So with that in mind, let me re-do the searching I had previously performed, only this time using the suggested method, and I'll let you know what the results are.  THANKS]

Posts 26529
Forum MVP
MJ. Smith | Forum Activity | Replied: Thu, Mar 28 2013 2:08 PM

Apexite:
At this point, this is my take-away

I'd word the take away this way. If you want accurate counts:

  1. always search a Bible in the same language as your search term - that eliminates differences in translation
  2. search for the actual lemma not some tagging scheme (disclosure:I hate the concept behind Strong's)
  3. examine your results to be sure that you've gotten what you expect re: word forms - morphology and etymology

Orthodox Bishop Hilarion Alfeyev: "To be a theologian means to have experience of a personal encounter with God through prayer and worship."

Posts 2964
tom | Forum Activity | Replied: Thu, Mar 28 2013 3:48 PM

MJ. Smith:

I'd word the take away this way. If you want accurate counts:

  1. always search a Bible in the same language as your search term - that eliminates differences in translation
  2. search for the actual lemma not some tagging scheme (disclosure:I hate the concept behind Strong's)
  3. examine your results to be sure that you've gotten what you expect re: word forms - morphology and etymology

Yes

Posts 63
Apexite | Forum Activity | Replied: Thu, Mar 28 2013 5:18 PM

I just finished doing the 3 sets of searches I originally posted data on, only this time using the method suggested [select a lemma and "search this resource"], and all came out with matching numbers!!  For that I'm thrilled (thanks for the suggestion, David).  In the future, I'll utilize that more original-language-oriented method, which follows MJ's take-aways 1 and 2.  It was a revelation to me that the lemma doesn't use the Strong's numbers (thanks for the heads-up, Graham).  

As an associated side note, concerning the use of the Enhanced Strong's Lexicon, I am wide open to anyone's insight as to it's qualitative usefulness compared to similar Logos tools.  Personally, I LOVE the concept behind Strong's.  I don't read Greek or Hebrew, and have no aspiration to do so.  I "grew up" using KJV and Strong's (hard copy).  To have the "enhanced" version that not only gives basic definitions, but also includes prefixes+root words (with Strong's #s), total occurrences, as well as individual translation counts - that has always been a HUGE blessing to me as a non-scholar who works strictly with transliteration.  But I'm wide open to change from ESL to another single tool (or integrated tool set) that will yield the same information (definitions/root words/occurrences/translation counts) in an accurate, easy-to-get-to process.

Posts 24930
Forum MVP
Dave Hooton | Forum Activity | Replied: Thu, Mar 28 2013 8:22 PM

Apexite:
But I'm wide open to change from ESL to another single tool (or integrated tool set) that will yield the same information (definitions/root words/occurrences/translation counts) in an accurate, easy-to-get-to process.

Because a "critical" Greek bible is used by modern translations as against the TR Greek used by the KJV the ESL has limited usefulness for occurrences/translation counts. But you can get occurrences via the lemma search of an original language bible (a reverse interlinear is not as accurate).

A reverse interlinear bible makes translation counts possible  e.g. lemma:λόγος ANDEQUALS word => find where λόγος is translated as "word". You can also find where λόγος is not translated as "word" e.g. lemma:λόγος NOTEQUALS word. Now, if you right click "word" in an interlinear (and some Greek bibles), you will finds its Root and can conduct a search on that in the same way you have done for the lemma.

Have a look at Original Language Lexicons on the upgrade page, The 3 "Dictionary of Biblical Languages" (DBL) available to the Starter package are very useful; they provide alternate meanings and can link to ESL via Strong's numbers or the lemma. An analytical lexicon provides more information e.g. Lexham Analytical Lexicon to the GNT, but not much definition; use them as a supplement to the DBL. A theological lexicon concentrates on words of significance and does not cover all words of the NT or OT.

Dave
===

Windows & Android

Posts 15805
Forum MVP
Keep Smiling 4 Jesus :) | Forum Activity | Replied: Thu, Mar 28 2013 9:13 PM

Apexite:
In studying the Greek verb, merimnao: (G3309).....

Logos can use transliteration for morph search entry:

g:merimnao

then choose lemma from drop down list.

Opened first search result, then right clicked on word in AV 1873, followed by clicking root, then morph search, which found more verses.

Apexite:
I don't read Greek or Hebrew, and have no aspiration to do so.

Caution: Greek is not English and both are not Hebrew.  The Greek verbal system is much more expressive and nuanced than English.  One option is learning about differences of verbal expression and use Logos Visual Filters to highlight morphological tagging.  The screen shot shows μεριμνάω is used in the imperative (command) mood in Matthew 6:25 and 6:28 while Matthew 6:27 is a participle that has noun and verb aspects (similar to a gerund in English).  Matthew 6:34 has two verbal uses: one is an aorist tense and the other is a future tense.  One "take" away is the lack of Enhanced Strong's Lexicon for Greek verbal range of expression.

3309 μεριμνάω [merimnao /mer·im·nah·o/] v. From 3308; TDNT 4:589; TDNTA 584; GK 3534; 19 occurrences; AV translates as “take thought” 11 times, “care” five times, “be careful” twice, and “have care” once. 1 to be anxious. 1A to be troubled with cares. 2 to care for, look out for (a thing). 2A to seek to promote one’s interests. 2B caring or providing for.

Strong, J. (2001).Enhanced Strong’s Lexicon. Bellingham, WA: Logos Bible Software.

For original language insights, Logos offers => http://www.logos.com/product/8831/learn-to-do-word-studies-with-logos-bible-software and => http://www.logos.com/product/5876/learn-to-use-biblical-greek-and-hebrew-with-logos-bible-software

For visual filters, wiki has examples => http://wiki.logos.com/Extended_Tips_for_Highlighting_and_Visual_Filters#Examples_of_visual_filters

Keep Smiling Smile

Posts 687
Jon | Forum Activity | Replied: Fri, Mar 29 2013 2:35 AM

Apexite:
As an associated side note, concerning the use of the Enhanced Strong's Lexicon, I am wide open to anyone's insight as to it's qualitative usefulness compared to similar Logos tools.  Personally, I LOVE the concept behind Strong's.  I don't read Greek or Hebrew, and have no aspiration to do so.  I "grew up" using KJV and Strong's (hard copy).  To have the "enhanced" version that not only gives basic definitions, but also includes prefixes+root words (with Strong's #s), total occurrences, as well as individual translation counts - that has always been a HUGE blessing to me as a non-scholar who works strictly with transliteration.  But I'm wide open to change from ESL to another single tool (or integrated tool set) that will yield the same information (definitions/root words/occurrences/translation counts) in an accurate, easy-to-get-to process.

I think the best tool for all that is Logos' Bible Word Study guide In particular, the Translation ring shows you how the word is translated for any English version you have reverse-interlinears for (hover over the bar at the top of the Translation ring section to see a Settings menu pop-up and choose a translation). You can then click on any part of the translation ring to give you numbers and show all the places where that word is translated that way.

Posts 63
Apexite | Forum Activity | Replied: Fri, Mar 29 2013 6:58 AM

Dave, KS4J, Jon.....THANK YOU for the extremely helpful information.  I will be incorporating the new practices into my study habits, based on what each of you pointed out.  The "put off...put on" process of change is considerably easier when someone leads you there.  Thanks again for your time and expertise.

Posts 853
Michael McLane | Forum Activity | Replied: Sat, Mar 30 2013 7:04 AM

The forums are, indeed, very helpful. I have always been thankful for my colleagues here who have come to my rescue. However, I have a different take on this. My initial thought was one of wondering why tech support off-loaded this problem to the unpaid forum users in the first place. Take my statement with a large grain of salt. I am one of those that has always been irritated and suspect of Logos's support philosophy (i.e. no manuals, depending on the community to develop and support a wiki, help, and tutorial system). Customer service, which includes tech support of all kinds, is in my opinion one of the most vital aspects of a tech company and one I think Logos gives marginal attention to supporting. Logos does a lot of things and has many products but doesn't seem to excel at any one of them. They're all just kind of... good enough. Now back to your regularly scheduled "programming". Zip it!

PS. Except for the forums, meaning the layout and execution, not the content (that's you people and it is great too). Its the best one I've seen. Don't get me wrong, I am a faithful Logos user. My personality, however, is one that would rather do one thing to my best rather than a dozen things just good enough. And, we all know that Logos (the software) could be more (for instance, quality control).

Posts 1649
Room4more | Forum Activity | Replied: Sat, Mar 30 2013 10:09 AM

Dave Hooton:

Apexite:
But I'm wide open to change from ESL to another single tool (or integrated tool set) that will yield the same information (definitions/root words/occurrences/translation counts) in an accurate, easy-to-get-to process.

Because a "critical" Greek bible is used by modern translations as against the TR Greek used by the KJV the ESL has limited usefulness for occurrences/translation counts. But you can get occurrences via the lemma search of an original language bible (a reverse interlinear is not as accurate).

A reverse interlinear bible makes translation counts possible  e.g. lemma:λόγος ANDEQUALS word => find where λόγος is translated as "word". You can also find where λόγος is not translated as "word" e.g. lemma:λόγος NOTEQUALS word. Now, if you right click "word" in an interlinear (and some Greek bibles), you will finds its Root and can conduct a search on that in the same way you have done for the lemma.

Have a look at Original Language Lexicons on the upgrade page, The 3 "Dictionary of Biblical Languages" (DBL) available to the Starter package are very useful; they provide alternate meanings and can link to ESL via Strong's numbers or the lemma. An analytical lexicon provides more information e.g. Lexham Analytical Lexicon to the GNT, but not much definition; use them as a supplement to the DBL. A theological lexicon concentrates on words of significance and does not cover all words of the NT or OT.

I have to admit, that I actually like what you wrote here Hooten. Very good.....maybe others should consider this.....

 

R4m.

DISCLAIMER: What you do on YOUR computer is your doing.

Page 1 of 1 (18 items) | RSS