[Utopian Stats] Analyzing Utopian Translation Moderation - Blockchain Business Intelligence

After analyzing the overall moderation time of all Utopian categories, this report drilled down into translations and looked into its moderation review time, rejection rate, and the moderator's user acceptance rate - the frequency of contribution acceptance of a moderator from a certain author.

This analysis aims to dig into the translations and see if any pattern emerges and if there's anything we can learn from it.

Outline

1. Translation Contributions Overview
   1. Monthly Contributions
   2. Average Review Time
2. Translation Moderators
   1. Rejection Rate
   2. Review Time
3. Translation Contributors
4. Conclusion

Scope of Analysis

The data extracted was from January 19, 2018 to March 13, 2018.

1. Translation Contributions Overview

Based on the the left table, there were 9846 translations that were reviewed from January 19 to March 13, 2018. Moderators took an average of around 32 hours to review each contribution.

The fastest review time was 35 seconds for this contribution while the longest review time was 323 hours (13.5 days) for this contribution.

23% (2305) of contributions were rejected, while 77% (7541) were accepted.

Based on the table on the left, the average review time per month was 40 hours for January, 28 hours for February, and 26.5 hours for March.

---

2. Translation Moderators

67 moderators reviewed the 9846 contributions. This amounted to an average of 147 contribution per moderator.

---

Based on the chart on the left, the average contribution per moderator was 63 for January, 100 for February, and 35 for March.

This would equate to around 2 contributions per day per moderator for January, 3 per day for February as well as for March (only 13 days as of date of data extraction).

---

2.1. Rejection Rate (Top 10 Moderators)

To make the figures readable, I decided to focus on the top 10 moderators to represent the whole translations moderation team.

Based on the right-most bar chart, for January there was a 35% rejection rate, 19% for February, and 16% for March.

This was what attributed to the high acceptance rate (77%) for this category.

@jmromero led all moderators by reviewing 680 contributions and rejecting 18%.

@rosatravels was at second with 628 contributions reviewed and rejecting 9%.

And @toffer on third reviewed 559 contributions and rejected 11%.

From Ranks 1 to 10 of the same table, 8 out of 10 moderators had a high acceptance rate - around 80% to 90%.

2.2. Review Time

The table on the left was the review time of the moderators sorted by number of contributions. The average review time of the top 3 moderators were: 21 hours for @jmromero, 18 hours for @rosatravels, and 52 hours for @toffer.

The chart on the right shows the moderators sorted by descending average review time. I was hoping to see a correlation between the rejection rate and the average moderator review time. Using this visualization, it seemed that there was no such correlation. But viewing it differently gave me this:

^{The x-axis (horizontal) is the moderation time from 0 hours to 160 hours. The y-axis (vertical) is the number of contributions.}

The bar chart plots the number of contributions reviewed (rejected or accepted) relative to the amount of review time (x-axis). The pink line chart is the % rejection in relation to review time.

From it we can conclude the following:

The shorter the review time, the higher the possibility of a contribution to be accepted.
The longer the review time, the higher the possibility of a contribution to be rejected.

Highlighting just the table from the above graphic.

If your translation contribution gets reviewed within the first day, there's an 18% possibility of your post being rejected. If it reaches the second day, it's 24%, on the third it's 28%, on the fourth it's 39%, and beyond the fourth day it's 55%.

Let us now look into the contributors.

---

3. Translation Contributors

There were a total of 1084 translators rewarded. @jmotip led the group with 152 translations at 0% rejection. His contributions were reviewed at an average of 14 hours.

@rooneey was at second with 124 translations at 2.36% rejection. His contributions were reviewed at an average of 11 hours.

And @shoganaii on third with 126 translations at 1% rejection. His contributions were reviewed at an average of 26 hours.

And lastly let's look into the moderator-author acceptance network.

4. Moderator-Contributor Network

The bigger the moderator's circle, the more times he or she has accepted or rejected an author's contribution.

Let's look at the individual stats.

16 moderators reviewed @jmotip's 152 translations. @felander reviewed 30% of it at 0% rejection reviewing the former's contributions at an average of 16 hours with 53 minutes as the fastest and 50 hours the longest.

---

15 moderators reviewed @rooneey's 127 translations. @joe28 reviewed it 35 times (28%) and rejected it 1 time (3%) reviewing the former's contributions at an average of 16 hours with 5 minutes as the fastest and 70 hours the longest.

---

And lastly into the third spot - 17 moderators reviewed @shoganaii's 130 translations. @plokmi reviewed it 22 times (17%) and rejected it 0 times reviewing the former's contributions at an average of 27 hours with 29 minutes as the fastest and 66 hours the longest.

---

5. Conclusion

From this analysis, we learned that there were 9846 translations reviewed from January 19, 2018 to March 13, 2018. 67 moderators reviewed this at an average of 2 to 3 contributions per day completing a review after around 32 hours.

We also learned that the translation category rejected contributions at 23%, lower than the 40% average rejection rate across all categories. We learned that this was due to the low rejection rate for the months of February and March at 19% and 16%, respectively.

We also learned that 8 out of the top 10 moderators in this category had a high acceptance rate - around 80% to 90%.

We also looked into the details of the review time of the top 3 moderators and learned that their average review times were 21 hours for @jmromero, 18 hours for @rosatravels, and 52 hours for @toffer.

Most importantly, we found out that for this category the possibility of your contribution being rejected gets higher as time passes. It had an 18% possibility of getting rejected on the first day, 24% on the second, 28% on the third, 39% on the fourth, and 55% beyond the fourth day.

And lastly we found out that there were 15 to 17 moderators who reviewed the contributions of the top 3 contributors. An idea that came to mind to mitigate abuse is perhaps adopt a system similar to what @curie does (although I have never curated for it) - a moderator is only allowed to review (accept or reject) a post from a certain only if at least (let's say) 24 hours has passed. Again, just an idea.

And one final thought that came to me while doing this report:

I look at moderating for Utopian as something noble. I've benefited from the open-source movement ever since I started my career as a software developer in 2003. Giving back to the open-source community is an honor in itself and Utopian is a tool that aims to reward those efforts.

Previous Reports

1. Analyzing Overall Utopian Moderation Time
2. Analyzing Development Moderation Time
3. Analyzing Translation Moderation (this report)

Tools

• Power BI for charts; M Query to transform *.json to Table

let
    Source = Json.Document(File.Contents("E:\develop\east\biutopian\utopian-io\posts_json-metadata_created_author_permlink.json")),
    #"Converted to Table" = Table.FromList(Source, Splitter.SplitByNothing(), null, null, ExtraValues.Error),
    #"Expanded Column1" = Table.ExpandRecordColumn(#"Converted to Table", "Column1", {"_id", "author", "permlink", "json_metadata", "created"}, {"_id", "author", "permlink", "json_metadata", "created"}),
    #"Expanded json_metadata" = Table.ExpandRecordColumn(#"Expanded Column1", "json_metadata", {"type", "moderator", "questions", "score"}, {"Meta.type", "Meta.moderator", "Meta.questions", "Meta.score"}),
    #"Expanded Meta.moderator" = Table.ExpandRecordColumn(#"Expanded json_metadata", "Meta.moderator", {"account", "time", "flagged", "reviewed", "pending"}, {"Mod.account", "Mod.time", "Mod.flagged", "Mod.reviewed", "Mod.pending"}),
    #"Filtered Rows" = Table.SelectRows(#"Expanded Meta.moderator", each ([Mod.account] <> null) and ([Mod.pending] = false)),
    #"Replaced Value" = Table.ReplaceValue(#"Filtered Rows","T"," ",Replacer.ReplaceText,{"Mod.time"}),
    #"Replaced Value1" = Table.ReplaceValue(#"Replaced Value","Z","",Replacer.ReplaceText,{"Mod.time"}),
    #"Changed Type" = Table.TransformColumnTypes(#"Replaced Value1",{{"Mod.time", type datetime}}),
    #"Replaced Value2" = Table.ReplaceValue(#"Changed Type","T"," ",Replacer.ReplaceText,{"created"}),
    #"Changed Type1" = Table.TransformColumnTypes(#"Replaced Value2",{{"created", type datetime}}),
    #"Renamed Columns" = Table.RenameColumns(#"Changed Type1",{{"Mod.account", "Moderator"}})
in
    #"Renamed Columns"

• Utopian data dump
• Studio 3T MongoDB for exporting data from local DB

db.posts.find({created: { $gte: '2018-01-19' }}, { 'json_metadata': 1, created: 1, author: 1 } )
db.posts.aggregate([
    { $match: { created: { $gte: '2018-01-19' } } },
    {"$group": { _id: "$json_metadata.type", 
                count:{$sum: 1} } },
    { $sort : { 'count' : -1}}
])

---

^{I am part of a Blockchain Business Intelligence community. We all post under the tag #BlockchainBI. If you have an analysis you would like carried out on Steemit or Blockchain data, please do contact me or any of the #BlockchainBI team and we will do our best to help you...}

^{You can find #blockchainbi on discord https://discordapp.com/invite/JN7Yv7j}

---

Posted on Utopian.io - Rewarding Open Source Contributors

---