State and country codes, again

With this new VCC sample, we are cleaning and standardizing the data in batches as we go, instead of waiting until we're finished with all data collection. So, some of the ladies here in the office are cleaning the most recent data that you've all collected. Remember, the better you input, the easier you make their lives.

They've noticed that there are still a lot of mistakes being made with state and country codes. We had Chris check which inputters were making the mistakes, and it turns out the problem is quite widespread. You should be checking your manual for the correct codes. Feel free to leave the manual on your desk open to that page for easy reference. With that in mind, here is a repeat of a post from July 13, 2015.

State and country codes

We've noticed incorrect state and country codes being used recently. It's not uncommon for this to happen. Please don't assume you know the two-digit code, especially all of you who are new. Here are some commonly confused state and country codes.

• Nebraska is NE not NB, which is New Brunswick.
• Arkansas is AR not AK, which is Alaska. We rarely input AK.
• England is EN not EG, which is Egypt.
• Indian Territory is RT not IT, which is Italy.
• Prussia is PU not PR, which is Puerto Rico.

Bonus fun fact: German states have their own codes. So if the person was born in Bavaria, you may type BV, Brandenburg is BR, etc.

If you're not sure what the correct code is, they can all be found in Appendix A of your manual.

Apostrophes

Just a reminder - Please don't type apostrophes in names. It is causing problems in the screens. If the name is O'CONNOR or O'NEAL, then type OCONNOR or ONEAL.

Checks

If a checker says in the checking document to ask Heather or Sandy about something, please go to them (or email Sandy) and ask unless you have been able to resolve the question yourself.  Do not assume that one of us is reviewing every check and will automatically answer the question.  We want to see, first of all, that you are actually going through the checks carefully.  Also, we don't want to stop and open up grids and documents if we don't have to because you have been able to resolve the issue already.  With some experienced inputters, we may not review the checks at all.

It is your responsibility to make sure that all questions from your checks have been resolved.

Assignment reminder

Please let me know when you're nearing the end of your assignment. When you reach your last one or two soldiers is plenty of notice. If you wait until you're finished with your assignment to let me know you need another one, then you might have to wait for it. I might be busy with other work, or I might have gone home ill. Most of you who work outside of the office are good about this. Some of you that work in the office are good as well. Just because you work in the office, please don't assume I'm checking your log every day and will see that you need a new assignment.

Giving me advanced notice that you will need a new assignment will help ensure that you always have work to do.

Please review music policy

Everyone needs to review the music policy stated in the September 1, 2015 blog post.

We have noticed that a lot of people in the office are wearing headphones. Trainees should never listen to music.

For everyone else, if I notice your quality of work goes down, you will not be able to listen to music either. If you are allowed to listen to music, you must turn the volume down. Humming, whistling, and tapping fingers and toes are not permitted. Please be courteous to those around you. Not only is this disruptive, it means you're paying more attention to the music than you are to your work.

Please email by the end of the day tomorrow (July 19) to let me know you have read and understand the blog post from September 1, 2015.

Progress report as of July 13

Here is our progress as of July 13, 2016.

	Project 1	Project 2 - USCT
	n = 8,500	n = 4,500
Total soldiers complete	3634	1063
Soldiers completed during last week	46	47
Soldiers w/o children (to be removed from sample)	379	191
% of completed soldiers w/o children	10.4	17.9
Soldiers with children complete	3255	872
% with children complete	38.3	19.4

When to use "Other" as a death source

We recently had a question about when to use the "Other" box in the Death sources section of the Death screen. Here are some guidelines.

Use "Other" when:

• You have added additional information to the Death screen from a third or fourth source, so you can't add the URL to the screens. We're trying to identify where all the information entered on the screen came from.
• There is not a check box for the source you're using. If you found a record on Ancestry, check the Ancestry box. If you found it on a Missouri death index, and there's no box, check "Other," and type the name of the source exactly as it appears.

Every single death source you find for an individual doesn't have to be added, especially when there is no additional information. ALL death sources should be attached to the family tree.

Input what you see

We've noticed that several people are leaving information out. Please input what you see. This has mostly been a problem on the death screen. Even if a word is difficult to read, you must type something. Do not type one word of a pair and leave the other out. You may ask for help if you can't read something. Please type all the information in the appropriate fields.

Question about hints

There has been some speculation as to whether the family tree hints are getting better? What do you all think? Have you noticed an improvement in the quality of hints Ancestry is able to generate? Have you noticed anything else interesting about hints? Please share in the comments.

June 2016 checking stats

In June, we checked 42 soldiers under our system. I've reviewed all of the checks, and I've tallied the number of differences. Some of these are errors, and some are judgment calls. Here are the categories and the total number of differences for each category. 

GRID Errors

• MILIN?/MAR? - 6
• Missing HH member - 8
• Duplicate people - 1
• Wrong person - 0
• Other - 4

Inferred Relationships

• Incorrect relationships - 14

Census Errors

• Name - 30
• Typo/Reading/Wrong - 68
• State Code - 0
• Missing/Wrong URL - 0
• Missing data - 24
• Additional finds - 27
• Quality Code - 13

Death Errors

• Typo/Reading/Wrong - 4
• Missing data - 11
• Missing/Wrong URL/Source - 2
• Quality Code - 2
• Additional finds - 12

Tree Errors

• Missing/Incorrect information/relationships - 10

The total number of differences for all 42 soldiers is 236. This is 100 fewer differences than last month. Great job! Some differences/errors affect the data more than others. If we checked other pensions, we'd probably find similar differences.

Alternate names on family trees

When you're adding records to your family trees, if the name on the record IN ANY WAY DIFFERS from the name you already have, please click the Alternate name button. Please do this for the soldier and all his family members. If you're not sure if you've already added a particular alternate name, go ahead and click it just in case.

This will help with analysis of some the name algorithms we've been working on (or something like that; I'm really not sure how it all works).