Updating the Grid

There are a couple of updates to the Grid that I'd like you to start making yourselves.

If you don't have a birth year, but you do have the marriage date of the parents, just enter black Xs on the Grid. I'd rather not add black dots in those cases. From a data user's perspective, they mean the same thing anyway.

You can delete death information yourselves. There is no reason to wait for me to do it. There is a Delete button at the bottom of the death screen.

If you do these things yourselves, then you won't have to wait for me to handle them for you.

Evidence that it's important to label your remarks

Recently, Professor Costa had me finding some information in an additional comments field in the disease data. These fields are like the Remarks fields in the VCC screens. Just as you are supposed to label your notes in the remarks with the field the data refer to, we did the same thing in disease data collection. If the information is labeled correctly, you know that what you're reading is the duration of the contributory cause of death and not something else.

Because the information in the general comments was tagged with the the appropriate question from the disease screens, I was able to more quickly and accurately find the information requested. Without those labels, I would have had some confusion about whether some of the comments answered Professor Costa's question.

So, it's important to label anything you type in remarks. It will prove useful later.

Results of Top-priority POW soldier redo

You all did really great work on the POW soldiers we had you check. Here are a few results:

• Total soldiers updated: 465
• New census links: 586
• New death links: 461
• Additional death causes: 482
• This is especially good, because we only had 2,200 death causes before the redo work

Thanks for all your work!

Mother-child relationships

This is just a reminder that you should always make sure the mother-child relationships on your family trees are correct. When Chris creates the GEDCOM files that you use for your trees, he has no way to put the correct child with the correct mother. We would like to add this information as a variable in our data. Chris will be going back to the trees to extract it.

It's really important that you take the time to make sure all the children are connected to the correct mothers on your family trees.

Progress as of May 24

Here is our progress report for May 24, 2017. In addition to this, you completed all the top-priority POW soldiers. There were 465 of them.

	Project 1	Project 2 - USCT
	n = 8,500	n = 4,500
Total soldiers complete	6551	2049
Soldiers completed during last  week	48	14
Soldiers w/o children (to be removed from sample)	596	351
% of completed soldiers w/o children	9.1	17.1
Soldiers with children complete	5955	1698
% with children complete	70.1	37.7

VCC People to be deleted from Grid

This is for those of you who use the VCC People to be deleted from the grid Google sheet.

If I delete a census decade, I'm happy to change the X to yellow. I will not automatically assume it should be yellow. If you want me to do it, then you need to tell me when you make the deletion request.

Please do not ask me to add black dots before you've done work on the individual. If you add census for that person, which we hope you do, the screens will add the black dots for you. Only ask me to add black dots if the screens have calculated incorrectly, or if you find no census records and all your Xs are yellow.

You don't have to do a lot of explaining or tell me that you or the previous inputter made mistakes. I understand that mistakes happen. Just make sure I understand what I need to do. Do add more explaining if what you're asking me to do is complicated or confusing.

Progress as of May 3

Here is our progress as of May 3, 2017.

	Project 1	Project 2 - USCT
	n = 8,500	n = 4,500
Total soldiers complete	6444	2012
Soldiers completed during last  week	61	28
Soldiers w/o children (to be removed from sample)	588	346
% of completed soldiers w/o children	9.1	17.2
Soldiers with children complete	5856	1666
% with children complete	69	37

Does anyone have questions about what this report means?

Quick checks

Recently, there has been a little confusion about checking. We check our work for two reasons:

1. Quality control - We want to produce the best possible data collection.
2. Ongoing training - Checking helps us figure out where the holes are in people's knowledge of the work. It also helps keep us all on the same page and following the same rules.

When you do a quick check, try not to spend more than 60 minutes. Check the important stuff first.

• Tree
• Grid
• Update Relations
• Search for missing census decades and death information
• Make sure the inputter updated HH0 (You should be able to figure this out by noting whether fields that are always wrong in the original data have been updated and not have to recheck all of it.)

After you've done all this, if you still have time, spot check the inputting. Stop when you reach 60 minutes.

On your checking documents, write a note about everything you check. If you didn't find any errors in a decade or death screen, write that down. Make sure we can tell what type of error it is by looking at what you've written. Follow all the formatting instructions that Sandy has given you in the past.

I promise, we don't make you do checking as a punishment. We want you all to have the tools you need to do great work. The reason we do a quick check and a full check instead of two full checks is to save time.

April 2017 checking stats

In April, we checked 46 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? - 11
Missing HH member - 8
Duplicate people - 0
Wrong person - 0
Other - 11

Inferred Relationships

Incorrect relationships - 10!

Census Errors

Name - 34
Typo/Reading/Wrong - 117
State Code - 0
Missing/Wrong URL - 1
Missing data - 47
Additional finds - 19
Quality Code - 10

Death Errors

Typo/Reading/Wrong - 18
Missing data - 30
Missing/Wrong URL/Source - 9
Quality Code - 4
Additional finds - 15

Tree Errors

Missing/Incorrect information/relationships - 13

The total number of differences for all 46 soldiers is 357. This is 66 more differences than January, when we checked 41 soldiers. Some differences/errors affect the data more than others. If we checked other pensions, we'd probably find similar differences. Some mistakes are inevitable, but please pay close attention so that the number of errors can be minimized.

Progress as of April 26

Here is our progress report as of April 26, 2017.

	Project 1	Project 2 - USCT
	n = 8,500	n = 4,500
Total soldiers complete	6383	1984
Soldiers completed during last  3 weeks	175	57
Soldiers w/o children (to be removed from sample)	582	342
% of completed soldiers w/o children	9.1	17.2
Soldiers with children complete	5801	1642
% with children complete	68.2	36.5