How do I target my records search to "ED 14 Election Districts 6-7 Hanna & Carbon Carbon city"
I'm trying to search the 1900 census of Hanna, Carbon, WY. For each person the system says the "event place" is "Election Districts 6-7 Hanna & Carbon Carbon city, Carbon, Wyoming, United States". How do I search all the people who have the same event place? I'm getting the input validation error "Input cannot contain the following characters: < > ~ ` + = % # ^ : & | ( ) { } [ ] \". I want to exclude all other results from Carbon.
Best Answer
-
That ED is 36 pages. The simplest way is to browse those 36 pages.
0
Answers
-
Ah, I see: Search - Records disallows the ampersand (&) in the placename, despite it being clearly present.
The workaround I can think of is to replace the ampersand with a wildcard (? or *), but it's not working: every variation I've tried either gives "No Results", or lists all sorts of places in Carbon county, none of them Carbon city.
It seems to me that the ampersand is messing things up from both ends: I searched for an exact name from one of the Carbon City pages (Ettie Cardwell), which came up with four matches: two on the page (mother and daughter), one in Texas, and one in Alabama. I tried the "Residence" filter on the search results, and my only choices were Texas and Alabama. Wyoming wasn't on the list. In other words, the ampersand is rendering this placename invisible to Search.
This needs to be brought to the attention of the engineers: either the algorithms need to be changed to allow the ampersand as a valid character in placenames, or all of the indexes and standardized placenames need to be edited to replace "&" with "and".
0 -
Thank you for the analysis. Now let's hope the problem gets recognized. This might affect quite a large amount of records.
0 -
@Alahärmä you ask, in a separate thread, how you can get this issue in front of a technician.
I guess, before we go down that path, I am wondering why @Áine Ní Donnghaile's response was not adequate. It appears that all of Election Districts 6-7 Hanna & Carbon Carbon city, and only that, is provided by the waypoint for which seem provided a link. This seems to meet all of your criteria.
I think the problem is more basic than was suggested. When you look at the Place search field for the 1900 United States census, notice that it specifies City, County, State etc. I tried searching (using the Residence field) for "Election District*" and got the result, "No Results Found." If there were any Place name that included "Election Disttrict," I believe that I would have gotten different results.
I also did the same search using Hanna, Carbon City, and Hanna and Carbon City, with wild cards - in all instances, I got the same result.
So I believe that using the Waypoint that was provided would be your only option.
I believe that engineers will tell us that the system is working the way it was designed; however, I will continue to pursue the issue until I can get a better answer.
1 -
@Mike357 @Áine Ní Donnghaile's response was inadequate because a manual search is slow and prone to errors. I am accustomed to adapting my behavior to the requirements of the systems I use to get adequate results, unless it is impossible. I'm sorry I'm not sure I understand what you mean by Waypoint. Is that a LDS term?
Edit: I'll come back to this.
0 -
Ah, ok, so the linked index pages are called or do contain "Waypoints". https://www.familysearch.org/search/image/index?owc=9BQY-L2Q%3A1031701301%2C1030636801%3Fcc%3D1325221
@Mike357 Do you think it is proper that if I do an exact search for "carbon, wyoming, united states", I get 28 pages of hits from Rawlins (and some Dexterville) and not from anywhere else?
For example "Leonard L Lawes" is found from Hanna/Carbon city only if I remove the exact match requirement.
Coincidentally Rawlins and Dexterville are the only districts that do not contain an ampersand in their name:
It seems that those other districts are not recognized as belonging to Carbon at all. I would bet that the locations have been discarded in the indexing process completely.
0 -
Somehow I've become the bad guy here. Not sure why, but I don't much like it.
1 -
I gave your comment a "Like" because I agree you have been unfairly criticised here - not because I think you are "the bad guy"!
On the general issue: in the past I have sat for hour upon hour, carefully examining microfilmed census records for a specific area (town/village). I believe indexes have made us a little unwilling to search in the way many of us had to in the past. Indeed, the problem with an index is that the very individual who has been missed (from the page) in the indexing process could well be the person we are seeking!
I think you offer good advice in saying, "The simplest way is to browse those 36 pages". If I thought my concentration might have slipped during a similar exercise, I have even searched some (or all) of the pages a second time.
Too many records have been overlooked during indexing. If the originals are available, always read through them, even if it takes (say) an hour or two.
1 -
@Alahärmä, I appreciate your comments. Based on them I reviewed all the Election Districts for Carbon County in the 1900 U.S. Census and tried to do an Exact search for all of the towns listed. The only ones that I could get any results were Rawlins City and Dexterville -- which is what you found.
It's interesting that, although I can find Dexterville results, I get nothing for Battle Lake in the same Election District.
I then did an Exact search for "Carbon, Wyoming" and only found results for Rawlins City and Dexterville. I then did a search for "Carbon, Wyoming" without the requiring an Exact search, and got the same result, and, after, subtracting the results that were not in Carbon County or in Wyoming, I found the same number of hits for the exact and non-exact search. This, of course, was just rehashing what you had done.
Looking at what effect the use of an ampersand might have in the Place name, I did a successful exact search for Evanston City, Unita, Wyoming. The place name for this search was "Election District 10 & 1 Evanston city, Uinta, Wyoming, United States," so this result suggests that the "&" in the place name doesn't seem to be the problem, notwithstanding this wasn't part of Carbon County.
However, we don't have to solve the problem. You have lead us to to the point that we know there is a problem, and so the next step is to move this to the engineers to figure out why almost 90% of the locations in Carbon County are not searchable.
At this point, I will put this issue in the queue that will be reviewed and forwarded to the engineers. As usual, we have no way to anticipate when the fix will be made. There are several issues related to the 1900 U. S. Census that are being worked by the engineers, although this issue does not seem to be related to the others.
0 -
@Mike357, I can't figure out what you searched for, and where the ampersand appeared.
If I search the 1900 census for residence = "Evanston City, Uinta, Wyoming", marked exact, I get 2265 results displayed as "Evanston, Uinta, Wyoming, United States". In the first thousand of those results, the only appearance of the word "city" is in my search input, and the only ampersand on the page is in the collection filter ("Collection: Censuses & Lists, ..."). Searching for any variation (that I could think of) adding "Election District" to the search term gave No Results.
(Aside: doesn't the E in ED stand for "enumeration", not "election"?)
The ampersand is definitely a problem on the search input end: the page does not allow it in any placename field.
Does anyone know of another placename that appears in search results with an ampersand? The Hanna & Carbon example has been shown to be invisible to Search; are there others?
0 -
@Julia Szent-Györgyi, just time to respond to your aside. Yes, my understanding is that ED is Enumeration District. The census seems to use Election District in a similar way that we often see "Precinct".
In the case of Frank W Crase, in the original census, I see Election District No. 1 and, separately, Enumeration District 58.
0 -
Thank you, @Paul W. That was my perspective.
[Going back in my bunker where I'm scrolling through pages of the 1950.]
0 -
@Áine Ní Donnghaile I'm sorry that I'm not able to express myself sensitively enough. English is not my first language and that causes at least some of the trouble. I understood that I was expected to state my position as clearly as possible to get the technical problem corrected in the system.
0 -
Ping @Mike357
Another indexing issue is with the Idaho Valley Lake precinct 1920 census. For some reason the original event location (Lake, ED 177, Valley, Idaho, United States) has been replaced by (Lake, Fremont, Idaho, United States) which is plain wrong. See https://www.familysearch.org/ark:/61903/1:1:MDC1-HHK for an example. This error excludes the Lake, Valley, Idaho, United States residents from the search results when searching the 1920 census for Valley, Idaho, United States residents.
Because of my poor knowledge of the area, I checked Wikipedia to make sure that this could not be some kind of a border case.
0 -
@Alahärmä, that Valley-versus-Fremont example is an autostandardization error. Notice that there are two Event Place fields, one of them labeled "(Original)". That's what was actually indexed. The other one is what the bot picked from the places database to go with that text. Yes, it's basically random: whatever came up at the top of the list is what got put in. There are many millions of such errors in FS's databases now, but we keep getting told to just report them individually.
2
