Working With Des Moines Register Obituaries Is More Messy Than Anyone Admits
I spend a lot of time going through digitized newspaper archives, and if you're pulling obituaries from the Des Moines Register for family history research, you will run into at least one problem that the databases won't warn you about. The digitization process for this particular paper has gaps that are not randomly distributed. There are years where the image quality is acceptable, years where OCR failed on about forty percent of the columns, and entire stretches where the archive simply stops and doesn't tell you why. Uncovering The Hidden Stories Of The Past A Dive Into Des Moines Register Obituaries requires patience and a working knowledge of what you are actually looking at. The microfilm is somewhere in Des Moines, but nobody wants to drive there unless they have to. The online databases help until they don't. Here is how most people approach it and where they get stuck.
Uncovering The Hidden Stories Of The Past A Dive Into Des Moines Register Obituaries
Start with Ancestry and Newspapers.com. Both carry the Des Moines Register, but they do not carry identical content. Ancestry has the earlier portions going back to the late eighteen hundreds, though the coverage between roughly 1910 and 1940 is patchy. Newspapers.com has stronger coverage for the mid-twentieth century but skips some sections entirely. If you are researching a death that happened in 1927, checking only one platform will make you think the obituary does not exist. It exists. You are just looking in the wrong place. The search interface on both platforms is adequate but not great. Use broad terms. Searching for "Des Moines Register" plus a full name will sometimes miss the hit because the OCR has garbled the surname. Searching for just the surname and narrowing the date range to plus or minus thirty days is faster. Then scan the results manually. The algorithm is not going to hand you the exact article. Here is the specific problem I hit last month that nobody online seems to document. I was looking for an obituary from October 1938. The database returned a result for that date, but when I opened it, the OCR had merged two separate columns into one block of text. The obituary for my ancestor was there, but it was interleaved with a food page advertisement and a municipal court listing. Reading it required manually reassembling the text column by column. I solved it by toggling to the raw image view instead of the text extraction, which let me read the actual scanned page layout directly. None of the databases advertise that you can do this, but both Ancestry and Newspapers.com have a toggle or button for viewing the original scan. Use it whenever the OCR output looks wrong.
Another issue is the death index itself. The Register published periodic death indexes in the late nineteen hundreds, but those indexes are not consistently digitized. I found a reference to an obituary in a 1972 index that did not appear in the digital collection until I went to the Library of Congress chronicling America website, which has a separate feed for some Iowa newspapers. It took me forty minutes to find something that should have been in the main archive. The workaround is to check the Iowa City University library digital repository and the State Historical Society of Iowa database as secondary sources when the main platforms come up empty. Pay attention to the difference between obituaries and death notices. The Register ran both, and they contain different information. Obituaries are longer, often written by family members or the funeral home, and include biographical details, survivors, and service information. Death notices are shorter, more standardized, and sometimes just list the name, age, date of death, and surviving family. If you are hitting a brick wall, the person you are researching may have only a death notice and no obituary. That changes your search strategy entirely. You need to look for the shorter format, which means expanding your date range and using different keyword filters. The real limitation most people overlook is that the Register's digitized archive does not include every edition. Weekends, holidays, and special sections are sometimes missing. I discovered this when a death occurred on a Sunday and the obituary was pushed to Monday's paper. The database indexed it under the Monday date, not the date of death. If you are searching strictly by date of death without adjusting for publication lag, you will miss it. Always search at least two days past the death date.
Get the Full Details
Image quality is another factor. Issues from the nineteen fifties onward are generally clearer, but the nineteen forties have significant degradation in many runs. The paper stock from that era was lower quality, and the scanning process does not always compensate for the yellowing and printing bleed. When the image is unreadable, try adjusting the contrast and brightness settings in the viewer. Both platforms have basic image tools. A twenty-second adjustment can make previously illegible text perfectly readable. If you need something the online databases do not have, the next step is the microfilm at the Central Library in Des Moines or the county historical society in the relevant county. Mail requests are sometimes an option, but response times vary from two weeks to two months depending on staffing. I usually request a photocopy of a specific page rather than the whole week's papers. It costs a few dollars and saves you from waiting for a full reel to be processed. The most useful technique I have developed is keeping a spreadsheet of every search attempt, including the platform, date range, exact terms used, and whether a result was found. Without this, you will waste hours resubmitting the same query and convincing yourself you have not found the record yet when you have just searched it before. Five minutes of documentation per session saves several hours per project.
There is also the matter of name variations. Immigrant families often appear under anglicized versions in later obituaries while the original spelling appears in earlier ones. A person might be listed as "Jan Kowalski" in a 1912 death notice and "John Kowalski" in a 1955 obituary. Search both. Assume variation and search accordingly. This applies to Irish, German, Polish, and Scandinavian names especially. The Des Moines Register obituary collection is not a complete record of who died in central Iowa. It is biased toward residents of the city and its immediate suburbs. Rural deaths often went through smaller papers like the Courier News in Adel or the Washington County Record. If your ancestor lived outside Des Moines proper, check those papers too. The Register may have a brief mention, but the full obituary could be elsewhere. I have found that combining the Register search with Iowa death certificates from the Department of Public Health produces better results than relying on either source alone. The certificate gives you the exact date, cause of death, and informant's name. The obituary gives you the narrative. Using them together fills the gaps in both. The certificate application process requires a valid reason and proof of relationship for deaths less than fifty years old, but records over fifty years are publicly accessible through the state archives.
The bottom line is that this work is iterative and slow. You will hit dead ends. You will find partial matches. You will discover that some obituaries were never digitized and exist only on microfilm. None of this is unique to the Des Moines Register, but the specific quirks I mentioned above are the ones that actually slow people down, so knowing them in advance saves time you would otherwise waste figuring them out the hard way.
