New Data Users: Getting Started

With your newly-extracted dataset, please be aware of a few basic problems you will likely encounter as a new data user along with some basic assumptions you will need to make in order to use the data correctly. The links below provide information that is relevant to all data users.

• Dates in the UA and USCT Data Sets
  
  Dates in the UA, USCT, and other datasets will need to be reformatted for use in your statistical software. Routines for Stata and SAS are provided, and can be modified based on your assumptions about missing information, etc.
  
  
• Complications Associated with Pension Law and History
  
  Second, the history of the pension complicates the use of these data. In a nutshell, prior to 1890, pensions could be legally awarded only to men who had sustained war-related injuries. After 1890, age-related conditions and other health problems were pensionable. For this reason, there's a large influx of men into the pension after 1890. Many researchers will find this selection to be a problem--i.e., if only the injured can be observed prior to 1890, the pre- and post-1890 pensioners aren't roughly equivalent. For that reason, many data users have limited their use of pension and post-war medical data (Surgeon's Certificates) to soldiers whose first exams occurred after 1890.
  
  A full explanation of the evolution of the Civil War pension and how it should impact individual studies using pension data is available from the link below :
  
  

  The Civil War Pension Law

  Linares, Claudia | #2001-6/CPE Working Paper Series
  
  
• Best Estimate of Soldiers' Ages
  
  The variable bestage_1891 represents our best estimate of the soldiers' age on January 1, 1891, based on the 82 possible indicators of age in the Union Army data. This date should be used to derive the soldiers' ages at any other date until you are comfortable with the Union Army data.
  
  
  
  
• Conclusion
  
  Finally, if the data extract you have created includes post-war medical data, you should be aware that the symptoms and descriptions of the conditions the soldiers presented at examination make up the bulk of the variables in the Surgeon's Certificates. There are no "diagnoses" per se, and these must be inferred from a combination of the symptoms and the surgeon's ratings (_rat and _rel) variables. If conditions were "rated" by a surgeon or "related" to another rated condition, it is often assumed that those conditions were present and disabling. If there's no rating associated with the symptoms and descriptions presented by the soldier, then it is probably the case that the surgeon did not find that the condition rose to the level of a disability under the law.
  
  We have provided some interpretation of the ratings variables (see ratings indexes for each set of disease variables), but because many rated conditions were grouped together into one rating for all ratable or related conditions (e.g., "Heart rated with diarrhea, 3/4), it is probably best as a beginning user to simply note the presence of a rating rather than trying to put the ratings into some consistent fractional or dollar value. The laws changed a lot over the lifetime of the soldiers, and the maximum rates for certain conditions changed over that time, as did the maximum pension awards, etc. Until you have a good grasp of pension law and medical history, much can be gleaned from simply creating dummy variables that represent if a condition was "rated" or not.
  
  It is also important to remember that a condition may be "rated" but have no information in the _rat variables. This is because the condition may have been "related to" a primary rated condition, or because the condition was "rated with" another condition. So, conditions appearing in the _rel variables must be considered "rated" conditions.
  
  Some articles describing pension law include the following from our scholarly articles archive:
  
  

  The Civil War Pension Law

  Linares, Claudia | #2001-6/CPE Working Paper Series
  
  Some articles describing civil war medical history include the following from our scholarly articles archive:
  
  

  Health and Labor Force Participation Over the Life Cycle: Evidence from the Past

  Costa, Dora L., ed. | Chicago: University of Chicago Press for NBER
  
  

  The Effect of Hernias on the Labor Participation of Union Army Civil War Veterans

  Song, Chen and Nguyen, Louis | In Dora Costa, ed., Health and Labor Force Participation Over the Life Cycle: Evidence from the past, ed. D.L. Costa, 253-310. Chicago: University of Chicago Press for NBER
  
  

  Understanding Mid-Life and Older Age Mortality Declines: Evidence from Union Army Veterans

  Costa, Dora L. | Journal of Econometrics 112: 175-92
  
  Some articles describing how other researchers have tackled the issues of "ratings" and "related conditions" and "diagnoses" include the following from our scholarly articles archive:
  
  

  Changing Chronic Disease Rates and Long-Term Declines in Functional Limitation Among Older Men

  Costa, Dora L. | Demography 39: 119-37
  
  

  Work and the Disability Transition in 20th Century America

  Wilson, Sven, Burton, Joseph and Howell, Benjamin
  
  

  Table of rates for specific disabilities

  Sanchez, Mario | CPE
  
  Please feel free to search our publications database at

  http://www.cpe.uchicago.edu/publication/publication.html

  or view a complete list of our scholarly publications here.
  
  Other suggested reading:
  
  

  Battle Cry of Freedom: The Civil War Era (Oxford History of the United States)

  McPherson, James M.
  
  

  Compendium of the Civil War

  Dyer, Frederick
  
  

  Back to top