After a few years doing research around military and veteran issues and populations, we’ve received many requests for information or assistance from students and other researchers interested in conducting research in the area.  Here’s some of what we’ve learned along the way that may be helpful. Some of what is below relates to studying the military, but it applies to veterans.  Veterans served in the military and maintain some of the culture of that institution after serving.

Data access

  • We don’t have any unique access to data or information. We rely on the same publicly available data as anyone else, including the Census, the CPS, ACS, and other nationally representative datasets.  Or we collect our own data through qualitative interviews, content analyses, survey-based or field experiments, and archival research.
  • Those who do have access to data may not be able to share it with you, or may not want to. People work long and hard to cultivate relationships that enabled their access to information, and become territorial.
  • The Defense Manpower Data Center (DMDC) hold access to military administrative data, but it is very difficult to gain access to their data, whether in or out of the DoD or government.  Some researchers have gained access to specific information through DoD FOIA requests  or VA FOIA requests. You typically need a formal DoD sponsor in order to get access to DMDC data.
  • Most public data include very little about the actual nature of an individual’s military experience. Rank, MOS, branch, combat exposure and other crucial distinctions within the military are simply not recorded in most national surveys.
  • Large, public data sources include: Census, ACS, CPS. Other nationally representative data include: the NLSY, NHIS,
  • Lack of data access and/or lack of high quality data is one of the major impediments to advancing social science research about veterans. Persistence and creativity are required to overcome the substantial hurdles to doing research on military and veteran populations.


Data collection

  • Your university’s IRB or human subjects review committee may require different rules when collecting data from military populations. Some will consider military populations just like anyone else, while others (we have heard) may require specific authorization from a military organization for your data collection, especially if you receive funding from the DoD. Be prepared for the unexpected.
  • Military populations are over-surveyed and under-studied (in our opinion). Military members and their families, and veterans are constantly receiving requests for surveys and other short data collections.  As someone who receives health care at a military treatment facility (MTF), I get a letter about 5 days after every visit to complete a satisfaction survey. There are internal researchers in DoD, the VA, think tanks, contracted firms, and academics all collecting data on the same relatively small population. Few of these studies are widely shared with the public, and fewer of these data sources are made available to others. The lack of sharing means duplicative collections and survey fatigue.


Military culture and structure

  • For those new to the military and veteran space, it is important to understand the distinction between the Department of Defense (DoD) and the Department of Veterans Affairs (VA). There are sharp organizational dividing lines between these two populations, their missions, their resources, their data, and their stakeholders. They are not a monolithic “military” organization. In fact, the separation of responsibilities across these two organizations is a major problem for those transitioning to civilian life.
  • The military and veteran population tends to feel most comfortable with outsiders who are military-connected (i.e. military spouse, military child, veteran, etc), especially outsiders who are wanting to conduct research. Military and veterans are  concerned about researchers they can trust (especially since they are part of research efforts so often) and a military affiliation not only aides the researcher in navigating the social and cultural dynamics in the community but also builds rapport with the population of interest. If you are not affiliated with a military or veteran status and you’re interested in conducting research, you may want to talk with someone in the community of interest about best ways to approach and interact with that population.
  • The Department of Defense and the Department of Veterans Affairs are not set up to help outside researchers conduct their research. They may be interested in what you find, but they do not exist to facilitate your research. Each organization has their own internal research systems and structures. Any research you are looking to do on the military population will likely need to be conducted outside of these institutions (I.e. by utilizing survey data, building relationships with veteran service organizations, etc), unless you have a formal relationship with these organizations.
  • When doing research on the military or veteran population it is important to consider and understand military culture. There are certain overarching features of military culture such as crude/abrasive humor, authoritative leadership styles, and discipline styles that dictate much of the social environment in the military community. There are also sub-cultures within different military communities based on service branch and MOS, so it is important to carefully understand military culture prior to conducting research, especially if it is qualitative or requires face-to-face interaction.


Military demographic realities

  • Military and veteran populations make up a fraction of the overall population.  They are a relatively small share (5-8% of adults depending on how you define the group and what data you consider.) And this share is declining rapidly as those who served when it was more normative succumb to the force of mortality. This means even in relatively large datasets, there will be relatively few veterans to observe.
  • If you want to understand a small group within the military and veteran population, it will be difficult to use even large national data sources. Women make up about 15% of the active military force, and were a smaller fraction among older cohorts of veterans. Immigrants and non-citizens serving are also relatively few, as are Muslims. If you want to understand small minority groups within the force, you will face tremendous data challenges.
  • Those who serve in the military are a selective subset of the population. This selection is both driven by self-selection (those who choose to enlist) and institutional selection (who the military allows to join.) This means military members reflect a skewed demographic: younger, more male, healthy, typically a high school graduate, more likely to have family who served, etc… These difference must be accounted for whenever comparing veterans and non-veterans.
  • If you approach studying veterans as poor saps who enlisted because they didn’t know any better, you’re starting on the wrong foot. Most contemporary research on enlistees suggests that those who serve tend to come from the middle of the socio-economic distribution. Young adults with plenty of options—like those from elite or wealthy families typically self-select into college rather than the military. But those with the least resources tend to not have the characteristics the military requires for enlistment. They may lack a high school diploma, have a criminal history, or be in poorer health.
  • If you want to study veterans, recognize that the population served by the VA may not be representative of the whole. Not all veterans qualify for VA healthcare, and of those who do qualify, not all choose to use it. VA uses a set of priority categories to determine eligibility for VA healthcare. These categories combine severity of service-connected disabilities and means testing. Users of VA services are by definition likely to be in Worse health and poorer than those who don’t because of the eligibility criteria. Other programs like education or home loan benefits have different eligibility criteria.

Suggested Readings:

%d bloggers like this: