Update II: Military Social Conservatives

Legacy blog posts Polls in the News

It gets more interesting.  Here is the promised update to the posts of the last two days, based on some fascinating analysis sent by Jason Dempsey, an Army Infantry officer and Columbia PhD candidate currently completing a teaching tour at West Point.  As part of his doctoral dissertation, Dempsey conducted a true random sample survey of the U.S. Army population in 2004.  The results show that Peggy Noonan’s comment that in her experience, “career military men” are rarely “social conservatives” reflects the reality of the Army population in some ways but not in others.   

Dempsey’s data confirm the findings I discussed yesterday showing that Army officers are far more conservative than the U.S. population, both in terms of their ideological self placement and their stands on specific “social” issues.  If by “career military” Noonan meant officers, then as Dempsey puts it, “that population is probably more socially conservative than she has experienced.” 

On the other hand, Dempsey finds that Army enlisted personnel, including those he classifies as “careerists,” have attitudes on political issues that largely resemble those of the larger U.S. population.   He also makes an important observation about the demographics of the enlisted Army personnel:

Enlisted soldiers also tend to be younger, much more diverse in terms of race and ethnicity, and come from more modest socioeconomic backgrounds than officers.  This correlates with more moderate or apolitical stances on larger political/social issues and less engagement in the political process. 

So in that sense, Noonan has a point. 

Dempsey drafted some analysis exclusively for MP readers that I have posted in full on the jump.  I have also created a PDF of his analysis suitable for printing.  Either way, it is well worth reading in full.  Thank you, Major Dempsey, for sharing it. 

PS:  See also this recent New Yorker “Talk of the Town” item on Dempsey and filmmaker Eugene Jarecki.

PPS:  A tip of the hat to GWU Professor John Sides, who originally alterted me to Dempsey’s research.

Jason Dempsey’s analysis:

Mark,

Having studied the social and political inclinations of members of the United States Army for some time I’m happy to comment on your reader’s question. 

First, you fairly accurately summed up the state of polls on the social and political attitudes of active members of the armed forces-they are few and far between. One you didn’t mention was the study conducted by the Triangle Institute for Security Studies (TISS) in the late 1990’s (http://www.poli.duke.edu/civmil/).  The TISS project was unique in that it got access to military schools and did some fairly extensive surveys of select groups of up-and-coming senior leaders.  It is worth looking at for an overview of the attitudes of higher-ranking officers across the various services.

In 2004 I was able to conduct a true random sample mail survey of the Army population focused primarily on the question of Hispanic integration that included questions on civic engagement, socialization, and attitudes towards social and political issues.  Everyone in the Army was eligible to be included in the survey with the exception of three groups 1) Generals and Command Sergeants Major (there are too few to guarantee confidentiality); 2) Privates (PV1 & PV2), who are notoriously hard to reach as they are predominantly still within their first year of service and are either in some form of training, moving between training, or moving to their first unit; and 3) those soldiers who were currently in a combat zone.  Due to the high turnover of the previous year excluding group three did not prevent us from surveying recent combat veterans.  Fully 376 of our 1189 respondents, or 32% of our sample, were veterans of either Operation Iraqi Freedom (2003-2004) or Operation Enduring Freedom in Afghanistan (2001-2004); and 143 indicated that they had been involved in direct ground combat in the previous two years.  Overall, the response rate for the survey was 45%.  All reported data is weighted to reflect the composition of the Army (minus excluded ranks) as of February 2004, when the sample was drawn.  Now, with the methodological disclaimers out of the way, on to your question…

First, I think that Noonan’s comment may have drawn more interest than usual because it is counterintuitive.  We tend to think of the military as an inherently ‘conservative’ institution.  And what is fascinating about her quote is that the way she describes the ‘career military man’s’ approach to such questions used to fit the definition of conservatism (an outlook that shunned government activism and approached such issues as “private and not subject to the movement of machines”). 

Indeed, Morris Janowitz (the father of military sociology) and Samuel Huntington (the father of civil-military relations theory) both described the conservative ethic, as practiced by the military in the 1950s and 1960s, as a distinctly nonpartisan means of political expression (See Janowitz, 1960/1971 p. 236, and Huntington, 1957).  That definition of conservatism clearly no longer holds so we just need to make sure that we are all comparing apples to apples in our approach.

In the Citizenship & Service survey the question I asked about conservative-liberal self-placement was worded, “In terms of politics and political beliefs, where would you place yourself?”  Respondents then had the option of placing themselves on the standard 7-point ‘extremely liberal’ to ‘extremely conservative’ scale.

On the whole, the Army looks remarkably like the civilian population on this question, as represented by the 2004 NES dataset (Note that this data is not directly comparable.  NES offered a “or have you not thought much about this” option that a large number of respondents chose.  The Army survey did not offer this option, and it might be assumed that a large number of the NES respondents who said they hadn’t thought much about the issue-had they been not given this option– would have chosen the moderate category):

However, the story gets a lot more complicated when you start breaking the Army up into key subgroups.  Among officers a clear majority, 63%, self-identify on the conservative end of the spectrum compared to only 32% of the enlisted ranks. (The overall numbers look the way they do because officers only make up about 14% of the Army).  This leads to the second question about Noonan’s statement:  does she consider a ‘career military man’ to be an officer, or anyone who commits to 20 years or more of uniformed service? 

Leaving open all possibilities, I present the following in terms of the entire Army, officers, enlisted, and ‘careerists’ (defined as anyone who has spent 15 or more years in the Army OR states that they will ‘definitely’ or ‘probably’ stay in the Army until retirement).  In the interest of brevity, I will leave out any discussion of differences among gender, racial and ethnic groups, but suffice to say that many of the same trends that one sees in the civilian population apply to the Army as well.

Among ‘careerists’, 21% self-identify as liberal while 40% identify as conservative.  This follows from a generally older and more established cohort (again pointing to the difficulty of broadly defining ‘conservatism’).

As for Noonan’s point, it would seem the Army data supports her anecdotal evidence, unless she was specifically referring to officers.

When it comes to the importance of religion in daily life, the Army data follows a similar pattern, with officers the most likely to say that religion provides ‘quite a bit’ or a ‘great deal’ of guidance.  Overall people in the Army are significantly less likely than civilians to state that religion provides ‘Quite a bit’ or a ‘Great deal’ of guidance in day-to-day living.   HOWEVER, the Army question is not directly comparable to the NES question because the NES question did not explicitly offer a ‘not religious’ or ‘no guidance’ option.

On a few ‘hot-button’ social issues I offer the following summary stats. I don’t have good comparisons with the civilian population handy at the moment, but the questions were worded as follows:

“Please indicate your position on the following domestic issues:

a. Banning the death penalty

c. Allowing prayer in public schools

d. Placing more restrictions on gun ownership

e. Outlawing abortion entirely”

Respondents were then given five options, Strongly Favor, Favor, Strongly Oppose, Oppose, and Don’t Know/No Opinion. 

Making rough comparisons from what we know historically from surveys of the general public, soldiers don’t appear to differ much on these issues (for example see Ben Page and Bob Shapiro’s, The Rational Public for an overview).

We do, however, have some comparable civilian data from the Chicago Council on Foreign Relations (CCFR) 2004 survey on preferences towards government spending on various social programs.  On these issues the Army is largely comparable to the civilian population, with officers again being the exception. 

Overall, the Army tends to track civilian attitudes and ideological self-placement.  Untold in this brief overview, however, are underlying stories of demographics in the Army.  The enlisted ranks drive overall attitudes due to their numbers relative to the officer corps.  Enlisted soldiers also tend to be younger, much more diverse in terms of race and ethnicity, and come from more modest socioeconomic backgrounds than officers.  This correlates with more moderate or apolitical stances on larger political/social issues and less engagement in the political process. 

However, I’d say that if by ‘career military men’ Noonan is referring to officers, then that population is probably more socially conservative than she has experienced.  At least by reported role of religion and ideological self-placement, the Army officer corps appears to exceed the rest of the Army and the civilian population in its ‘conservatism’.  This would make sense as the Army officer corps, and particularly its senior ranks, is predominantly white and male and comes from a middle- to upper-middle-class background (officers are required to have a Bachelor’s degree before commissioning).   Officers also deviate from the rest of the Army in terms of attitudes towards hot-button social issues, and they do appear more frugal when it comes to spending on programs related to social welfare (although some of this may be due to the Army’s retirement package, which might reduce an officer’s concern over the state of health care or Social Security).  A tighter definition of ‘social conservatism’ and more detailed multivariate analysis (as well as tests of statistical significance) are therefore clearly needed to take this question any further. 

Anyway, hope this sheds some light on an otherwise under-studied area. 

Anyone interested in more nuanced analyses of these and related questions is welcome to a paper presentation I will give at the American Political Science Association convention on September 1st.  Our paper on Army racial and ethnic attitudes, as they relate to Hispanic integration (co-authored with Bob Shapiro) should also be coming out in an edited volume from the Tomas Rivera Policy Institute this fall.  Additionally, we have several papers in the works on voting and partisanship (co-authored with Craig Cummings and Mat Krogulecki) that should get out this fall and winter.

Initial funding and support for this project was provided by the Tomas Rivera Policy Institute through the Institute for Social and Economic Research and Policy at Columbia University.  Additional funding was provided by the Academic Research Division of the United States Military Academy, the Saltzman Institute for War and Peace Studies and a fellowship grant from the Eisenhower Foundation.

Jason Dempsey is an Army Infantry officer currently completing a teaching tour at West Point.  This research is part of his doctoral dissertation in political science at Columbia University.  The views expressed in this article are those of the author and do not reflect the official policy or position of the United States Military Academy, the Department of the Army, or the Department of Defense or U.S. Government.

Mark Blumenthal

Mark Blumenthal is the principal at MysteryPollster, LLC. With decades of experience in polling using traditional and innovative online methods, he is uniquely positioned to advise survey researchers, progressive organizations and candidates and the public at-large on how to adapt to polling’s ongoing reinvention. He was previously head of election polling at SurveyMonkey, senior polling editor for The Huffington Post, co-founder of Pollster.com and a long-time campaign consultant who conducted and analyzed political polls and focus groups for Democratic party candidates.