Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2018 Aug 1.
Published in final edited form as: Psychol Trauma. 2016 Nov 28;9(Suppl 1):12–18. doi: 10.1037/tra0000191

Assessing Civilian Perceptions of Combat Veterans: An IAT Study

Cade Schreger 1, Matthew Kimble 2
PMCID: PMC5621655  NIHMSID: NIHMS815280  PMID: 27893266

Abstract

Objective

Evidence suggests that civilians may have considerable ambivalence to returning veterans. While civilians are frequently grateful for the service of military personnel, they can often be wary of the mental health and stability of returning veterans. If civilians do hold such negative biases towards veterans, whether implicit or explicit, this may have a significant impact on the ability of military personnel to integrate back into society. The goal of the study was to test whether participants held an implicit bias of mental instability towards veterans.

Method

In this study, forty-eight participants took an adapted version of the Implicit Association Test (IAT: Greenwald et al., 1998). Pictures of veterans and civilians were paired with words that either reflected mental stability (safe, sane, reliable, responsible) or instability (crazy, dangerous, unstable, unpredictable).

Results

The results demonstrated a finding of moderate effect size for an association between veterans and instability. Participants had significantly shorter reaction times for IAT blocks in which veteran pictures and negative words were paired.

Conclusions

This is the first study in the literature to demonstrate implicit biases of veterans as unstable. If implicit biases do in fact exist, the re- integration issues that veterans face may be due, at least in part, to a negative bias they face upon their return.

Keywords: Trauma, PTSD, Veterans, Bias, IAT, implicit association test

Literature Review

A difficult, yet understudied, aspect of the experience of the combat veteran is that which occurs after active duty and deployment – the process of assimilation back into civilian life. Prior self-assessment and demographic research has demonstrated high rates of veteran dissatisfaction with respect to their re-entry into civilian life (Taylor et al., 2011) as well as disproportionate rates of suicide (Brenner & Barnes, 2012), homelessness (Bureau of Justice Statistics, 2007; Veterans Inc, 2010), unemployment (Ostovary & Dapprich, 2011), and alcohol/substance abuse (Burnett-Ziegler et al., 2011). Most of this research has emphasized how characteristics of the veterans themselves, such as presence of depression or PTSD contribute to these issues (Morin, 2011). In contrast, there is a surprising lack of research concerning societal attitudes towards veterans and how this might contribute to difficulties assimilating.

When this type of work has been done, it has focused exclusively on explicit attitudinal data measures through self-reports. The most thorough and recent report was conducted by Taylor et al. (2011) for the Pew Charitable Trust and reports on the experiences and attitudes of 1,853 veterans (including 712 who were post 9/11 veterans) and 2,003 civilians. This work revealed that civilian respondents largely disapproved of the U.S.’ recent combat efforts in both Afghanistan and Iraq, and believed that over half of post-9/11 veterans suffer from posttraumatic stress disorder (PTSD). This figure is not only far higher than the 10-20% prevalence rate typically reported for PTSD specifically, but double the rate of any type of reported mental illness in returning veterans including PTSD, depression, and substance use (Tanielian & Jaycox, 2008; Dursa, Reinhard, Barth, S. K. & Schneiderman, 2014; Thomas, Wilk, Riviere, McGurk, Castro, & Hoge, 2010; Ramchand, Schell, Karney, Osilla, Burns, R. M. & Caldarone, 2010). Such discrepancies reveal the troubling truth that civilians may hold negative perceptions of veteran mental health that exceed actual concerns. This is particularly problematic in that studies have shown that a large proportion of the American population considers a diagnosis of PTSD to be somewhat synonymous with the characteristics of instability or labels such as “violent” or “crazy” (Mittal et al., 2013). This is likely to have significant functional impact on veterans as prior research has demonstrated that negative perceptions associated with PTSD have dissuaded veterans from seeking treatment for a host of combat-related issues (Dickstein et al., 2010).

There are now methods to assess attitudes towards veterans that are implicit rather than explicit. Implicit attitudes are beliefs or appraisals that an individual holds that they are not aware of. The advantage of assessing implicit attitudes regarding an issue is that implicit attitudes are less influenced by social desirability biases and yet still influence overt behaviors (Wicker, 1969; Cunningham & Zelazo, 2007). This body of work has spawned more integrative theories of attitudinal/perceptual formation and their effects on behavior, some of which emphasize the role of attitudinal fluidity, particularly in the case of varying contexts (see “Iterative Reprocessing Model”, Cunningham & Zelazo, 2007).

The emergence of the Implicit Association Test (IAT; Greenwald et al., 1998) has become a primary method for measuring implicit attitudes and perceptions. The IAT is a speed-based categorization task used to measure the associative strength between a given target-concept (e.g., black, white, male, female) and an attribute dimension (e.g., pleasant, unpleasant, positive, negative). The premise behind the IAT is that participants react quicker to attributes that match a given concept as compared to attributes that do not match the concept. The speed of reaction is thought to be a measure of associative strength between the concept and the attribute (Williams & Themanson, 2010). While there has been some debate about the interpretation of the “IAT effect,” most have come to the conclusion that it represents an implicit bias on issues related to prejudice, stereotype, or discrimination (see Oswald, Gregory, Hart, James & Tetlock, 2015 for a constrasting view). The bias is thought to be implicit as the effect has been reliably found even when explicit self-reports do not show an indication of bias (Nosek & Smyth, 2007). For example, in one of the earliest versions of the IAT which looked at bias based on skin color, numerous studies found that participants responded significantly quicker when images of dark-skinned people were paired with negative words than when the same faces were paired with positive words (see Williams & Themanson, 2010; Greenwald, Banaji, and Nosek, 2015 for review). This effect was reversed with light-skinned people. Similar biases have been revealed with respect to other categories such as “gay-straight,” “fat-thin,” and “young-old.” (Nosek, Greenwald, & Benaji, 2005)

Whether the public would have similar negative associations of veterans is an interesting question. This is due to the fact that civilian reports of veterans reflect some ambivalence. While on the one hand civilians are grateful to veterans for their service, there is also reason to believe that civilians may be wary and cautious of veterans because of perceived mental health issues (Taylor et al., 2011). In this study, participants completed a computerized IAT task with pictures of veteran models vs. pictures of civilians acting as the two target-discrimination categories. The words paired with these target pictures either represented mental stability or instability thus allowing for possible implicit associations between veterans and mental illness.

While the IAT has previously been utilized to analyze myriad associations and biases (for a review, see Williams & Themanson, 2010; Nosek & Smyth, 2007), none of these have investigated civilian perceptions of combat veterans. However, based on what little explicit literature that is present, and the exaggerated association between “veterans” and “PTSD,” we hypothesized that civilians would have a negative perception of combat veterans as unstable - an effect that should manifest itself through significantly shorter reaction times for IAT blocks in which veteran pictures and negative words were paired. Additionally, we expected to find no significant correlations between explicit and implicit measures of attitudes towards veterans, due to what should be a strong social desirability bias in participants. Thus, the present study was the first to attempt to analyze the role of implicit attitudes and perceptions of civilians towards combat veterans using an IAT paradigm.

Methods

Participants

Participants for this study were 48 students in an introductory psychology class. All participants received coursed credit. Participants completed an informed consent form before beginning the IAT task and the questionnaire. Additionally, eight photo participants were used for the image stimuli in this paradigm. Age and race varied among the eight photo participants in order to protect against confounding IAT biases. However, all photo participants were male to better reflect combat veterans as a group. Photo participants were compensated $10 for their time. Photo models dressed in both civilian and military clothing provided by the researchers in order to preserve the continuity of faces across conditions within a given task version (Figure 1). Digital images used in this study were taken using a Canon EOS T5 digital SLR camera.

Figure 1.

Figure 1

Open in a new tab

Examples of civilian and veteran model images (top row) above an image from a combined trial of the IAT paradigm. Photo models were dressed in both civilian and non civilian clothes. For counterbalancing purposes, half of the participants would have received a version of the task with a photo on the top left while the other would have received a version of the task with the rightmost picture. Below the two images, is an example as would be seen by a participant in which they might be asked to push a button as quickly as possible to categorize the image as a “veteran” or “civilian.” Other trials (as can be seen in Figure 2) might have the a word, such as “insane,” in the middle of the screen and the participant would have to categorize it as “negative” or “positive”.

Procedure

After completing the consent form, participants were seated ~3 feet from a Dell computer screen and presented with on-screen instructions. All participants were divided among the four task versions described below, such that an equal number of participants completed each version of the task (n=12). The computerized IAT was then administered using Superlab 4.0 (Superlab: Cedrus Corporation, San Pedro, California) on a Windows XP desktop computer and participants used a Cedrus RB-730 (Cedrus Corporation) response pad to record their categorizations.

For any given trials, the categories were continuously presented in the upper left and right corners of the screen and the target words/images were presented in the center of the screen, one at a time. See for example, the first screen of Figure 2 in which categorization words are shown on the top left and right and the picture is centered in the screen. The participant's task was to respond as quickly as possible with a left or right button press as to which category the picture or word presented in the middle of the screen fit with. On the example trial shown on the first screen of Figure 2, a participant would push the top left button to indicate that the picture was that of a veteran. If the participant provided an incorrect response, the word “incorrect” appeared 300 pixels below the target word or image for 200 ms and the participant then selected the correct category before the program automatically moved on to the next trial. Following correct categorization, the next target word/image was presented after a 50 ms interval.

Figure 2.

Figure 2

Open in a new tab

Visual schematic of the Veteran/Stability IAT. The top-leftmost image is an image from Block 1, where participants were asked to categorize solely images into 2 categories. The top-middle image is an image from Block 2, where participants were asked to categorize solely words into 2 categories. The top-rightmost picture is an image from Blocks 3/4, where participants were asked to categorize both words and images into all 4 categories. Participants might be asked during this block to categorize a picture as “veteran” or “civilian” or categorize a word as “positive” or “negative.” The bottom-left image is an image from Block 5, where participants were again asked to categorize images into 2 categories, however those categories had now switched sides from their positioning in Blocks 1, 3, and 4. The bottom-right image is an image from Blocks 6/7, where participants were asked to categorize both words and images into all 4 categories, however in their new configuration (words having flipped sides from Blocks 2-4). An incongruent trial is shown in the top rightmost image in which a “veteran” and “positive” are mapped onto the same response key. A congruent trial is shown on the bottom rightmost image in which “veteran” and “negative” are mapped onto the same response key.

The example illustrated above was the simplest block of trials. There were 7 blocks in all (see Figure 2). Block 2 had trials in which words (such as “insane” or “responsible”) were presented in the center of the screen and categorized as either “positive” or “negative” with a button press. In Blocks 3 and 4 (see top-rightmost image on Figure 2), the task became more complicated for the participant. Participants were asked to categorize both words and images into four possible categories. Some trials were “inconsistent” in which “veteran” and “positive” are mapped onto the same response key. On “consistent” trials “veteran” and “negative” were mapped onto the same response key. In Block 5, participants were again asked to categorize images into 2 categories, however those categories had now switched sides from their positioning in Blocks 1, 3, and 4 for the purposes of counterbalancing. In Blocks 6/7, participants were again asked to categorize both words and images into all 4 categories, however in a new counterbalanced configuration (words having flipped sides from Blocks 2-4). Data for analyses were taken from Blocks 3,4,6, and 7 with the mixed trials in which speeded responses between certain pictures and words would indicate stronger associations and implicit bias.

The eight target words were selected from the Corpus of Contemporary American English (COCA; Davies, 2008) and matched – across the two category types, positive and negative – for length and frequency in the English language. These were used for discrimination blocks. The words representing the two category types were “safe”, “reliable”, “responsible”, and “sane” for positive and “crazy”, “dangerous”, “insane”, and “unpredictable” for the negative category.

While the IAT procedure has been modified and adapted to fit the needs of numerous experiments since its original publication, the methods used in this research have been adapted from three studies in particular which sought to create a singular methodology that yields the greatest construct validity (Greenwald et al., 1998; Greenwald et al., 2003, Nosek et al., 2005). The target-concept image categories each consisted of 4 pictures and the attribute dimensions each contained 4 words to represent mental stability and instability – prior research has shown that categories of this size maintain internal validity and overall effect size (Nosek et al., 2005). Additionally, order of the test trial blocks were counterbalanced by participant to prevent against the IAT order effect (Greenwald et al., 1998). A visual schematic of this paradigm is shown in Figure 2. Upon completion of the IAT, participants filled a brief demographic form . This brief survey asked about basic demographics such as age, gender, race, history of military experience, and hometown. It also included Likert scales (from 1 to 11) regarding how “mentally stable” they felt civilians and veterans to be. Participants were then debriefed; the purpose of the task was explained and any questions that they might have were answered. Participants received course credit for their participation.

Behavioral Data Acquisition & Analysis

When the IAT was first published, IAT effects were computed by comparing the logarithmic values of the average response times for trials 3-40 for blocks 4 and 7 (the combined and reverse-combined test blocks) (Greenwald et al., 1998). However, several years later this scoring method was adapted into a new variable – D (not to be confused with Cohen's d) – that proved to have greater internal consistency, construct validity, and effect sizes (Greenwald et al., 2003). In comparison to the conventional method of IAT scoring, the adapted method not only includes the first 2 trials of blocks 4 and 7, but also includes all trials from blocks 3 and 6 (the combined/reverse-combined practice blocks). This method also includes different exclusion boundaries for excessively slow and fast responses (<400ms and >10,000 ms). In this study, participants for whom ≥20% of all trials were incorrect were discounted. Additionally, the D measure affords for greater power than conventional IAT scoring methods, such that a sample size of 39 participants yields a power of .80 to reject the null hypothesis when α = .05. On a final note, the D variable has been shown to reduce the contaminating effects of prior IAT exposure when compared to typical scoring methods (Greenwald et al., 2003). When possible, as basis for comparison, we also provide the more widely recognized Cohen's d as an estimate of effect size.

Results

Participants

In total, 48 participants completed the experiment, comprising of 16 males and 32 females. No participants were excluded due to poor behavioral data. Mean age of the participants was 19.0 (SD=.71). Thirty-six participants defined themselves as White/Caucasion, 7 as Asian, and 5 as either multi-racial or other. Twenty were from the northeast, 5 from the southeast, 5 from the Midwest, 1 from the southwest, 11 from the west, and 6 were international students. Ten of the participants had been exposed to some version of an IAT previously. Thirty-one of the participants (65%) had a first or second degree relative who had served in the military.

Explicit Self-Report

Participants reported a mean of 2.33 (SD=0.52) on a five points scale with respect to the “mental stability” of veterans. A mean of 2.33 corresponds closest to an endorsement that veterans are “somewhat more mentally unstable than civilians” and was between that endorsement and “neither more stable or unstable than civilians.”

IAT Effect/Behavioral Correlations

The average D score as well as t-tests between the congruent and incongruent blocks suggests a significant association between veterans and mental instability. A paired-samples t-test was conducted comparing the averages within participants between mean response times for the congruent and incongruent blocks. There was a significant difference, such that individuals responded more quickly in trial blocks where veteran images and negative words were mapped onto the same response key as opposed to differing keys, t(47) = −4.38, p < .001. (Figure 3). Among the 48 participants, the average D score – for which a positive value signifies an effect in the predicted direction was 0.33 (SD=.43). Cohen's d demonstrates an effect size of .61 for the comparison in which veteran images and negative words were mapped onto the same response key as opposed to differing keys.

Figure 3.

Figure 3

Open in a new tab

Paired samples t-test comparing mean reaction time between consistent (veteran/negative) and inconsistent (veteran/positive) blocks, t(47)= −4.38, p<.001, d=0.61.

Two Pearson's bivariate correlations were also performed between IAT effect and both means in order to assess the validity of presenting both implicit measures. Significant correlations were found between IAT effect and both congruent (r(48) = −.32, p = .03) and incongruent (r(48) = .50, p = <.001) mean reaction times.

A Pearson's bivariate correlation was also computed between explicit measures of attitude and IAT effect (D). There was no significant correlation between the explicit measure of veteran stability and the IAT effect (r(48) = −.01, p = .95).

Independent sample t-tests revealed that IAT effects were not impacted by sex, t(46)=0.21, p>.05) , political affiliation, t(41)=0.80, p>.05, family history of military service, t(46)=1.4, p>.05, or previous exposure to an IAT, t(46)=−1.1, p>.05).

Discussion

The data above provide evidence towards the initial hypothesis that participants would have an implicit bias of mental instability in combat veterans. Participants’ response times were significantly quicker when the categories “Veteran” and “Negative” were paired than when “Veteran” and “Positive” were paired, a result which suggests that there is a greater cognitive association between the “Veteran” and “Negative” categories (Greenwald et al., 1998). With respect to practical effect size, an IAT D score of 0.33 is typically considered to be a moderate effect (Nosek et al. 2005). This conclusion is supported by a Cohen's d score of 0.61 between consistent and inconsistent trials (Cohen, 1988). However, both of these measures of effect size are smaller than what are typically found in “Black-White” or “Gender-Science” IAT studies where an IAT D ranging from .45 to .54 would be more typical (Nosek, Greenwald, & Banaji, 2005). The effects in this study are closer to those found in Asian-White IATs and larger than negative biases found in “Native-White” comparisons. While this veteran effect is relatively moderate, it represents the first report of an IAT-based implicit bias towards veterans.

The findings of bias in this study were not impacted by demographic characteristics. Neither the IAT effect, nor the explicit ratings, were impacted by sex, political affiliation, previous exposure to an IAT, or family history of military service. This suggests that this IAT effect, as has been found to be the case in IATs of other characteristics, is fairly robust and may represent a broadly held societal value. This was true even among the participants who had military personnel in their family and who had had the personal experience of knowing a member of the military fairly well. It does not appear that close proximity to a member of the military decreases implicit bias or, if this is the case, it is counterbalanced by a subgroup of individuals for whom personal experience with a veteran increases implicit bias in instability.

The fact that the IAT effect did not correlate significantly with the explicit measure of stability is consistent with much of the existing literature and suggests that the IAT may be tapping into biases held by a subset of participants who are either unwilling to report it, unaware of an unconscious bias, or both. This result was expected, as one would predict some social desirability effects to emerge. While the likelihood of a Type II error was increased due to the difference in measurement specificity between IAT score and Likert scale data, the robustness of the results make this conclusion highly unlikely. Additionally, the significant correlations between D score and both consistent and inconsistent mean reaction times provide backing for the validity of utilizing both measures, particularly with respect to the directionality of the effects. D score was positively correlated to inconsistent block reaction time while negatively correlated to consistent block reaction time.

If an implicit bias in citizens in fact exists, it may play a role in the challenges military personnel face when returning to civilian life. Often the struggles of returning veterans have been attributed to veteran behavior, specific symptoms, or internal characteristics (i.e., PTSD, aggression, substance use). Far less attention has been paid to societal influences. Few have discussed how individuals in society may hold biases of veterans that make readjustment more difficult. These would be particularly difficult to identify if the biases are implicit. From related literature, we know that bias in general, and discrimination in particular, can result in lost opportunities for housing and work and is known to result in poorer quality health care provision and differential treatment in the criminal justice system (Mittal et al., 2013). Mental health stigma among military personnel may also prevent veterans from seeking treatment after deployment (Ghaffari, 2011; Majette, 2013; Flick, 2011). Some combination of these factors may have a negative impact on veterans upon their return.

Limitations

Our sample consisted of college undergraduates. While the sample was fairly geographically diverse, with 58% of the sample coming from outside the northeast, the average age sample was 19 and therefore is limited in its diversity with respect to education, military exposure, life experience, and social economic status. This data from undergraduates can not be generalized to age matched non university students or adult communities with a larger proportion of veterans. While there have not generally been generational effects reported on IATs, and they appear to be fairly robust in that that they represent largely internalized societal values, this paradigm would need to be implemented in older samples for comparison.

This study was concerned with the possible effects of task version on IAT effect. Four versions of our paradigm were administered equally among the 48 participants. Two were used to counterbalance against the possibility of certain faces being perceived as inherently more “stable/unstable” than others and the other 2 were used to additionally counterbalance against the IAT order effect. Interestingly, no significant differences were found between IAT scores of any task version. This is likely the result of the implementation of the D measure to assess IAT effect as the D measure has been shown to drastically reduce the magnitude of the order effect when compared to conventional measures of IAT analysis (Greenwald et al., 1998; Greenwald et al., 2003). This result comes as a beneficial surprise as it suggests that use of a single version of the task may be sufficient for subsequent implementation.

Future Directions

This study is additionally valuable in that it can be easily adapted to study other aspects of biases regarding veterans. As mentioned earlier, this study could be implemented in different generational groups. Furthermore, this task could also be presented following a prime – such as a pro-veteran service announcement or anti-veteran news story to assess how certain proximally-encountered stimuli might affect one's attitudes and perceptions. In addition, it would be valuable to assess whether perceptions change as a function of the regalia that the veterans are asked to wear. In this study, all models were dressed in fatigues. Would the IAT effects be nullified if veterans were presented in formal regalia with medals? If this were the case, would implicit biases often seen against African Americans be inverted if they were placed in formal military regalia?

As a whole, this study yielded some valuable findings in a largely understudied area. The data acquired in this study demonstrate a moderate sized implicit bias of mental instability in veterans. While these findings are preliminary, similar studies could serve to promote greater awareness among the general populace and scientific community towards the existence of a negativity bias towards veterans. Such information could lead to strategies to overcome biases to improve the process of reassimilation.

Supplementary Material

1

Acknowledgments

The project was funded in part by an AREA Grant (R15 MH081276-03) from the National Institute of Mental Health to Matthew Kimble

Contributor Information

Cade Schreger, Program in Neuroscience, Middlebury College.

Matthew Kimble, Department of Psychology, Middlebury College.

References

  1. Booth-Kewley S, Larson GE, Highfill-McRoy RM, Garland CF, Gaskin TA. Factors Associated With Antisocial Behavior in Combat Veterans. Aggressive Behavior. 2010;36:330–337. doi: 10.1002/ab.20355. [DOI] [PubMed] [Google Scholar]
  2. Brenner LA, Barnes SM. Facilitating Treatment Engagement During High-Risk Transition Periods: A Potential Suicide Prevention Strategy. American Journal of Public Health. 2012;102(Suppl 1):S12–S14. doi: 10.2105/AJPH.2011.300581. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Burnett-Ziegler I, Ilgen M, Valenstein M, Zivin K, Gorman L, Blow A, Duffy S, Chermack S. Prevalence and correlates of alcohol misuse among returning Afghanistan and Iraq Veterans. Addictive Behaviors. 2011;36(8):801–806. doi: 10.1016/j.addbeh.2010.12.032. [DOI] [PubMed] [Google Scholar]
  4. Corrigan PW, Watson AC. The paradox of self-stigma and mental illness. Clinical Psychology: Science and Practice. 2002;9(1):35, 53–35, 53. doi: http://dx.doi.org/10.1093/clipsy/9.1.35. [Google Scholar]
  5. Cunningham WA, Zelazo PD. Attitudes and evaluations: a social cognitive neuroscience perspective. Trends in Cognitive Science. 2007;11(3):97–104. doi: 10.1016/j.tics.2006.12.005. [DOI] [PubMed] [Google Scholar]
  6. Davies M. The Corpus of Contemporary American English: 450 million words, 1990-present. 2008 Available online at http://corpus.byu.edu/coca.
  7. Dickstein BD, Vogt DS, Handa S, Litz BT. Targeting Self-Stigma in Returning Military Personnel and Veterans: A Review of Intervention Strategies. Military Psychology. 2010;22:224–236. [Google Scholar]
  8. Drake B. On Memorial Day, public pride in veterans, but at a distance. 2013 May 24; Retrieved from Pew Research Center, FactTank site: http://www.pewresearch.org/fact-tank/2013/05/24/on-memorial-day-public-pride-in-veterans-but-at-a-distance-2/
  9. Dursa EK, Reinhard MJ, Barth SK, Schneiderman AI. Prevalence of a Positive Screen for PTSD Among OEF/OIF and OEF/OIF-Era Veterans in a Large Population-Based Cohort. Journal of Traumatic Stress. 2014;27:542–549. doi: 10.1002/jts.21956. [DOI] [PubMed] [Google Scholar]
  10. Flick JB. 2011. A conceptualization of treatment stigma in returning veterans (Psy.D.). Available from ProQuest Dissertations & Theses Global. (904416701)
  11. Ghaffari AS. 2011. Stigma and psychological help-seeking of veterans (Ph.D.). Available from ProQuest Dissertations & Theses Global. (904553398)
  12. Granello DH, Pauley PS, Carmichael A. Relationship of the Media to Attitudes Toward People With Mental Illness. Journal of Humanistic Counseling, Education, and Development. 1999;38:98–110. [Google Scholar]
  13. Greenwald AG, McGhee DE, Schwartz JLK. Measuring Individual Differences in Implicit Cognition: The Implicit Association Test. Journal of Personality and Social Psychology. 1998;74(6):1464–1480. doi: 10.1037//0022-3514.74.6.1464. doi:10.1037/0022-3514.74.6.1464. [DOI] [PubMed] [Google Scholar]
  14. Greenwald AG, Nosek BA, Banaji MR. Understanding and Using the Implicit Association Test: I. An Improved Scoring Algorithm. Journal of Personality and Social Psychology. 2003;85(2):197–216. doi: 10.1037/0022-3514.85.2.197. doi:10.1037/h0087889. [DOI] [PubMed] [Google Scholar]
  15. Greenwald AG, Uhlmann EL, Poehlman TA, Banaji MR. Understanding and Using the Implicit Association Test: III. Meta-Analysis of Predictive Validity. Journal of Personality and Social Psychology. 2009;97(1):17–41. doi: 10.1037/a0015575. [DOI] [PubMed] [Google Scholar]
  16. Greenwald AG, Mahzarin R, Banaji M, Nosek B. Statistically small effects of the implicit association test can have societally large effects. Journal of Personality and Social Psychology. 2015;108(4):553–561. doi: 10.1037/pspa0000016. [DOI] [PubMed] [Google Scholar]
  17. Hoit K. The “Dangerous” Veteran: An Inaccurate Media Narrative Takes Hold. 2012 Retrieved from Vantage Point: Official Blog of the U.S. Department of Veterans Affairs: http://www.blogs.va.gov/VAntage/6026/the-“dangerous”-veteran-an-inaccurate-media-narrative-takes-hold/
  18. Majette TD. 2013. Self-stigma as a barrier to seeking mental health care for veterans (Ph.D.). Available from ProQuest Dissertations &amp; Theses Global. (1430909473)
  19. Mittal D, Drummond KL, Blevins D, Curran G, Corrigan P, Sullivan G. Stigma associated with PTSD: perceptions of treatment seeking combat veterans. Psychiatric Rehabilitation Journal. 2013;36(2):86–92. doi: 10.1037/h0094976. [DOI] [PubMed] [Google Scholar]
  20. Morin R. The Difficult Transition from Military to Civilian Life. 2011 Dec 8; Retrieved from Pew Research Center, Social Trends site: http://www.pewsocialtrends.org/2011/12/08/the-difficult-transition-from-military-to-civilian-life/
  21. Mumola CJ, Noonan ME. Veterans in State and Federal Prison. 2007 2004 Apr 29; Retrieved from Bureau of Justice Statistics: http://www.bjs.gov/index.cfm?ty=pbdetail&iid=808.
  22. Myers DG. Social Psychology. 3rd ed. McGraw Hill; New York: 1990. [Google Scholar]
  23. Newport F. U.S., 24% of Men, 2% of Women Are Veterans. 2012 Retrieved from Gallup Polling, Politics section: http://www.gallup.com/poll/158729/men-women-veterans.aspx.
  24. Nosek BA, Banaji MR, Greenwald AG. Harvesting Implicit Group Attitudes and Beliefs From a Demonstration Web Site. Group Dynamics: Theory, Research, and Practice. 2002;6(1):101–115. [Google Scholar]
  25. Nosek BA, Greenwald AG, Banaji MR. Understanding and Using the Implicit Association Test: II. Method Variables and Construct Validity. Personality and Social Psychology Bulletin. 2005;31(2):166–180. doi: 10.1177/0146167204271418. doi:10.1177/0146167204271418. [DOI] [PubMed] [Google Scholar]
  26. Nosek BA, Smyth FL. A multitrait-multimethod validation of the Implicit Association Test: Implicit and Explicit attitude are related but distinct constructs. Experimental Psychology. 2007;54:14–29. doi: 10.1027/1618-3169.54.1.14. [DOI] [PubMed] [Google Scholar]
  27. Ostovary F, Dapprich J. Challenges and Opportunities of Operation Enduring Freedom/Operation Iraqi Freedom Veterans with Disabilities Transitioning into Learning and Workplace Environments. New Directions for Adult and Continuing Education. 2011;132:63–73. [Google Scholar]
  28. Oswald FL, Gregory M, Hart B, James J, Tetlock P. Using the IAT to predict ethnic and racial discrimination: Small effect sizes of unknown societal significance. Journal of Personality and Social Psychology. 2015;108(4):562–571. doi: 10.1037/pspa0000023. [DOI] [PubMed] [Google Scholar]
  29. Oswald FL, Gregory M, Hart B, James J, Tetlock P. Predicting ethnic and racial discrimination: A meta-analysis of IAT criterion studies. Journal Of Personality And Social Psychology. 2013;105(2):171–192. doi: 10.1037/a0032734. [DOI] [PubMed] [Google Scholar]
  30. Phelan JC, Link BG, Stueve A, Pescosolido BA. Public Conceptions of Mental Illness in 1950 and 1996: What Is Mental Illness and Is It to be Feared? Journal of Health and Social Behavior. 2000;41(2):188–207. [Google Scholar]
  31. Psimas D, Crawford I. Post 9/11 Veterans Face Continuing Misperceptions on Homecoming. 2012 Jun 21; Retrieved from gotyour6: http://www.gotyour6.org/post-911-veterans-face-continuing-misperceptions-on-homecoming/
  32. Rabak-Wagener J, Eickhoff-Shemek J, Kelly-Vance L. The Effect of Media Analysis on Attitudes and Behaviors Regarding Body Image Among College Students. Journal of American College Health. 1998;47(1):29–35. doi: 10.1080/07448489809595616. [DOI] [PubMed] [Google Scholar]
  33. Ramchand R, Schell TL, Karney BR, Osilla KC, Burns RM, Caldarone LB. Disparate prevalence estimates of PTSD among service members who served in Iraq and Afghanistan: Possible explanations. Journal of Traumatic Stress. 2010;23:59–68. doi: 10.1002/jts.20486. [DOI] [PubMed] [Google Scholar]
  34. Tanielian T, Jaycox L, editors. Invisible Wounds of War: Psychological and Cognitive Injuries, Their Consequences, and Services to Assist Recovery. RAND Corporation; Santa Monica, CA: 2008. [Google Scholar]
  35. Taylor P, et al. The Military-Civilian Gap: War and Sacrifice in the Post 9/11 Era. 2011 Oct 5; Retrieved from Pew Research Center, Social and Demographic Trends project site: http://www.pewsocialtrends.org/files/2011/10/veterans-report.pdf.
  36. Thomas JL, Wilk JE, Riviere LA, McGurk D, Castro CA, Hoge CW. Prevalence of Mental Health Problems and Functional Impairment Among Active Component and National Guard Soldiers 3 and 12 Months Following Combat in Iraq. Archives of General Psychiatry. 2010;67(6):614–623. doi: 10.1001/archgenpsychiatry.2010.54. [DOI] [PubMed] [Google Scholar]
  37. Veterans Inc. Statistics. 2010 Retrieved from Veterans Inc.: http://www.veteransinc.org/about-us/statistics/#homelessness.
  38. Veterans Administration 2012 http://www.blogs.va.gov/VAntage/6026/the-%E2%80%9Cdangerous%E2%80%9D-veteran-an-inaccurate-media-narrative-takes-hold/
  39. Villani S. Impact of Media on Children and Adolescents: A 10-Year Review of the Research. Journal of the American Academy of Child & Adolescent Psychiatry. 2001;40(4):392–401. doi: 10.1097/00004583-200104000-00007. [DOI] [PubMed] [Google Scholar]
  40. Wicker AW. Attitudes versus Actions: The Relationship of Verbal and Overt Behavioral Responses to Attitude Objects. Journal of Social Issues. 1969;25(4):41–78. [Google Scholar]
  41. Williams JK, Themanson JR. Neural correlates of the implicit association test: evidence for semantic and emotional processing. Social Cognitive and Affective Neuroscience. 2010;6(4):468–476. doi: 10.1093/scan/nsq065. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS815280-supplement-1.docx (522.9KB, docx)

RESOURCES