Abstract
Objectives. We hypothesized that chronic stressors associated with an everyday social role (work) would interact with a traumatic macrosocial stressor (the terrorist attacks of September 11, 2001) in predicting mental health status (during the fall of 2001).
Methods. We used mail surveys returned as part of wave 3 of a workplace cohort study, both before and after September 11, 2001, to assess decision latitude, sexual harassment, generalized workplace abuse, psychological distress, and alcohol use. We also used regression analyses to assess the main effect of September 11 and interactions between September 11 and stressors, after control for baseline mental health.
Results. The main effect of September 11 on elevated alcohol use was significant for women but not for men. For women, work stressors significantly interacted with experiencing the events of September 11 to affect alcohol use and anxiety outcomes.
Conclusions. Women experiencing chronic work stressors were most vulnerable to elevated psychological distress and alcohol use after September 11, 2001.
Studies of the effects of the events of September 11, 2001, on New Yorkers and residents of surrounding communities,1 as well as on national samples,2–4 have demonstrated various mental health outcomes (e.g., depression, anxiety, posttraumatic stress disorder, increased alcohol consumption). In addition, demographic risk factors for experiencing negative outcomes related to September 11 have been identified (female gender, low education level, single marital status).5 However, much of this research has been cross-sectional in design, making the determination of causality between the experience of September 11, 2001, and mental health consequences problematic. In addition, the effects of experiencing September 11 have not been studied in conjunction with other ongoing life stressors, especially those that encompass other elements of lack of control over life experiences.
We derived this report from a longitudinal workplace cohort study. The wave 3 mail survey was carried out in the fall of 2001, with some respondents returning the survey pre–September 11 and other respondents returning the survey post–September 11. We assessed the main effect of experiencing September 11–related events on men and women and examined the interaction of these events with 3 chronic stressors that could be viewed as manifestations of lack of control—in this case, with respect to an everyday social role, that of work. These aspects of lack of control over work include low decision-making latitude and sexual harassment and generalized workplace abuse.
We derived this study from the stress paradigm delineating the impact of psychosocial stressors on mental health status. In particular, we addressed the combined effects on participants of a traumatic macrosocial stressor6 and everyday stressful experiences in the workplace, which are often chronic in nature7 and which may be viewed as being low in controllability.8 Moreover, the events of September 11 have been viewed as constituting an apocalyptic moment, producing feelings of powerlessness and victimization on a massive scale,9 in contrast to other feelings of powerlessness generated by everyday experiences of interpersonal victimization in the work-place, such as those generated by sexual harassment and generalized workplace abuse.7,10 In addition, stress researchers have depicted the negative consequences of cumulative adversity wherein acute events combine with chronic or earlier stressors to have an adverse effect on psychological well-being.11,12 We thus hypothesized that there would be negative mental health consequences resulting from the joint effect of experiencing September 11 and ongoing stressors involving other sources of lack of control, such as those experienced in the workplace. In particular, we hypothesized that individuals who experienced September 11 would be at greater risk for deleterious mental health outcomes if they also were subjected to everyday chronic stressors. However, given the greater vulnerability of women to psychological distress following September 11 found in previous studies,2,3,5 our central interest was to explore the extent to which, for women, chronic stressors further contributed to the deleterious effects of experiencing September 11.
METHODS
Sampling and Data Collection
Data were obtained from an ongoing longitudinal mail survey of employees initially selected from an American Midwestern urban university during the fall semester of 1996. The sample was stratified by gender and occupation into 8 groups. Initial wave 1 occupational groups included faculty, graduate student workers/trainees, clerical/secretarial workers, and service/maintenance workers. Employees (2416 men and 2416 women) were sampled from the university payroll database. Data collection used Dillman’s13 total design method for mail surveys, but additional follow-up strategies were used as well (supplementary reminder postcards, 2 additional mailings, reminder e-mail messages, and follow-up telephone calls). Respondents were compensated for their time and effort ($20 at waves 1 and 2 and $30 at wave 3). The final wave 1 sample comprised 2492 employees (response rate = 52%). The lower-than-desired response rate was typical of questionnaires that are self-administered and contain highly sensitive material and identifiers for subsequent tracking.14 Comparison of the sample with known characteristics of the total population indicated no significant racial/ethnic differences within each occupational stratum. Gender differences between this sample and the total population were also very small and insignificant for 2 of the 4 strata (service workers and student trainees). However, men were overrepresented in the clerical group by a difference of 8.3%, and women were overrepresented in the faculty group by a difference of 11.3%.15
One year after wave 1 (during the fall semester of 1997), the wave 1 respondents were resurveyed, producing a sample of 2038 employees, with an 82% retention rate. Five years after wave 1 (during the fall semester of 2001), the sample was again surveyed, producing a sample of 1730, with a 70% retention rate of eligible respondents from the wave 1 survey (e.g., minus those who had died). Wave 3 responders did not differ significantly from wave 3 dropouts in levels of sexual harassment, workplace abuse, or decision-making latitude at wave 1. Nor did they differ on 2 of the 3 relevant indices of alcohol consumption (quantity of consumption and escapist motives for alcohol use), but they drank more frequently at wave 1 (5.3 vs 4.3 days per month, P < .01). In addition, wave 3 completers manifested significantly lower levels of depression and anxiety symptoms (P < .01) compared with dropouts. Wave 3 completers were slightly more likely to be older (mean age 42 vs 37 years, P < .001) and White (75% vs 55% Asian, 71% African American, and 59% Hispanic) than dropouts. Because wave 3 respondents’ job status varied in terms of whether they were still employed in their original university jobs or were employed at other positions outside the initial university setting, we did not carry out between the sample and the general university population comparisons.
Measures
Low decision-making latitude was measured with the 9-item scale of the Job Content Questionnaire.16 The Decision Latitude sub-scale has 2 highly correlated components: Decision Authority, which assesses the degree to which one perceives the freedom to make decisions and choice of how the job is done, and Skill Discretion, which measures the extent to which the job involves learning new things, developing skills, variety, creativity, and lack of repetitiveness. The α coefficients were .86 for both men and women.
Sexual harassment was measured with a modified version of the Sexual Experiences Questionnaire (SEQ),17 reworded to make items applicable to both men and women. The modified SEQ contains 19 items that behaviorally depict 3 types of sexual harassment measured by 6 items each: gender harassment, unwanted sexual attention, and sexual coercion. Gender harassment encompasses crude sexual comments or comments that demean the target’s gender. Unwanted attention encompasses unwelcome touching and repeated requests for dates. Sexual coercion involves demands for sexual favors, which imply job-related consequences. One item additionally assessed sexual assault. Respondents rated each experience in their current job during the past year as occurring never, once, or more than once. The overall α coefficients were .82 for both men and women.
Generalized workplace abuse was measured by a 29-item instrument developed from transcripts of focus groups conducted, separately for men and women, with representatives of each of the occupational groups later surveyed in the study (see Rospenda et al.18 for a description of the focus groups). This instrument assesses 5 dimensions of workplace abuse: verbal aggression, disrespectful behavior, isolation/exclusion, threats/bribes, and physical aggression. Verbal aggression (9 items) consists of hostile verbal exchanges involving yelling, swearing, and the like. Disrespectful behavior (9 items) encompasses demeaning experiences such as being humiliated publicly or being talked down to. Isolation/exclusion (5 items) involves having one’s work contributions ignored or being excluded from important work activities. Threats/bribes (3 items) encompasses subtle or obvious bribes to do things deemed wrong or threatened with retaliation for failing to do such things. Physical aggression (3 items) involves being hit, pushed, or grabbed. Experiences were rated in a manner similar to that of the SEQ items. The overall α coefficients were .92 for both men and women.
Experiences in the 2 instruments were scored positively if they occurred more than once, with the exception of sexual coercion, sexual assault, and physical aggression (which, given their severity, were scored positively if they occurred only once). Respondents were categorized as having been harassed or abused during the past year based on the above rules. This scoring method is derived from research on perceptions of sexual harassment indicating that people perceive behaviors involving sexual coercion and assault as more serious than hostile-environment manifestations of sexual harassment.19 In addition, given empirical evidence for the distinctiveness of the sexual harassment and workplace abuse instruments,20 they are used as measures of separate phenomena.
We focused on the mental health outcomes involving depression, anxiety, and changes in alcohol use, which have shown an association with the experience of September 11 in previous studies. However, we excluded measures in the data set that assessed problem-related alcohol use and alcohol use to intoxication, because these measures covered the past-year time frame and thus we would be unable to determine whether these alcohol use–related outcomes occurred before or after September 11. Depressive symptomatology occurring during the past week was measured on the basis of 7 items from the Center for Epidemiologic Studies Depression Scale, with a scale range of 0 to 21.21 The α coefficients were .84 for men and .86 for women. Anxiety during the past week was measured by means of the 9-item tension/anxiety factor of the Profile of Mood States, with a score range of 0 to 36.22 The α coefficients were .80 for men and .82 for women.
Alcohol-related outcomes included frequency of consumption, quantity of consumption, and escapist motives for alcohol use. Frequency was assessed on the basis of response to the question, During the last 30 days, about how many days did you drink any type of alcoholic beverage? Quantity was assessed on the basis of response to the question, During the last 30 days, how many drinks did you usually have per day? Escapist motives for alcohol use were measured with 5 items developed by the Alcohol Research Group23: I drink for (1) cheering up, (2) to escape, (3) when tense, (4) to forget things, and (5) to forget worries. The αcoefficients were .88 for both men and women.
Analyses
Given our focus on the effects of September 11, 2001, a variable was created to differentiate respondents in terms of when they returned their questionnaire during the wave 3 data collection period. We used September 18, 2001, as the cutoff on the basis of our best estimate of the lag between when the questionnaires were mailed and when they were subsequently logged into the data collection system. We then used χ2 tests and analyses of variance to contrast men and women in our sample in terms of sociodemographic characteristics, stressor characteristics (the experience of September 11, decision latitude at work, and experiences of sexual harassment and workplace abuse), and mental health at wave 3 (depressive and anxiety symptomatology, quantity and frequency of alcohol use, and escapist motives for alcohol use). Next, we used linear regression analyses to examine, for men and women separately, the main effect of experiencing September 11 on each mental health variable at wave 3, controlling for age, race/ethnicity, education, and the relevant wave 1 baseline mental health variable. Finally, we used linear regression analyses to examine, for men and women separately, interactions between experiencing September 11 and each of the work characteristics (decision latitude, sexual harassment, and generalized workplace abuse) as they affected mental health outcomes, controlling for sociodemographic variables and baseline mental health. Because our outcome variables were skewed, we also ran all analyses using square root transformations. Because the directionality and significance of the findings remained the same, the nontransformed findings are shown for greater ease of interpretation.
RESULTS
We contrast men and women in the sample in terms of sociodemographic factors, stressors, and mental health status at wave 3, assessed during the fall of 2001 (Table 1 ▶). Men were more likely to be White, whereas women were more likely to be African American; in addition, men had attained a higher education level and were significantly older. In terms of stressors, there was no significant difference between genders in terms of having completed the wave 3 questionnaire before versus after September 11. By contrast, men experienced greater sexual harassment during the past year, whereas women experienced greater generalized workplace abuse. In terms of mental health outcomes, the data show gender differences that are generally found in community samples: men manifested more frequent and heavier alcohol consumption, whereas women were higher on depressive and anxiety symptomatology.
TABLE 1—
Sociodemographic, Stressor, and Mental Health Characteristics of Wave 3 Participants
| Wave 3 Respondents (n = 1594) | ||||
| Men (n = 724), No. (%) | Women (n = 870), No. (%) | Men (n = 724), Mean No. (SD) | Women (n = 870), Mean No. (SD) | |
| Sociodemographic characteristics | ||||
| Age, mean, y | ...(...) | ...(...) | 42.6 (12.4) | 40.7 (10.8)** |
| Race/ethnicity | ||||
| African American | 104 (14.5) | 224 (26.2)*** | ...(...) | ...(...) |
| Asian/Pacific Islander | 114 (15.9) | 72 (8.4) | ...(...) | ...(...) |
| Hispanic | 41 (5.7) | 62 (7.3) | ...(...) | ...(...) |
| White/other | 458 (63.9) | 496 (58.1) | ...(...) | ...(...) |
| Education | ||||
| Less than high school graduate; | 70 (9.7) | 111 (12.8)*** | ...(...) | ...(...) |
| high school graduate or | ||||
| General equivalency diploma; | ||||
| technical or trade school | ||||
| Some college or college graduate | 116 (16.1) | 220 (25.4) | ...(...) | ...(...) |
| Some graduate school or Completed graduate school | 534 (74.2) | 534 (61.7) | ...(...) | ...(...) |
| Stressor characteristics | ||||
| Survey return date | ||||
| Before September 11 | 561 (77.5) | 696 (80.0) | ...(...) | ...(...) |
| After September 11 | 163 (22.5) | 174 (20.0) | ...(...) | ...(...) |
| Work environment in wave 3 | ||||
| Low decision latitude | ...(...) | ...(...) | 38.4 (6.5) | 37.7 (6.4) |
| Experienced sexual harassment | 190 (29.7) | 201 (25.3)* | ...(...) | ...(...) |
| Experienced generalized workplace abuse | 322 (53.3) | 447 (60.2)** | ...(...) | ...(...) |
| Mental health characteristics | ||||
| Depression | ...(...) | ...(...) | 3.4 (3.9) | 4.0 (4.5)** |
| Anxiety | ...(...) | ...(...) | 6.4 (5.5) | 7.3 (6.6)** |
| Drinking frequency | ...(...) | ...(...) | 8.34 (8.9) | 5.3 (7.1)*** |
| Drinking quantity | ...(...) | ...(...) | 1.87 (1.2) | 1.6 (1.0)*** |
| Escape motives for drinking | ...(...) | ...(...) | 7.1 (2.9) | 7.0 (3.1) |
* P < .05; **P < .01; ***P < .001.
The one main effect manifested by women but not men as a consequence of experiencing September 11 was that women completing the survey after September 11 used more alcohol than did women completing the survey before September 11 (Table 2 ▶). There were no differences on any of the other variables for women, and men showed no effects of experiencing September 11.
TABLE 2—
Main Effect of September 11
| Survey Returned After September 11 | Alcohol Use Quantity, β(SE) |
| Men only | − 0.04 (0.12) |
| Women only | 0.28 (0.10)* |
Note. Survey returned before September 11 served as the referent category (coded 0). Control variables included wave 1 alcohol use quantity, age, race/ethnicity, and education.
* P < .01.
When we examined the interaction effects of experiencing September 11 in conjunction with the 3 work stressors, the data again showed that there were no significant effects on any of the mental health outcomes for men (Tables 3 ▶ and 4 ▶). By contrast, for women, decision latitude interacted with experiencing September 11 in relation to quantity of alcohol consumed, escapist alcohol use motives, and anxiety (Table 3 ▶). Moreover, experiences of sexual harassment and work-place abuse interacted with exposure to September 11 in relation to quantity of alcohol consumption. We show the nature of the interaction effects by means of the comparative mean scores on the outcomes, looking at the combination of pre–September 11 versus post–September 11 occurrence with low versus high decision latitude and presence versus absence of sexual harassment and generalized workplace abuse (Table 4 ▶). The major commonality across outcomes involves the combination of experiencing September 11 and the lack of control over the work role (as manifested by either low decision latitude, sexual harassment, or generalized workplace abuse), leading to the highest scores on quantity of alcohol consumption, escapist motives for alcohol use, and anxiety. However, other patterns varied. In the case of decision latitude, experiencing September 11 appeared to make little difference for the high-decision-latitude group in terms of predicting quantity of alcohol consumption. Similarly, for the high-decision-latitude group, experiencing September 11 had little effect on escapist alcohol use motives. The final finding regarding decision latitude was that the high-decision-latitude group had the lowest level of anxiety even after experiencing September 11. In terms of sexual harassment, experiencing September 11 did not affect quantity of alcohol use when sexual harassment was absent. Finally, in the absence of generalized workplace abuse, quantity of alcohol use was slightly lower after experiencing September 11. In summary, experiencing September 11 was most salient as a predictor of alcohol use outcomes and anxiety when it was coupled with chronic work stress involving lack of control over various aspects of the work role.
TABLE 3—
Interaction Effects of Work Stressors and September 11
| Alcohol Use Quantity, β(SE) | Escape Motives for Alcohol Use, β (SE) | Anxiety, β (SE) | |
| Men only | |||
| Decision latitude (job stress) | −0.01 (0.01) | −0.02 (0.02) | 0.00 (0.04) |
| Survey return date before vs after September 11 | −0.30 (0.90) | −2.87 (1.62) | −0.31 (3.50) |
| Survey return date × decision latitude | 0.01 (0.02) | 0.07 (0.04) | 0.00 (0.09) |
| Sexual harassment | 0.165 (0.12) | 0.59 (0.24)* | 1.51 (0.51)** |
| Survey return date before vs after September 11 | −0.02 (0.15) | −0.39 (0.29) | −0.46 (0.60) |
| Survey return date × sexual harassment | −0.06 (0.25) | 0.12 (0.52) | 0.25 (1.06) |
| Generalized workplace abuse | 0.02 (0.12) | 0.40 (0.23) | 1.14 (0.50)* |
| Survey return date before vs after September 11 | −0.12 (0.20) | −0.06 (0.37) | −0.53 (0.78) |
| Survey return date × generalized workplace abuse | 0.14 (0.25) | −0.55 (0.48) | 0.27 (1.03) |
| Women only | |||
| Decision latitude (job stress) | 0.01 (0.01) | 0.02 (0.02) | 0.00 (0.04) |
| Survey return date before vs after September 11 | 3.16 (0.70)*** | 4.77 (1.60)** | 7.81 (3.82)* |
| Survey return date × decision latitude | −0.07 (0.02)*** | −0.12 (0.04)** | −0.24 (0.10)* |
| Sexual harassment | 0.03 (0.10) | 0.05 (0.25) | 3.01 (0.56)*** |
| Survey return date before vs after September 11 | 0.08 (0.13) | 0.22 (0.30) | −1.05 (0.68) |
| Survey return date × sexual harassment | 0.53 (0.21)* | 0.35 (0.55) | 0.50 (1.22) |
| Generalized workplace abuse | 0.08 (0.09) | 0.12 (0.22) | 1.49 (0.53)** |
| Survey return date before vs after September 11 | −0.16 (0.19) | 0.18 (0.40) | −2.24 (0.93)* |
| Survey return date × generalized workplace abuse | 0.65 (0.22)** | −0.08 (0.50) | 2.12 (1.18) |
Note. Survey returned before September 11 served as the referent category (coded 0). Control variables included wave 1 alcohol use and mood variables and demographic variables (age, race/ethnicity, and education).
* P < .5; **P < .01; ***P < .001.
TABLE 4—
Mean Scores for Significant Interactions Before and After September 11, by Outcome and Work Stressor: Women Only
| Alcohol Use Quantitya | Escape Motives for Alcohol Useb | Anxietyc | ||||
| Before September 11 | After September 11 | Before September 11 | After September 11 | Before September 11 | After September 11 | |
| Decision latitude | ||||||
| High | 1.51 | 1.53 | 7.11 | 7.22 | 7.43 | 6.13 |
| Low | 1.61 | 2.29 | 6.53 | 7.82 | 7.43 | 7.84 |
| Sexual harassment | ||||||
| No | 1.55 | 1.57 | ... | ... | ... | ... |
| Yes | 1.65 | 2.14 | ... | ... | ... | ... |
| Generalized workplace abuse | ||||||
| No | 1.54 | 1.32 | ... | ... | ... | ... |
| Yes | 1.66 | 2.00 | ... | ... | ... | ... |
Note. Means adjusted for age, race/ethnicity, education, and baseline measures.
aAlcohol use quantity scores ranged from 0 (none) to 7 (≥ 6 drinks per day).
bScores on 5 items developed by the Alcohol Research Group23 ranged from 5 to 20.
cScores on the 9-item tension/anxiety factor of the Profile of Mood States22 ranged from 0 to 36.
DISCUSSION
The findings from this study suggest that women who are already experiencing chronic stress (in this case, particular work-related stressors involving lack-of-control issues) are most vulnerable to experiencing elevated alcohol use and anxiety in the face of a traumatic macrosocial stressor such as the events related to September 11, 2001. Future studies could be useful in extending this particular investigation to examine the salience of other chronic stressors, such as those related to family social roles, and to examine a wider array of psychiatric outcomes as well as clinical disorders beyond the elevation of symptomatology per se. Nevertheless, these findings have important implications for the further specification of those most at risk for negative alcohol use and anxiety outcomes and thus represent a potential focus for future interventions aimed at helping individuals cope with the experience of macrosocial stressors such as terrorist attacks and their ramifications.
An interesting issue for further research is that only women, not men and women, manifested both anxiety and alcohol use negative outcomes, especially when the experience of September 11 was coupled with lack of control in the work setting. We might speculate that, given the greater involvement of women in social networks and in the provision of social support to the members of their networks,24 they may have been particularly distressed as a consequence of fears for both their own safety and the safety of their friends and families. This explanation would be consistent with Kessler and McLeod’s25 argument that greater involvement by women compared with men in social networks entails an emotional drain, or “cost of caring.” Moreover, chronic work stressors might have been viewed as functioning to inhibit the caretaking activities that women believed they needed to perform in the post–September 11 period. Alternatively, given economic insecurities after September 11, 2001, women with work stressors might have had additional worries about job security issues. Another interesting issue for future study involves the extent to which job characteristics such as decision-making latitude are protective in helping individuals withstand macrosocial stressors. Alternatively, to what extent are individuals with particular psychological traits differentially selected into jobs that embrace high decision-making latitude? Future research should examine the role personality traits might play in the links between occupational role incumbency and protection from the detrimental effects of macrosocial traumatic stressors.
In our initial analyses of the data set, we examined the effect of returning the questionnaire fairly immediately after September 11 (e.g., within 1 month after the events) versus later in the semester. These analyses produced no additional or different findings, so we present the results focused on differences before–September 11 versus after–September 11. However, future research should continue to examine the extent to which the experience of various macrosocial stressors along with other forms of stress has immediate versus longer-term consequences for mental health. In addition, there is a slight possibility that some of the questionnaires returned after September 11 were actually completed before September 11. Thus, future research addressing the impact of a macrosocial stressor should accurately record timing of assessments.
Finally, we should note other limitations in the interpretation of these findings. First, we used mail surveys, whose limitations include (1) biases from reading or language difficulties of some of the intended respondents and (2) the potential for greater initial nonresponse by individuals uncomfortable with sensitive questions such as those involving work stressors they may have experienced. Second, although other studies have shown negative mental health effects related to September 11 across the country, one could speculate that our findings might have been even stronger in a sample closer to the areas most directly affected by the terrorist attacks. In addition, our particular sample, initially derived from a university workplace, had fewer minority group participants by wave 3. In addition, although wave 3 responders were similar to the initial sample in terms of stressors and most alcohol use outcomes, they showed fewer and less severe symptoms of anxiety than the original participants. Thus, our findings are less generalizable to non-Whites and may have underestimated anxiety reactions to September 11. Finally, our study compared mental health characteristics of respondents before versus after September 11, so we had no direct measures of the particular meanings that September 11 had for those respondents who returned their surveys after that date. Nonetheless, this study contributes knowledge regarding additional risk factors for anxiety and altered alcohol consumption behaviors following a macrosocial stressor—in this case, the terrorist attacks of September 11, 2001. Moreover, the longitudinal nature of the data set provided the means to control for prior mental health status in the determination of the links between the experience of September 11, chronic stressors, and anxiety and alcohol use outcomes.
Acknowledgments
This research was supported by National Institute of Alcohol Abuse and Alcoholism (grant R01–AA09989).
We thank the University of Illinois at Chicago Survey Research Laboratory for data collection activities.
Human Participant Protection This research was approved by the University of Illinois at Chicago institutional review board.
Peer Reviewed
Contributors J. A. Richman conceptualized the study and wrote most of the article. J. S. Wislar assisted with the study and completed all of the data analyses. J. A. Flaherty contributed to the conceptualization of the study and the interpretation of findings. M. Fendrich contributed to the study design and interpretation of the findings. K. M. Rospenda contributed to the study design and interpretation of the findings.
References
- 1.Centers for Disease Control and Prevention. Psychological and emotional effects of the September 11 attacks on the World Trade Center—Connecticut, New Jersey, and New York, 2001. JAMA. 2002;288:1467–1468. [PubMed] [Google Scholar]
- 2.Schuster MA, Stein BD, Bradley D, et al. A national survey of stress reactions after the September 11, 2001, terrorist attack. N Engl J Med. 2001;345:1507–1512. [DOI] [PubMed] [Google Scholar]
- 3.Schlenger WE, Caddell JM, Ebert L, et al. Psychological reactions to terrorist attacks: findings from the national study of Americans’ reactions to September 11. JAMA. 2002;288:581–588. [DOI] [PubMed] [Google Scholar]
- 4.Silver RC, Holman EA, McIntosh DN, et al. A nationwide longitudinal study of psychological responses to September 11. JAMA. 2002;288:1235–1244. [DOI] [PubMed] [Google Scholar]
- 5.Cardenas J, Willams K, Wilson JP, Fanouraki G, Singh A. PSTD, major depressive symptoms, and substance abuse following September 11, 2001, in a mid-western university population. Int J Emerg Ment Health. 2003;5:15–28. [PubMed] [Google Scholar]
- 6.Wheaton B. Sampling the stress universe. In: Avison WR, Gotlib IH, eds. Stress and Mental Health. New York, NY: Plenum; 1994:77–114.
- 7.Rospenda KM, Richman JA, Wislar JS, Flaherty JA. Chronicity of sexual harassment and generalized work-place abuse: effects on drinking outcomes. Addiction. 2000;95:1805–1820. [DOI] [PubMed] [Google Scholar]
- 8.Herbert TB, Cohen S. Measurement issues in research on psychosocial stressors. In: Kaplan HB, ed. Psychosocial Stress. San Diego, Calif: Academic Press; 1996:295–332.
- 9.Denzin NK, Lincoln YS. Introduction: 9/11 in American culture. In: Denzin NK, Lincoln YS, eds. 9/11 in American Culture. Walnut Creek, Calif: Altamira Press; 2003:xiii–xxi.
- 10.Richman JA, Rospenda KM, Nawyn SJ, Flaherty JA. Workplace harassment and the self-medication of distress: a conceptual model and case illustrations. Contemp Drug Probl. 1997;24:179–299. [Google Scholar]
- 11.Turner RJ, Lloyd DA. Lifetime traumas and mental health: the significance of cumulative adversity. J Health Soc Behav. 1995;36:360–376. [PubMed] [Google Scholar]
- 12.Mirowsky J, Ross CE. Social Causes of Psychological Distress. 2nd ed. New York, NY: Aldine de Gruyter; 2003.
- 13.Dillman DA. Mail and Telephone Surveys: The Total Design Method. New York, NY: John Wiley and Sons; 1978.
- 14.Sudman S, Bradburn N. Improving mailed questionnaire design. In: Lockhart DC, ed. Making Effective Use of Mailed Questionnaires. San Francisco, Calif: Jossey-Bass; 1984:33–47.
- 15.Richman JA, Rospenda KM, Nawyn SJ, et al. Sexual harassment and generalized workplace abuse among university employees: prevalence and mental health correlates. Am J Public Health. 1999;89:358–363. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Karasek RA, Theorell T. Healthy Work: Stress, Productivity, and the Reconstruction of Working Life. New York, NY: Basic Books; 1990.
- 17.Fitzgerald LF. Sexual harassment: the definition and measurement of a construct. In: Paludi MA, ed. Ivory Power: Sexual Harassment on Campus. Albany, NY: State University of New York Press; 1990:21–44.
- 18.Rospenda KM, Richman JA, Nawyn SJ. Doing power: the confluence of gender, race, and class in contrapower sexual harassment. Gend Soc. 1998;12:40–60. [Google Scholar]
- 19.Popovich PM, Jolton JA, Mastrangelo PM, et al. Sexual harassment scripts: a means to understanding a phenomenon. Sex Roles. 1995;32:315–335. [Google Scholar]
- 20.Fendrich M, Woodward P, Richman JA. The structure of harassment and abuse in the workplace: a factorial comparison of two measures. Violence Vict. 2002;17:491–505. [DOI] [PubMed] [Google Scholar]
- 21.Mirowsky J, Ross CE. Control or defense? depression and the sense of control over good and bad outcomes. J Health Social Behav. 1990;31:71–86. [PubMed] [Google Scholar]
- 22.McNair DM, Lorr M, Droppleman L. Profile of Mood States. San Diego, Calif: Educational and Industrial Testing Service; 1981.
- 23.Temple M. Trends in collegiate drinking in California: 1979–1984. J Stud Alcohol. 1986;47:274–282. [DOI] [PubMed] [Google Scholar]
- 24.Flaherty JA, Richman JA. Gender differences in the perception and utilization of social support: theoretical perspectives and an empirical test. Soc Sci Med. 1989;28:1221–1228. [DOI] [PubMed] [Google Scholar]
- 25.Kessler RC, McLeod JD. Sex differences in vulnerability to undesirable life events. Am Sociol Rev. 1984; 49:620–631. [PubMed] [Google Scholar]