Authors: Rana DellaRocco, Ryan D. Krone, Nancy L. Wayne
Categories: General, Bullying, Civil servant status, Gender, Harassment, Forensic scientists
Source: Forensic Science International: Synergy
Authors: Rana DellaRocco, Ryan D. Krone, Nancy L. Wayne
The purpose of this study was to investigate, for the first time, the extent and nature of harassment and bullying in forensic science laboratories associated with law enforcement in the United States. We hypothesized that the most common harassment would be that of male sworn civil servants working in forensics targeting women non-sworn forensic scientists. Further, we assessed whether levels of harassment were associated with workplace performance, absenteeism, and intention to leave. There were three types of self-reported harassment work performance-related, interpersonal, and physical intimidation. There were three categories of lab technician, manager/supervisor, and “Other” that included students, administrative support personnel, assistants, and consultants. Results showed that levels and types of harassment varied depending upon the inter-relationship between the harassed and the harasser, with higher levels of harassment reported by participants coming from harassers above them in authority. While the majority of harassment reported was between women, due to the overwhelmingly female respondent pool and gender of those employed in forensics, the effect of gender on harassment may be skewed. Work-related harassment was associated with a decrease in work performance and increased intention of participants to leave their agencies. Intention to leave was mitigated by the employee maintaining high work performance despite the harassment. This study can be valuable to forensic scientist employers seeking to identify those most likely to experience harassment so that agencies may intervene, prevent decreases in productivity and job satisfaction, and to address employee retention.
According to the latest data reported by Forensic Science & Technology [1], women dominate the field of forensic science and technology. It was reported that in 2017, 72 % of graduating students in this field from five of the largest institutions were women. Not surprisingly, this gender imbalance favoring women (by the numbers) is also reflected in forensic science laboratories—most of which are associated with law enforcement agencies, crime laboratories associated with city, county or state governments, and federal agencies.
Even though women now make up the majority of forensic scientists, there is growing evidence that there are problems for them working in an environment that is embedded within policing organizations that are male dominated. Holt and colleagues [2] found that women forensic scientists were more stressed than their male counterparts. They looked at a number of different factors associated with stress (long work hours, less supervisory or managerial support, poor relationships with prosecutors and the courts, and high levels of role conflict). What wasn't investigated was the potential of harassment and bullying of the women forensic scientists that contributed to the stress they were experiencing. For example, supervisors giving excessive or menial work, providing insufficient support, and providing unclear or insufficient standards for a task (i.e., role conflict)—all factors associated with stress in the Holt study [2]—are well-known forms of work harassment [3].
Although harassment and gender discrimination of women police officers and women in Science, Technology, Engineering, and Medicine (STEM) fields have been well documented [[4], [5], [6], [7], [8], [9], [10]], we are unaware of any published work investigating harassment of women forensic scientists working in laboratories associated with law enforcement. Studies show that women in policing experience harassment and stress at much higher levels than their male counterparts [[4], [5], [6], [7]]. Our study explores the experiences of women forensic scientists in an otherwise homogeneously male workplace within law enforcement agencies.
The outcome of this initial study will be instrumental in determining the nature and extent of harassment in order to develop evidence-based interventions and improvements to the work climate of forensic science laboratories.
The purpose of this proposed study is to investigate, for the first time, the extent and nature of harassment and bullying in forensic science laboratories associated with law enforcement. We hypothesize that the most common harassment will be that of male sworn civil servants working in forensics targeting women non-sworn forensic scientists, regardless of their authority in the organization. In addition, a secondary aim of this study is to assess whether levels of harassment were associated with workplace performance, absenteeism, and intention to leave. We hypothesize that higher levels of harassment are associated with greater intention to leave, absenteeism, and poorer workplace performance.
This study is a quantitative non-experimental correlational research design exploring mean differences in a victim's level of harassment between group characteristics and their interactions in addition to associations between levels of harassment with a respondent's workplace performance, absenteeism, and intention to leave. All procedures were reviewed and approved by the Institutional Review Board of the University of California at Los Angeles for the protection of human subjects. Informed consent was obtained for experimentation with human subjects.
The target population for this study is any employee who works at or supervises a forensic laboratory in the U.S. This includes administrative staff and lab technicians as well as sworn and non-sworn officers. This study reached out to directors of forensic laboratories associated with law enforcement, contacts at the National Institute of Justice, the American Academy of Forensic Sciences, as well as regional groups and discipline-specific groups to recruit forensic scientists for the survey. Therefore, the sampling technique was convenience sampling with no random component. According to the U.S. Bureau of Labor Statistics, there were about 17,000 forensic science technicians in the U.S. in 2021, the year that we collected the survey data (https://www.bls.gov/oes/current/oes194092.htm). With a final sample size of 479 respondents, we recruited approximately 3 % of the total target population for this study.
An a priori power analysis was conducted using G∗Power v3.1.9.4 to determine the minimum sample size required to find significance with a desired level of power set at .80, an alpha (α) level at .05. For the initial research aim (determinants of levels of harassment) the analysis assumed an interaction effect model analysis of variance (ANOVA) with 16 groups, 15 numerator degrees of freedom (df), and a moderate effect size of f = .25 (Cohen's d = .50), the sample size estimate was 314. For the second research aim (levels of harassment on work performance, absenteeism, and intention to leave) the analysis assumed a logistic regression with a moderate effect size of OR (odds ratio) = 2.50 (Cohen's d = .505), the sample size estimate was 108. Based on this analysis, it was determined that a minimum of 314 participants were required to ensure adequate power [[11], [12], [13]].
A Forensic Scientist Harassment Survey was designed using a slight modification of survey questions described by Einarsen and coworkers [14] in “Measuring exposure to bullying and harassment at Validity, factor structure and psychometric properties of the Negative Acts Questionnaire-Revised”. Questions used in the Einarsen survey were validated using Cronbach's alpha showing a score of .90, indicating excellent internal consistency and reliability. The NAQ-R was designed to establish “a reliable, valid, comprehensive, yet relatively short scale, tailor-made for use in a variety of occupational settings”.
At the start of the survey, we defined the three main appearances of harassment/bullying behavior as described by Raineri and coworkers [3]. The basic 22 item NAQ-R was, in fact, a good fit for the purposes of this Forensic Scientist Harassment study and was used for the Forensic Science Harassment Survey. The NAQ-R test measures exposure to three subscales of bullying (work related, person related, physical intimidation) within the last 12 ‘‘Never,’’ ‘‘Now and then,’’ ‘‘Monthly,’’ and ‘‘Weekly or Daily”. An overall scale score is also constructed using all 22 items of the NAQ-R. For any participant who answered affirmative for bullying, there were additional questions about the gender of the harasser, level in the organization of the harasser, and sworn or non-sworn employee status of the harasser. This study also used additional questions described in the Einarsen survey [14] that asked about absenteeism (no days off, 1–3 days, 4–10 days, 11–20 days, more than 20 days), intention to leave (never, rarely, sometimes, quite often and very often), and work performance (varying from 100 % to less than 50 % of normal capacity) within the last 12 months. This study included additional demographic questions about gender, race/ethnicity, age, educational attainment, region of the country, years working in the organization, and job role as sworn or non-sworn forensic science employee. Survey questions are located in Appendix A in Supplementary Materials.
Altogether, these questions were used to determine the impact of harasser gender, level in the organizational hierarchy, and civil servant status (sworn or non-sworn) on harassment, and the impact of harassment on absenteeism, intention to leave, and work performance in forensic science laboratories.
SurveyMonkey® was used for uploading the online survey instrument. As mentioned before, the recruitment effort was aided by contacts at the National Institute of Justice, the American Academy of Forensic Sciences, as well as regional groups and discipline-specific groups. The respondents did provide some identifying information such as region, gender, and race/ethnicity and so they were not completely anonymous, but the data were confidentially encoded and entered into a secure database on a secure computer. Two waves of data collection were completed to achieve the desired sample size required for acceptable power. The first wave of data collection was conducted from May 26, 2021 to August 31, 2021 and collected a total of 530 respondents. The second wave of data collection was conducted from October 20, 2021 to October 31, 2021 and collected a total of 91 respondents, for a total of 622 initial respondents.
Once data collection was complete, the data was screened to ensure validity of cases [15], a missing values analysis was conducted [16], and the variables of interest were subjected to univariate assumptions testing to assess normality and identify extreme outliers [17]. The screening process removed a total of 143 cases due to non-consent, they did not work at or supervise a forensic science lab, they skipped inclusion criteria questions, or they did not answer the harassment questions. The analytic sample for this study was 479 respondents. Additionally, there was a skip logic issue related to the harasser questions such that if a respondent answered that they were not harassed in the last 12 months, any analysis using harasser characteristics would drop that respondent due to the absence of a harasser — leading to missing by design. Consequently, for the first research aim which uses the harasser characteristic variables, this reduced the useable sample size down to 195 respondents.
Another consideration important to this study was the classification of the “Other” category for the respondent's role. The primary groups that were coded were lab technician, manager/supervisor, and “Other”. The “Other” group included the following roles that did not ‘fit’ within the lab technician or manager/supervisor administrative secretary, student, support specialist in administration, office assistant, grant fund researcher, and consultant. Lastly, reliability testing was conducted on each of the harassment scales using inter-item correlation and reporting Cronbach's alphas to provide the degree of reliability for each scale.
Further information about the data management can be found in Statistical Analysis Details in Supplementary Materials.
Once the data was cleaned and prepared for analysis, descriptive statistics were conducted on key variables to describe the sample. Frequencies and percentages were conducted for categorical and ordinal variables and means and standard deviations were conducted for continuous of variables of interest. Preliminary bivariate analysis was conducted using cross-tabulations, analysis of variance, and correlations to explore the data. The analysis of the first aim was conducted using a generalized linear model assuming a normal distribution and identity link. P-values less than .05 were considered statistically significant.
A mediation relationship was discovered when exploring the relationships between harassment and work performance and intention to leave. The mediation analysis was conducted following Baron and Kenny [18] and involved separate regression estimations of the c, a, and b, c’ paths to observe potential full or partial mediation effects. Testing of the indirect effect was conducted using a bootstrapping method developed by Tofighi and MacKinnon [19] which produced 95 % confidence intervals to assess statistical significance (application located https://amplab.shinyapps.io/MEDCI/). The data preparation and analysis were conducted in SPSS version 27 (IBM Corp. 2021) and the alpha level for this study was set at α = .05.
Further information about the data analysis can be found in Statistical Analysis Details in Supplementary Materials.
Because this study limited respondent identification characteristics, kept strict confidentiality protocols, and data was collected from around the United States without specific identifying locations, there were no risks to the respondents participating in this study in terms of having their identity revealed. For any forensic scientist participating in this study who had been harassed or bullied, there was a benefit to knowing that they are being heard in a safe and confidential way. Answering questions about workplace harassment has the potential risk of making the survey participant feel uncomfortable and/or trigger emotionally- and psychologically-charged memories and feelings. The authors provided a list of accessible online resources for dealing with workplace harassment at the end of the survey that the participants could access at any time while taking the survey.
Sample descriptive statistics were conducted prior to exploring the research aims. Frequencies and percentages of categorical study variables are provided in Table 1 in Supplementary Materials. A description of the sample indicated that the majority of the study participants did not or rarely considered leaving their current job within the last 12 months (59.5 %). Similarly, a majority of the sample indicated that they had less than three days absence from work over the last 12 months (60.1 %).
In terms of work performance capacity, the largest proportion of respondents indicated that they were at 75%–99 % normal capacity (44.7 %) while approximately one-third of the sample indicated that their work performance was at 100 % of normal capacity (35.9 %). A large proportion of the sample indicated that they had experienced some sort of harassment in the previous 12 months (40.7 %).
In terms of the harasser characteristics, the vast majority were non-sworn employees (83.2 %), the majority were above the respondent in terms of authority (56.3 %) with a third of the sample peers (33.2 %), and the majority were female (63.9 %). In terms of respondent characteristics, most were female (79.7 %), White/Caucasian (85.2 %), Laboratory technicians (64.1 %) and had non-sworn civil servant status (92.7 %). Respondent age groups were fairly evenly distributed. In terms of respondent work experience, the largest group was less than 4 years (23.4 %) and of this group the largest percentage consisted of 23–30-year-old (57.1 %). The majority of respondents had a master's degree (57.2 %) with the next largest group being a bachelor's degree or lower (32.8 %). Most respondents were located in the Southern (36.3 %) and Midwest (24.4 %) regions of the U.S. A majority of respondents indicated that COVID-19 influence was negligible or minimal (54.7 %).
Means and standard deviations were conducted for the continuous study variables (see Table 2 in Supplementary Materials). All of the continuous variables were scale or subscales of the harassment measure and reliability testing was conducted on each of these scales to establish internal consistency. The scales were averaged, and all ranged from 1 to 4 (four-point Likert) which measured frequency (Never, Now and Then, Monthly, or Weekly/Daily) of a particular kind of harassment so that higher values on the scale indicate greater frequency of harassment. The overall harassment scale had a mean of 1.45 (SD = .52), the person-related harassment subscale had a mean of 1.42 (SD = .57), the work-related harassment subscale had a mean of 1.64 (SD = .63), and the physical intimidation subscale had a mean of 1.19 (SD = .36). The frequency of harassment occurrence was lower than the theoretical midpoint of the scale (M = 2.5) for the overall harassment scale suggesting that harassment was not a frequent occurrence for this sample. For the subscales, person-related harassment also approached overall harassment's mean level with work-related harassment exhibiting greater frequency (approaching the midpoint) and physical intimidation much less frequent. The Cronbach's alpha was very strong for the overall scale and the person-related and work-related subscales, with Cronbach's alphas ranging from .828 to .944, indicating excellent reliability. The physical intimidation subscale exhibited adequate reliability using inter-item correlation.
In further exploration of the descriptive data, it was noted that the sample contained considerably more female (80.3 %) than male (19.7 %) survey participants. When looking at the gender breakdown of harassers for those participants that were harassed (see Fig. 1), a large proportion of harassers were female (63.9 %) compared to male (36.1 %). It is important to note that there was a larger proportion of male harassers compared to males in the study and conversely a smaller proportion of female harassers compared to females in the study. This may indicate that the occurrence of harassment by male sworn civil servants working in forensics is not significant or was undetected in this sample due of the respondent pool's characteristics. Future work could include male sworn civil servants who work outside of forensics as detectives, commanders or other sworn stakeholders who interact with forensics but do not work in the field.Fig. 1Proportion of Participants by Gender in the Overall Study Compared to the Gender Proportions of Harassers for those Participants who were Harassed. Descriptive visualization showing disproportionately fewer female harassers identified in the study compared to overall percent of female respondents (63.9 % versus 80.3 %), as well as disproportionately more male harassers identified in the study compared to overall percent of male respondents (36.1 % versus 19.7 %).Fig. 1
In addition, there were considerably less sworn officers (6.9 %) compared to non-sworn participants (93.1 %). When looking at the sworn status of harassers for those participants who were harassed (see Fig. 2), a larger proportion of harassers are non-sworn (83.2 %) compared to sworn (16.8 %). Again, there is a larger proportion of sworn harassers compared to sworn participants in the study and conversely a smaller proportion of non-sworn harassers to non-sworn participants in the study. Once again, this outcome is very possibly a result of the respondent pool and its majority female participants.Fig. 2Proportion of Participants by Gender in the Overall Study Compared to the Gender Proportions of Harassers for those Participants who were Harassed. Descriptive visualization showing disproportionately more sworn harassers identified in the study compared to overall percent of sworn respondents (16.8 % versus 6.9 %), as well as disproportionately fewer non-sworn harassers identified in the study compared to non-sworn respondents (83.2 % versus 93.1 %).Fig. 2
Last, it was important to note the differences in levels of harassment across the three different types in the NAQ-R. Notably, the means of each of the three types of harassment showed considerably less levels of harassment for physical compared to personal or work-related harassment. Paired sample t-tests were conducted to assess whether this was a statistically significant difference with the results shown in Table 3 in Supplementary Materials. The results indicate that physical intimidation (M = 1.19) was significantly less than work-related harassment (M = 1.64), p < .001, dz = .845. Similarly, physical intimidation (M = 1.19) was significantly less than person-related harassment (M = 1.42), p < .001, dz = .512.
The first research aim explored whether sworn harassers targeted female respondents more than male respondents, while controlling for respondent characteristics. A linear regression analysis was conducted using each harassment subscale and the overall scale as the outcome of interest and an interaction between harasser civil servant status (sworn vs. unsworn) and the gender of the respondent (male or female). There were no statistically significant effects for this interaction for any of the harassment subscales or the overall harassment scale (all p-values were greater than .05). This suggests that there was no difference in levels of harassment of males or female targets whether the harassers were sworn or unsworn. The initial research aim and its corresponding hypothesis tested whether sworn harassers working in forensics targeted females more so than males. This hypothesis was not supported. Nevertheless, as noted in Fig. 1, Fig. 2, there were disproportionately more male harassers and sworn harassers compared to their representation in the study. This may indicate that further investigation into whether sworn male harassers not working in forensics would show a statistically significant difference in the occurrence of harassment reported by forensic scientists.
A baseline analysis of the harassment scales with the harasser and respondent characteristics as covariates did find interesting significant results. Linear regression analyses were conducted utilizing both harasser and respondent characteristics as predictors of the harassment scales. Estimated marginal means were calculated and reported in order to provide a sense of statistically significant group differences while controlling for all covariates in the model.
For the overall harassment scale, the only statistically significant predictors in the model were harasser authority, respondent age, and respondent experience. The results indicate that a respondent experienced lower levels of harassment from a subordinate harasser (M = 1.65, SE = .18) compared to a higher authority harasser (M = 2.05, SE = .13), p = .012. Similarly, a respondent experienced lower levels of harassment from a peer harasser (M = 1.77, SE = .14) compared to a higher authority harasser, p = .006. In addition, respondents who were 50 years of age or older experience greater levels of harassment (M = 2.08, SE = .16) compared to the youngest age group (23–30 years) (M = 1.61, SE = .19), p = .020. Lastly, respondents who had 20+ years of experience on the job had lower levels of harassment (M = 1.57, SE = .18) compared to respondents that had less than 4 years of experience (M = 2.10, SE = .17), p = .005. This seemingly contradictory finding of older respondents experiencing greater levels of harassment while those with more experience having lower levels of harassment is commented on in the Discussion section.
For the person-related harassment subscale, the same statistically significant predictors in the model were harasser authority, respondent age, and respondent experience. Respondents experienced less harassment from a subordinate harasser (M = 1.66, SE = .17) compared to higher authority harassers (M = 2.03, SE = .15), p = .038. In addition, respondents experienced less person-related harassment from a peer harasser (M = 1.79, SE = .15) compared to higher authority harassers, p = .033. In terms of age, respondents experienced more harassment if they were 50+ years old (M = 2.16, SE = .19) compared to the youngest age group (M = 1.61, SE = .22), p = 017. In terms of experience, respondents experienced significantly less person-related harassment if they had 4–6 years' experience (M = 1.85, SE = .19), 7–9 years' experience (M = 1.81, SE = .18), 13–15 years' experience (M = 1.75, SE = .23), and 20+ years' experience (M = 1.50, SE = .21) compared to if they had less than 4 years’ experience (M = 2.21, SE = .19), all ps < .05.
For the work-related harassment subscale, the statistically significant predictors in the model were authority of harasser and age only. Respondents experienced less work-related harassment from a subordinate harasser (M = 1.80, SE = .21) or peer harasser (M = 1.91, SE = .15) compared to a higher authority harasser (M = 2.34, SE = .17), all ps < .05. In terms of age, respondents experienced more work-related harassment if they were 50+ years old (M = 2.25, SE = .19) or 46–50 years old (M = 2.29, SE = .20) compared to the 23–30 years old age group (M = 1.73, SE = .22), all ps < .05. Years of experience was not a significant predictor in the model for this outcome.
For the physical intimidation harassment subscale, the statistically significant predictors in the model were role of the respondent, gender, and experience. A respondent whose role on the job was “Other” experienced higher levels of physical harassment (M = 1.79, SE = .21) compared to respondents who were lab technicians (M = 1.20, SE = .11), p = .007. Respondents who were male (M = 1.56, SE = .12) experienced higher levels of physical harassment than females (M = 1.20, SE = .11), p = .027. Respondents who had 16–20 years of experience had higher levels of physical harassment (M = 1.66, SE = .13) compared to respondents with less than 4 years of experience (M = 1.34, SE = .13), p = .029.
Due to harasser authority indicating a significant effect in most of the models, the estimated marginal means of the outcomes by authority group are presented in Fig. 3. For overall, personal, and work-related harassment outcomes, the harassment a respondent experiences is usually greater if coming from a person in authority compared to a peer or subordinate. However, for physical harassment, while the levels of harassment are much lower by comparison, there were no significant differences in means by authority group, suggesting that there was no difference in physical harassment depending on the authority of the harasser.Fig. 3Levels of Harassment a Respondent Experiences by Type of Harassment and by the Authority of the Harasser. Authority of the harasser is indicated 1) A subordinate (black histograms); 2) A peer (shaded histograms); and 3) Above me in authority (white histograms). The data are shown as the estimated marginal means ( ± SE) of harassment outcome. Different letters above the histogram error bars denote statistically significant differences within each type of harassment. For example, respondents experienced greater overall harassment levels by harassers above them in authority compared to that by harassers who were subordinates or peers.Fig. 3
To expand on the initial findings, additional interaction analyses were conducted that explored whether the harasser's civil servant status (sworn or unsworn officer) effect on levels of harassment were moderated by the respondent's role (lab technician, manager/supervisor, or “Other”). To reiterate, respondents within the “Other” group consist of roles such as administrative secretary, student, support specialist in administration, office assistant, grant fund researcher, and consultant. These models were conducted using overall harassment and each of its subscales as dependent variables.
For the overall harassment scale, once again, harasser authority, age, and experience were all significant predictors in the model and had the same directionality and similar coefficient effects as the prior models. The main effects of harasser civil servant status and respondent role were not statistically significant, however, the interaction of both of them was significant. Fig. 4 provides a visualization of the impact of the interaction effect using the estimated marginal mean comparisons of the interacted groups. For those respondents whose role was classified as “Other” and whose harasser was non-sworn, these respondents had much higher levels of overall harassment (M = 2.30, SE = .29) compared to lab technicians with non-sworn harassers (M = 1.56, SE = .15) and manager/supervisors with sworn harassers (M = 1.46, SE = .20), all ps < .05.Fig. 4Interaction of Role of Respondent and Civil Servant Status of Harasser on Levels of Overall Harassment a Respondent Experiences. Civil servant status of the harasser is indicated 1) Sworn harasser (black histograms), and 2) Non-sworn harasser (white histograms). The data are shown as the estimated marginal means ( ± SE) of overall harassment. Different letters above the histogram error bars denote statistically significant differences across all data bars. For example, Laboratory Technicians experienced significantly less levels of overall harassment by non-sworn harassers than those respondents with the role of “Other” forensic scientist.Fig. 4
For the personal harassment scale, harasser authority, age, and experience were significant predictors of harassment. In addition, this interaction of harasser civil servant status and respondent role in the organization was significant for the personal harassment subscale. Fig. 5 provides an illustration of the impact of the interaction effect using the estimated marginal mean comparisons of the interacted groups. The results indicate that those respondents whose role was classified as “Other” and whose harasser was non-sworn had much higher levels of personal harassment (M = 2.36, SE = .34) compared to lab technicians with non-sworn harassers (M = 1.56, SE = .17) and manager/supervisors with sworn harassers (M = 1.41, SE = .23), all ps < .05.Fig. 5Interaction of Role of Respondent and Civil Servant Status of Harasser on Levels of Personal Harassment a Respondent Experiences. Civil servant status of the harasser is indicated 1) Sworn harasser (black histograms), and 2) Non-sworn harasser (white histograms). The data are shown as the estimated marginal means ( ± SE) of personal harassment. Different letters above the histogram error bars denote statistically significant differences across all data bars. For example, Laboratory Technicians experienced significantly less levels of personal harassment by non-sworn harassers than those respondents with the role of “Other” forensic scientist.Fig. 5
For physical intimidation harassment, only the servant status and role interaction and respondent experience were statistically significant. Fig. 6 provides a visualization of the interaction effect. Respondents whose role was classified as “Other” and whose harasser was non-sworn had much higher levels of physical harassment (M = 1.98, SE = .23) compared to all other managers/supervisors with non-sworn harassers (M = 1.35, SE = .12), lab technicians with non-sworn harassers (M = 1.15, SE = .12), lab technicians with sworn harassers (M = 1.29, SE = .13), managers/supervisors with sworn harassers (M = 1.23, SE = .15), and “Other” respondents with sworn harassers (M = .90, SE = .46). In sum, lab technicians were trending towards experiencing a higher level of harassment by sworn harassers with decreasing levels of harassment for managers/supervisors and then the “Other” group. The respondents in the “Other” group experienced the highest level of harassment by non-sworn harassers with decreasing levels of harassment for managers/supervisors and then lab technicians. The “Other” group experienced the highest level of harassment by non-sworn harassers and the lowest level of harassment by sworn harassers.Fig. 6Interaction of Role of Respondent and Civil Servant Status of Harasser on Levels of Physical Harassment a Respondent Experiences. Civil servant status of the harasser is indicated 1) Sworn harasser (black histograms), and 2) Non-sworn harasser (white histograms). The data are shown as the estimated marginal means ( ± SE) of physical harassment. Different letters above the histogram error bars denote statistically significant differences across all data bars. For example, Laboratory Technicians experienced significantly less levels of physical harassment by non-sworn harassers than those respondents with the role of “Other” forensic scientist.Fig. 6
The second research aim explored whether levels of harassment impacted a respondent's work performance, absenteeism, and intention to leave while controlling for harasser and respondent characteristics. A generalized linear regression analysis using a logit link function was conducted using the binary coded work performance, days absent, and intention to leave as the outcomes of interest. For work performance, the significant predictors of the model were work-related harassment scale, respondent role, respondent education, and COVID-19 influence. No other covariates in the model were statistically significant. The findings suggest that a respondent who experiences a one-unit increase in work-related harassment is 2.51 times less likely to achieve strong work performance (B = −.920, OR = .399, p = .012). A respondent who is a manager/supervisor level in the organization is 3.18 times more likely to achieve strong work performance compared to a lab technician (B = 1.158, OR = 3.183, p = .003). A respondent who has a master's degree is 3.55 times less likely to achieve self-reported strong work performance compared to a bachelor's or less (B = −1.267, OR = .282, p = .001). Lastly, a respondent that was highly impacted by COVID-19 was 5.47 times less likely to achieve strong work performance compared to if COVID-19 had no influence on a respondent at all (B = −1.700, OR = .183, p < .001).
For days absent, the significant predictors of the model were physical intimidation harassment scale and respondent role in the organization. No other covariates in the model were statistically significant. The findings suggest that a respondent who experiences a one-unit increase in physical intimidation harassment is 2.97 times less likely to call in absent to work (B = −1.088, OR = .337, p = .008). A respondent who is a manager/supervisor level in the organization is 1.779 times less likely to call in absent to work compared to a lab technician (B = −.576, OR = .562, p = .022).
Lastly, for intention to leave, the only significant predictor in the model was work-related harassment. These findings suggest that a respondent who experiences a one-unit increase in work-related harassment is 4.046 times more likely to seriously consider leaving their job (B = 1.398, OR = 4.046, p < .001).
Due to work-related harassment's significant impact on both work performance and intention to leave, marginal probabilities were estimated to visualize the probability of these outcomes occurring (performing well or deciding to leave the job) given a value of work-related harassment holding all other variables in the model constant. The results are shown in Fig. 7. This figure clearly shows that as work-related harassment increases, the probability of work performance decreases. However, work performance does not seem to go below the 50 % threshold (the likelihood of good work performance being a 50/50 chance) even at the highest levels of work-related harassment. It's also noteworthy that for work performance, the standard errors of the estimate expand as levels of work-related harassment increase, adding to greater uncertainty at higher levels of harassment.Fig. 7Probability of Outcome Occurring Given Levels of Work-related Harassment for Both Work Performance and Intention to Leave Outcomes. Data on the y-axis show the marginal probability ( ± SE) of self-reported work performance (dotted line) and intention to leave (solid line). Data on the x-axis show the work-related harassment scaled from 1 (None) to 4 (High). As work-related harassment increases, the probability of high work performance decreases while intention to leave increases.Fig. 7
Conversely, as work-related harassment increases, the probability of intention to leave increases with much tighter standard errors regardless of harassment level. At the highest levels of work-related harassment, the probability of intention to leave approaches 100 % (an almost certain likelihood that a respondent would leave their job), highlighting the substantial impact work-related harassment has on the mindset of respondents towards job retention.
Due to respondent role being a significant effect in most of the models, estimated marginal probabilities of the outcomes were conducted by respondent role holding all other variables in the model constant. The results are shown in Fig. 8. The probability of strong work performance is highest among manager/supervisor level with a 95 % probability (SE = .28) and lowest among lab technicians with an 86 % probability (SE = .41) with a statistically significant difference in probabilities between these two groups, p = .022. The probability of calling in absent to work and intention to leave is highest among lab technicians compared to the other two roles, however there were no statistically significant differences between those groups.Fig. 8Probability of Work Performance, Days Absent, and Intention to Leave by Respondent Role. Data are shown as the marginal probability of the occurrence ( ± SE) of high work performance, days absent, and intention to leave as a function of type of role of the respondents (Laboratory technician in black histograms, Manager/supervisor in shaded histograms, and Other forensic scientist in white histograms). Different letters above the histogram error bars denote statistically significant differences across all data bars. For example, Laboratory technicians have significantly lower levels of self-reported work performance compared to Managers/supervisors.Fig. 8
In exploring the relationships between harassment and performance related measures, work-related harassment seemed to stand out as a predictor in most of the models. In particular, the relationship between work-related harassment with work performance and intention to leave outcomes was a recurring theme and these relationships suggested a potential mediation effect. Accordingly, a mediation analysis was conducted using work-related harassment as the independent variable, work performance as a mediator, and intention to leave as the outcome of interest. Following Baron and Kenny [18], three separate models were conducted in order to test the c, a, b, and c’ paths of the mediation model; harasser and respondent demographic covariates were included in each of the models as controls. The results of the mediation analysis are shown in Fig. 9.Fig. 9Mediation Analysis of Work-related Harassment on Intention to Leave via Work Performance. Three possible pathways of mediation are shown. The a path shows an association between work-related harassment and work performance that is negative and significant (i.e., high work-related harassment is associated with poor work performance). The b path shows an association between work performance and intention to leave that is negative and significant (i.e., high work performance is associated with decreased intention to leave). The c’ path shows an association between work-related harassment and intention to leave that is positive and significant (i.e., high work-related harassment is associated with increased intention to leave.Fig. 9
For the c path model (work-related harassment on intention to leave), the results indicate that the relationship between work-related harassment and intention to leave is positive and significant (B = 1.875, OR = 6.524, p < .001), as work-related harassment increases, the likelihood of a respondent deciding to leave their job increases. This suggests that there is a direct association of work-related harassment on intention to leave.
For the a path model (work-related harassment on work performance), the results show that the association between work-related harassment and work performance is negative and significant (B = −1.293, OR = 3.643, p < .001), as work-related harassment increases, the likelihood of a respondent's strong work performance decreases. This suggests that there is potential that work performance mediates the effects of work-related harassment on intention to leave the workplace.
For the b and c’ paths model (work performance and work-related harassment on intention to leave), the results indicate that the b path is negative and significant (B = −.959, OR = 2.608, p = .005) and the c’ path remains positive and significant (B = 1.752, OR = 5.764, p < .001) in the presence of this mediating effect. As work performance increases, the likelihood of intention to leave decreases and as work-related harassment increases, the likelihood of intention to leave increases.
Following Fig. 9 diagram, as work-related harassment increases, the likelihood of good work performance decreases (path a) and as the likelihood of good work performance increases, the likelihood of intention to leave also decreases (path b). This captures the mediation or indirect effect. As work-related harassment increases, the likelihood of intention to leave increases (path c’). However, the presence of a still statistically significant direct relationship between work-related harassment and intention to leave suggests that a partial mediation effect exists given the still significant c’ path in the model.
To estimate the indirect effect the coefficients of the a and b paths are simply multiplied with each other, and a test of statistical significance requires a bootstrapping method. The calculated estimated indirect effect was 1.24 (SE = .498) and the distribution of the product of coefficients bootstrapping method generated a 95 % confidence interval of [.348, 2.301]. With zero outside the 95 % confidence interval, this indicates a statistically significant indirect effect of work-related harassment on work performance and work performance on intention to leave. The results of this mediation analysis suggest that work-related harassment has both a direct impact on intention to leave in addition to an indirect path via work-related harassment's impact on decreasing work performance and thereby increasing intention to leave.
Harassment and bullying in the workplace are common phenomena that affect many industries with 30 % of U.S. workers reporting they have experienced some form of bullying in their working lives [20]. Forensic science is in a unique position with a largely homogeneously female workforce performing work, typically, within a largely homogeneously male organization. Additionally, among all STEM fields, forensic science is the only one that has a majority of female graduates [21]. In contrast to this, the U.S. Federal Bureau of Investigation's UCR 2017 table of Full Time Law Enforcement employees (which includes both sworn and professional staff) shows that males make up 73 % of all law enforcement employees, while women represent approximately 27 % (https://ucr.fbi.gov/crime-in-the-u.s/2017/crime-in-the-u.s.-2017/tables/table-74). More specifically, of over 670,000 sworn officers, 87.5 % are men while among the population of over 287,000 non-sworn law enforcement employees, 60.4 % are women (https://ucr.fbi.gov/crime-in-the-u.s/2017/crime-in-the-u.s.-2017/tables/table-74).
Our original hypothesis regarding the harassment of non-sworn female forensic scientists at the hands of male sworn officers who are working in forensics was not supported. However, the findings suggest that harassment does occur in forensic sciences at higher levels than experienced by most employed adults and that within this industry, the perpetrators are mostly female. Over 40 % of the study respondents report experiencing harassment and bullying in the past 12 months as compared to a 2021 survey by the Workplace Bullying Institute [20] in which 13 % of employed adults in the U.S. reported experiencing bullying over a 12-month period. This high rate of harassment is similar to the experiences of women in nursing who report in one study [22] that over 44 % of them have experienced bullying within the previous year. Interestingly, the vast majority of our respondents who experienced harassment in the present study reported that their harassers were women. Due to the unusually high concentration of women in forensics working within largely male police organizations (homogeneous v. homogeneous), their status as minorities may contribute to the levels of harassment they perpetrate. Studies on workplace violence in a similarly “homogeneous v. Homogenous” workplaces, such as healthcare, have shown higher levels of harassment for female employees. A 2020 study demonstrates that women in the healthcare industry experience gender-based harassment at higher rates with nearly 44 % reporting experiencing it with the vast majority of perpetrators being people they work with [23].
It should be noted that sexual harassment was not evaluated in this study but was mentioned by study participants in the open-ended data. It is a limitation of this study and one worth pursuing in the future. Due to the high levels of reported sexual harassment of women in other studies focused on policing [24] and healthcare workers [25], it is likely to also be a significant factor in the forensic sciences. However, several studies showing high levels of sexual harassment of women in the traditionally male workplaces suggest that women in non-traditional fields may be less inclined to report sexual harassment then those in other fields due to fear of retaliation, concerns about career advancement and being viewed as an outsider [[26], [27], [28], [29]]. Forensic science may well qualify as a non-traditional field existing at the intersection of STEM and police work, possibly resulting in underreporting of the occurrence of sexual harassment in this field. Further investigation into this phenomenon may reveal interesting insights, particularly if employees included in the study were not directly working in forensics but were instead stakeholders such as detectives or commanders.
Additionally, the authors acknowledge the binary nature of the demographic survey analysis pertaining to gender (male vs. female, with “other” gender being so small that it could not be analyzed) and recognize this as a limitation in the work. According to the 2024 Williams Institute study [30], nearly 60 % of nonbinary employees report experiencing harassment at work. The majority of these employees also report not being out to their supervisors and/or downplaying their LGBTQ status and largely represent employees in the under 35 group. Our study seems to show that employees over the age of 50 report the highest levels of harassment and therefore, we may have missed an important demographic by limiting our survey analysis to male or female.
Finally, several respondents' comments noted that the study's timeframe limiting harassing experiences to the past 12 months caused them to answer “no” to our survey questions. These employees report that they would have answered “yes” had the timeframe been extended. Our goal in attempting to identify recency of the harassment event(s) may have had the effect of discounting the lived experiences of those with longevity or those with experiences related to their entry into this field.
We did also note disproportionality in the male and sworn categories as harassers as compared to their presence in the study's population. A higher proportion of male harassers compared to men in the study was observed compared to a smaller proportion of women harassers to women in the study. There was also a larger proportion of sworn harassers than sworn participants and a smaller proportion of non-sworn harassers than non-sworn participants. Given that the study participants were largely female (79.7 %) and non-sworn (92.7 %), reflecting the population of forensic scientists in the U.S., these disproportionalities may warrant further investigation.
Harassment, in general, was experienced most often from women supervisors directed toward those lower in the organizational chart in the forensic science laboratory. According to a recent American Psychological Society survey, high levels of reported harassment in any field have the ultimate effect of causing employee stress, intention to leave, and, in certain employee groups, higher levels of absenteeism and lower levels of self-reported work performance [31]—that was supported by our second hypothesis, specifically in the forensic sciences. We found that the highest levels of harassment were experienced by those lower in the chain of command in a forensic science service. This was consistent except in the area of physical harassment where the predictor of the status of the harasser was not significant.
Interestingly, we found that employees who self-reported as being over the age of 50 reported higher levels of harassment, suggesting ageism. We also found that employees with more than 20 years of experience reported less harassment. Experiential seniority may have a mitigating effect on ageism, but a tipping point may exist between the two variables of experience in forensics and age of the practitioner. Forensic science is a rapidly evolving field. Changes in technology, regulations, educational requirements, standards, and methodologies happen quickly. Those in the 50+ age category may feel discouraged to keep up or lack the motivation to embrace the changes and therefore have a sense that their work “as is” is undervalued or unappreciated. Conversely, those with higher levels of experience may tend to be in leadership and therefore have greater or more positive engagement with these changes resulting in a decrease in the perception of harassment.
When employees experience work and interpersonal harassment, our study shows that absenteeism, low performance, and intention to leave increase. Surprisingly, we found a counter-intuitive relationship between physical harassment and absenteeism in a category of employee we named “Other” in our data. When these respondents reported physical harassment, they were surprisingly less likely to report that it affected their attendance at work.
In the “Other” worker category, we combined job descriptions such as quality managers, office assistants, laboratory support, and other non-technical positions. The experiences reported by the “Other” in this study were high in the physical harassment category with no predictors related to the rank of their harasser. When harassment is experienced, except as noted above, workers generally tend to have higher levels of absenteeism, more intention to leave employment, and poorer self-reported work performance. This is typical in many other workplaces globally and has been identified in other studies [[32], [33], [34]]. Currently, we have no plausible explanation as to why forensic scientists who experienced physical intimidation in the “Other” category showed lower rates of absenteeism than those in the other employee categories who also experienced physical intimidation.
An interesting mitigating factor we found in this study was that when forensic science employees believed their work performance was good or high, they had less desire to leave their employment. A perception of high work performance here seems to have a protective function for employees experiencing difficult workplace situations involving harassment. This effect can be useful for agencies to retain employees by creating a culture of high performance and fostering greater levels of professional development in the discipline.
Participants in the survey were also afforded an opportunity to describe their harassment and the effect it had on their work in a narrative section. Narrative comment sections in surveys are often used to share experiences not addressed by the survey questions presented. Though their comments should be considered anecdotal, it is important to summarize them here.
Several survey respondents reported various barriers to reporting and retaliation after notifying their organization. Comments from respondents included references to intimidation by management, lack of belief and support from supervisors, and a lack of action when harassment was reported. One respondent stated, “The supervisor ignored me when I approached her about the aggressive peer … and due to this bias in my direct hierarchy, I have been forced to fight management via the Union and HR.” Another respondent reported that her harasser was promoted “despite several complaints of bullying … and an investigation by sworn [investigators] … The department chose to promote him to be the Director”. Several respondents also reported a failure to act on reports of harassment including one who stated that despite having over “50 plus pages of documentation of bullying … no action is taken by the union or workplace”. Other studies indicated that respondents experienced similar barriers to reporting, naming “the perception that nothing would change, not wanting to be seen as a trouble-maker, the seniority of the bully and uncertainty over how policies would be implemented and bullying cases managed” as the most common reasons for non-reporting [35]. This is reflected in the comments from our respondents, as well.
There were also a few respondents who seemed to dismiss bullying and harassment as a significant problem in their workplace. One respondent stated that harassment doesn't happen to them because they “don't let people push them around”, while another called harassment “operationally necessary”. Of note, many female-identifying respondents reported occurrences of gender or sexual-based harassment in the narrative section, which we did not address in this study. Mirroring these narrative responses, in a study on sexual harassment in Law Enforcement, women were reluctant to describe themselves as having been sexually harassed due to the perception that this labeled them as “victims”. This victim status did not align with their self-perception as strong individuals who can manage a situation without involving supervision or entering into a formal complaint process [36]. These women also reported that sexual harassment is “part of the job” and therefore, they didn't have a reason to file complaints or have a right to a workplace free of this harassment.
Additionally, these and other comments may hint at a more significant issue - the lack of awareness of what qualifies as harassing behaviors. Comments such as “I have not experienced harassment but I have experienced having my opinions being ignored” or “We aren't bullies - just aggressive women” were reported in the present study. These seem to indicate the respondent's inability to identify harassing behaviors, a phenomenon that may be unique to forensic science, or perhaps more broadly, police organizations. This phenomenon was also reported in a National Institute of Justice [37] report in which respondents noted that harassment is so normalized they can no longer identify it. As one respondent noted in the report, “I don't think of myself as harassed, ever, but a lot of the things I experience would be considered harassment in another field.“. The normalization of more aggressive behaviors and lack of education on harassment and bullying may stem from a workplace culture—such as in law enforcement—that expects paramilitary compliance, values employees who can “handle it”, and may not focus resources on creating a wellness-forward or supportive workplace culture.
The open-ended responses showed that of those who chose to share experiences in a narrative section were experiencing harassment but were not able to answer some of the questions due to the formatting of the questions themselves. The comments above highlight the alarming nature of some of the harassment experienced by respondents outside of our 12-month timeframe, by employees who did not fall into the defined roles in the study, or by those who don't recognize that they have been harassed. Especially concerning is the acceptance of behaviors that are harassment but not defined by the respondent as harassment, or the notion that they should expect harassing behaviors when working in this field. In a study on conceptual challenges of what constitutes harassment [38], the authors state that certain types of harassment are often viewed as legal concepts rather than actual experiences of “hostile workplace” conditions. They describe this as a type of “iceberg” model where only the most egregious behaviors—those rising to the level of legal action and above the water level—are defined as harassment, while the vast majority of behaviors which create hostile work environments remain unseen and unacknowledged.
There are several reasons why harassment levels in forensic science may be higher than what is typically seen in other industries or fields. The U.S. Equal Employment Opportunity Commission lists several risk factors on their website as significant in a workplace that may make it a breeding ground for harassment (https://www.eeoc.gov/chart-risk-factors-harassment-and-responsive-strategies, accessed May 30, 2024). Two that may be relevant to forensic science are being in a largely homogeneous workplace, and significant power/role disparities between certain types of employees. As noted earlier in the Discussion, this describes the work environment of forensic science largely female and made up, primarily, of non-sworn, highly educated scientific professionals. Policing, in general, is largely male and is considered a skilled blue-collar job, though educational exceptions among the rank and file do exist. Forensic scientists make up just a small percentage of the overall workforce of police agencies— in the United States, there are approximately 17,000 forensic scientists (mostly women) and over 708,000 police officers (mostly men). The workplace for many forensic scientists looks like a homogeneous group of highly educated women working with, and at the same time separate from, another homogeneous and mostly skilled, blue-collar, male workforce. These are significant risk factors for unequal power dynamics and homogeneity, both external and internal to the forensic science services, and are at play in this situation.
Though we did not find strong direct evidence of harassment by sworn officers working within forensic science, consideration should be given to the indirect effect of non-sworn forensic scientists integrated within a police agency. This has been acknowledged by forensic science organizations for several years. The National Academy of Sciences (NAS) controversially recommended in a 2009 seminal report [39] that it was critical for forensic science “to maximize independence from or autonomy within the law enforcement community … for the purpose of removing all public forensic science laboratories and facilities from the administrative control of law enforcement agencies …“.
Though these and other best practice recommendations urge the separation of the forensic science provider and police agencies, many have not done so. This may be a result of the police agency itself being unwilling to lose direct control of a high-functioning and impactful service, the inability of the forensic science service itself to gain the necessary political buy-in or power for a separation, or the inability to acquire the significant resources necessary to actuate the change.
Separation may not only create the desired effects of autonomy and independence the NAS report [39] seeks for forensic science, but also serve to alleviate the issues surrounding the unusually high levels of reported harassment in the field as well.
Forensic science can often be a source of conflict not only because of the demographics of the workforce but also due to the stark differences between high-alert police investigative practices and the methodical pace of scientific analysis. Managerial and supervisory forensic scientists are more frequently exposed to these conflicting desires. They are also more involved in creating quality assurance policies and ensuring consistent rule-following by their subordinates. Police investigators are frequently frustrated by what they perceive to be the result—an unnecessarily slow process and a “lack of urgency” by the forensic science service providers they must work with. This conflict has led to many high-profile examples where laboratory directors, supervisors, or managers may have been terminated or demoted to achieve what police investigators believe will be greater levels of compliance with their emergent investigatory needs and less compliance with perceived “overburdensome” regulatory rules.
The pressures experienced by mostly female forensic scientists to suit the urgent needs of a mostly male police agency may explain some of the harassment experienced by those on the lower rungs of the chain of command. As female managers experience pressure to perform in and for homogeneous workplaces where they are the minority in numbers, educational achievement, and status, especially where it conflicts with important regulatory and accreditation standards as found in forensics, leaders may refer that stress to their subordinates as a coping mechanism. A 2018 survey conducted by the Pew Research Center [27] reported that women employed in majority-male workplaces were less likely to see the fair treatment of other women. This has also been noted concerning female police officers in a male-dominated policing environment. It is easy to infer that this unfair treatment would translate to forensic scientists.
Holt and colleagues [28] reported that female forensic scientists experienced higher stress levels than their male colleagues. These stressors can lead to managers creating stressful and harassing cultures within their hierarchy. Research has shown that stressed-out bosses make unrealistic demands, are abusive, and may try to regain control over their situation by exerting more dominance over their employees [40,41]. The mostly female managerial teams in forensic science may simply be reacting to the daily pressure of working under the high-stress conditions they experience as women scientists in the definitively male workplace of policing.
At typically reported rates of harassment in the workplace, it has been shown by the Workplace Bullying Institute in a 2017 survey [42] “that targets leave their jobs to escape more mistreatment 23 percent of the time; transfer to a different job or location with the same employer 11 percent of the time, or are forced to quit 12 percent of the time because of worse condition”. Our second hypothesis, that harassment would be positively associated with intention to leave or absenteeism while being negatively associated with workplace performance was supported by our data.
Our findings also show that harassment, specifically workplace harassment, is related to work performance, absenteeism and an employee's intention to leave their job. The data demonstrates that as work-related harassment increases, the intention to leave also increases. Interestingly, we found that this intention to leave is mediated by feelings of good or high work performance. Where work performance is perceived to be high, the intention to leave decreases. In other words, when an employee perceives themselves to be doing a good job, even when they are experiencing high levels of harassment, their inclination to leave employment is mitigated.
This mitigating effect could be one of the reasons there are so many women in forensics in the first place. Women in forensics want to feel like they are doing something good for their communities and society. In a 2013 report, The Pew Research Society [43] found that more women than men feel that working in a job that helps society is extremely important. Our data shows that if a forensic scientist feels they are doing good work, even in the face of very high levels of stress and harassment, they are more likely to stay.
Forensic scientists, the majority of whom identify as female, experience higher levels of workplace harassment than most other working adults in the United States and they experience it primarily at the hands of other women. This gendered aspect of the results may be due to the overwhelming number of women working in forensic science. This may also be due to the presence of known harassment risk factors in their workplace—homogeneity and power imbalances. Homogeneity applies on two fronts for forensic scientists as minority members of a police organization as well as from within as forensic science services are mostly female.
Our findings can be important for those in or aspiring to forensic science leadership toward creating welcoming and inclusive workplaces where harassment is taken seriously, and effort is taken to reduce its occurrence. The results of this study show that leaders in forensic science are perceived as perpetrators of harassment. Awareness of this can help leaders identify ways to diminish its occurrence and mitigate its effects by modeling non-harassing workplace behaviors or supporting workers who are experiencing it.
Recommendations for mitigation include training, encouraging management to value and promote employee wellness, and creating a workplace culture that adheres to the concept of psychosocial safety climate (PSC) where, policies, practices, and procedures for the protection of worker psychological health and safety are in place. This concept can be particularly helpful in forensic science as it places high value on properly resourcing and designing the work such that employees feel that they have the capacity to actually perform. This falls to managers to ensure that the work is organized in such a way as to encourage wellness over productivity, stress reduction over urgency, and provide avenues for employees to perform work feeling supported and valued. A recent study on the concept shows that PSC is negatively associated with harassment, bullying, discrimination and abuse and positively associated with well-being, personal resilience and hope [44].
An easy way to begin creating a workplace with the PSC mindset is to train managers to learn about it and write policies in their units to support it. These can be things such as offering debriefs following difficult crime scenes, providing peer support counselors, ensuring adequate rest hours between shifts by limiting overtime, or allowing team members a few hours a month to dedicate to personal wellness. A more difficult, but crucial step, is for top management to focus efforts on obtaining more resources for their labs and employees. This can, of course, come in the form of more staffing and more funding for their work. However, management can also do more with what they already have by doing such things as creating small “down-time” rooms, tapping into officer wellness programs that also support non-sworn law enforcement, creating more avenues for two-way communication between management and employees, and asking for more recognition from the commanders and stakeholders who use forensic science services.
One strong mitigating factor identified in the study is when scientists report that they feel they are doing a good job, they are less likely to leave their employment even in the face of high levels of harassment. This may be a result of forensic scientists' community service mentality, their desire to help victims of crime, and evidence of the effect of PSC. Forensic science leadership can work to ensure that employees feel more valued and are properly resourced to do good work. By being connected to the impact of their work in the community and having a sense of support from these groups, forensic scientists may have greater resilience in the face of instances of harassment. Connection to the overall mission can help employees to feel that the work they are doing is important and valued. This can be achieved in several ways while simultaneously considering the impact of bias and reducing direct knowledge of case specifics. Methods such as community service activities, outreach to STEM students, and recognition of group achievements can be employed to demonstrate value to forensic employees.
This link between job satisfaction and resilience shows that forensic science leaders should strive to create opportunities for employees to find value and meaning in the performance of their duties in an effort to offset perceived harassment in the workplace. Importantly, the results of the present study can lead to improvements in policies and procedures that deal with harassment and bullying in forensic science laboratories. There is also a benefit to society to have workers in forensic science laboratories who feel secure and happy in their jobs, without having to deal with the stress of being bullied by peers, subordinates, or management/supervisors. Studies have shown that stress can lead to work mistakes—and in the field of forensic science, mistakes can have significant negative impact on society. According to Almazrouei and colleagues [45], acute elevated or repeated exposure to stress impairs cognitive abilities and well-being of forensic scientists. Armed with this awareness, those in leadership can begin to take ownership and work to end harassment that may exist withing their organizations. This can be a powerful tool toward boosting employee retention, feelings of career contentment, and an overall sense of professional well-being within a forensic science service.
Rana DellaRocco: Writing – review & editing, Writing – original draft, Visualization, Project administration, Methodology, Investigation, Conceptualization. Ryan D. Krone: Writing – review & editing, Writing – original draft, Visualization, Validation, Methodology, Investigation, Formal analysis, Data curation. Nancy L. Wayne: Writing – review & editing, Writing – original draft, Visualization, Supervision, Resources, Project administration, Methodology, Investigation, Funding acquisition, Data curation, Conceptualization.
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.