Bias-Based Bullying in California Schools: The Impact of the 2016 Election Cycle

Abstract

The purpose of this study is to examine the trends in bias-based bullying between 2013 and 2019 among California youth overall and by type of bias-based bullying and explore the extent to which Trump’s announcement of his candidacy for U.S. President in June 2015 impacted these bullying outcomes. We pooled the student-level survey data from multiple waves of the California Healthy Kids Survey. The final study sample included 2,817,487 middle- and high-school students (48.3% female, 47.9% male, and 3.7% not reported). We examined five specific types of bias-based bullying and any bias-based bullying overall. We employed logistic regression and calculated odds ratios to compare differences in the odds of bias-based bullying before and after Trump announced his candidacy for U.S. President. Between 2013 and 2019, approximately one in four students reported experiencing at least one type of bias-based bullying, based on race, ethnicity, or national origin being the most commonly reported. Trump’s announcement for candidacy was inconsistently associated with differences in the odds of bias-based bullying. Counties in which a higher proportion of the electorate voted for Trump had slightly higher odds of bullying for any bias-based bullying and for all specific types of bias-based bullying. Findings highlight the need for a commitment to protecting students from bullying regardless of their identity. Public health and education researchers and practitioners should draw on our growing understanding of the different dimensions of bullying in designing, implementing, and evaluating intervention approaches that address bias-based bullying, a particularly important cause given the growing polarization in the United States and the increasing salience of identity in the lead-up to and since the 2016 and 2020 elections.

Keywords

bullying media and violence violence exposure youth violence

Introduction

Bias-Based Bullying

Bullying has long been considered a serious issue faced by school-aged youth. It is generally defined as a form of aggressive behavior that occurs repeatedly; intended to cause physical, social, or psychological harm; and characterized by an imbalance of power between the parties involved (Smith, 2016). Bullying is linked to negative short- and long-term physical and mental health, psychosocial, and academic outcomes (Benedict et al., 2015; Eisenberg et al., 2003; Hong & Espelage, 2012; Juvonen et al., 2003, 2011; Li et al., 2020; McDougall & Vaillancourt, 2015; Smith, 2016). While studies of general bullying prevalence suggest either a downward trend or no change over time (Kennedy, 2021; Li et al., 2020), bias-based bullying as a specific, distinct form of bullying has emerged as a growing concern in recent years (Horton, 2021).

Bias-based bullying refers to bullying based on dimensions of an individual’s identity, such as their actual or perceived race, ethnicity, national origin, religion, gender, sexual orientation, and/or disability status. Data on bias-based bullying are less widely available than those on general bullying; however, some estimates of the prevalence of bias-based bullying are as high as 35%–45% (Eisenberg et al., 2021; Russell et al., 2012). In comparison, a recent study of “traditional” forms of bullying estimated its prevalence at approximately 20%, with no significant changes in prevalence between 2011 and 2019 (Li et al., 2020). In addition, evidence suggests that bias-based bullying is more strongly associated with poor outcomes. Youth reporting experiences of bias-based bullying had higher odds of substance use and poor mental health during adolescence than did victims of non-bias-based bullying (Russell et al., 2012). These students also report poorer grades and lower school connectedness, and experiencing bias-based bullying is associated with increased absenteeism from school due to feeling unsafe (Baams et al., 2017; Nishina et al., 2005).

Impact of the Social Context

Much of the bullying research literature has drawn on the social ecological model to understand this phenomenon (Bronfenbrenner, 1977, 1979). However, most studies focus on the microsystem (e.g., individual traits, home environment) and mesosystem (e.g., school environment) in their exploration of bullying participation and victimization (Horton, 2021). In contrast, the broader levels of the social ecological model are less frequently considered in empirical studies of bullying, despite their important contributions to the behavior (Newman & Fantus, 2015); these levels include the exosystem (e.g., community and media environments), the macrosystem (e.g., broader social environment, dominant culture, and norms), and the chronosystem (i.e., key events and changes over time) (Horton, 2021).

The few studies that have considered broader levels of influence highlight just how important social context can be for bullying outcomes. For example, Hatzenbuehler et al. (2019) found evidence of higher rates of homophobic bullying among California youth during the lead-up to a voter referendum, negatively targeting the LGBTQ+ community. The 2016 presidential election of Donald Trump represents another political event that contributed enormously to the exosystem, macrosystem, and chronosystem and thereby had potential to affect bias-based bullying behaviors. The election was covered extensively by the media, a key component of the exosystem, and a position of power such as the presidency certainly shapes cultural norms, which is a key component of the macrosystem. According to the social ecological model, it follows that acceptance or tolerance of a presidential candidate’s violent and divisive language and actions in the media could directly or indirectly affect the beliefs and behavior of the school-aged youth in those communities. Conversely, critiques and rejections of the candidate’s behavior could equally signal that these behaviors or views are not acceptable within the dominant culture and also have an effect on the youth.

In the wake of the 2016 presidential election, anecdotal evidence of increased bullying (Carroll, 2016; Loeb, 2016; Pollock, 2017) and educators’ concerns about bullying (Costello, 2016; Pollock, 2017) has been explored empirically to some extent. A study of middle schoolers in Virginia found that while there were no meaningful differences in bullying between Democratic- and Republican-leaning counties prior to the election, there were higher adjusted rates of bullying in counties supporting the Republican candidate after the 2016 election (Huang & Cornell, 2019). In 2018, LGBTQ+ students in more politically conservative districts in Washington State reported experiencing more bullying and less teacher intervention than students in less politically conservative areas (Hobaica et al., 2021). A qualitative study of undocumented students in California reported increased experiences with bullying around the time of the 2016 election (Valdivia et al., 2021), and a qualitative participatory study with Muslim children highlighted the impact of anti-Muslim rhetoric during the election on their experiences with and understanding of Islamophobia (Farooqui & Kaushik, 2021). Finally, this social context resulted in identity-based aggressive behavior beyond bullying; analyses of newspaper coverage of hate crimes, as well as national hate crime statistics found as high as a 226% increase in their incidence after election rallies supporting Trump and since his election in 2016 (Feinberg et al., 2019; Levin et al., 2018; Rushin & Edwards, 2018; Warren-Gordon & Rhineberger, 2021). Taken together, these findings suggest that exosystem-, macrosystem-, and chronosystem-level factors can be key contributors to bullying trends, and additional research is needed to understand the extent to which these factors can impact bullying participation and victimization.

The Present Study

In this study, we used data from a statewide survey of middle- and high-school students in California to examine trends in specific types of bias-based bullying among youth, and to assess whether and to what extent the 2016 election cycle impacted bias-based bullying outcomes among these youth. The specific types examined included bullying based on (1) race, ethnicity, or national origin; (2) religion; (3) gender; (4) sexual orientation; and (5) disability, as well as bias-based bullying overall, all of which align with the definition of bias-based bullying provided above. Given the theoretical and empirical evidence underpinning the role of broader levels of the social ecological model and their potential to affect bullying outcomes, as well as the impact of the 2016 election on the media and social environment, we hypothesized that bias-based bullying would be positively associated with the election. Trump’s official announcement of his candidacy for president, which took place in June 2015, was used as the “index date” for this analysis, allowing us to compare bias-based bullying during two school years before the announcement to four school years after the announcement. As divisiveness has continued during and beyond the 2020 presidential election, this study provides an important insight into the extent to which racist, homophobic, xenophobic, and other election rhetoric may result in “trickle-down bullying” (Johnson, 2017, p. 448) in a large diverse state.

Methods

Data and Study Sample

To examine bias-based bullying, secondary data for middle- and high-school students who completed at least one wave of the California Healthy Kids Survey (CHKS; WestEd, 2022) during the study period were used. The CHKS is an anonymous survey fielded by California schools each academic year that asks questions that are relevant for guiding school improvement, including questions on student demographics, school connectedness, school culture and climate, and school safety. Specific information regarding data collection processes is provided elsewhere (WestEd, 2022), as are details about the validity and reliability of the instrument (Mahecha & Hanson, 2020). Student-level data from CHKS were linked with county-level 2016 presidential election returns data from MIT Election Lab (MIT Election Data and Science Lab, 2018) to assess the political affiliation of counties in which CHKS schools were seated. The original CHKS data included 3,111,403 students who completed at least one survey wave between the 2013–2014 and 2018–2019 school years. From this, students were excluded from the sample if they did not answer at least one of the CHKS bias-based bullying questions (n = 182,058, 5.85% of original sample), or if they had missing school connectedness data or their school could not be linked to county-level election returns data (n = 111,858, 3.60% of original sample), resulting in a final study sample which comprised of 2,817,487 middle- and high-school students.

Variables

Main dependent variables

We examined five specific types of bias-based bullying, including bullying based on (1) race, ethnicity, or national origin, (2) religion, (3) gender, (4) sexual orientation, and (5) disability; we also examined any bias-based bullying overall. The CHKS asks students how many times in the last 12 months they were harassed or bullied on school property for the listed reasons. Students were provided the following explanation of bullying: “You were bullied if you were shoved, hit, threatened, called mean names, teased, or had other unpleasant physical or verbal things done to you repeatedly or in a sever way. It is not bullying when two students of about the same strength or power quarrel or fight.” The five specific types of bullying were treated as binary variables and coded as “yes” if a student reported experiencing one or more incidents of that type of bullying in the last 12 months, and “no” if they reported experiencing no incidents of that type of bullying in the last 12 months. From these five binary variables, we created the sixth binary variable that was coded as “yes” if a student reported experiencing any incident of bias-based bullying in the last 12 months, and “no” if otherwise. Some students did not have complete information for all of the individual bias-based bullying questions. Therefore, sample sizes for the five specific types of bias-based bullying are slightly smaller than the sample size for any bias-based bullying. The CHKS bias-based bullying measure is validated and has demonstrated good internal consistency reliability (Mahecha & Hanson, 2020). Past-year measures of bullying experiences are commonly used (Hamburger et al., 2011), although they may be susceptible to recall bias, as discussed in the limitations below, particularly because bullying tends to be underreported. We examined bullying as a dichotomous variable to align with previous studies of bullying and the social context, including in the wake of the 2016 election (Hatzenbuehler et al., 2019; Hobaica et al., 2021; Huang & Cornell, 2019).

Independent variables

Our main independent variable was the year in relation to Trump’s candidacy announcement for U.S. President. Baseline (coded as 1) included the 2 years prior to Trump’s candidacy announcement (2013–2014 and 2014–2015 school years); early-post (coded as 2) included the first years directly after Trump’s candidacy announcement (2015–2016 and 2016–2017 school years); and later-post (coded as 3) for the second 2 years after Trump’s candidacy announcement (2017–2018 and 2018–2019 school years).

Control variables

We included demographic variables for sex (male, female, and not reported); sexual orientation (straight, gay/lesbian/bisexual, other, not sure, and not reported); transgender status (no, yes, not sure, and not reported); and race/ethnicity (Hispanic/Latino and non-Hispanic White, Black/African American [AA], Asian, American Indian/Alaskan Native [AI/AN], Native Hawaiian/Pacific Islander [NH/PI], Multiple Race, and not reported). We also included education variables for grade level (6th grade through 13th grade and not reported); and grades received in school (mostly A’s, A’s and B’s, mostly B’s, B’s and C’s, mostly C’s, C’s and D’s, mostly D’s, mostly F’s, and not reported). In addition, we included socioeconomic status variables for living arrangement (parent’s/guardian’s home, relative’s home, multifamily home, friend’s home, foster home/group care, hotel/motel, transitional/temporary housing, other, and not reported); and parent education level (did not finish high school, graduated high school, some college, graduated college, and don’t know/not reported). Next, following the approach outlined by WestEd—a nonpartisan, nonprofit agency that develops measurement instruments on school climate—we included a variable on school connectedness (categories included low, moderate, and high connectedness), which measures a student’s sense of belonging in school (Austin et al., 2011). This measure has been validated and has demonstrated internal reliability (Mahecha & Hanson, 2020). Finally, we included a continuous variable for the proportion of the county in which the school was seated that voted for the Republican candidate during the 2016 U.S. Presidential election, and a continuous time trend variable for school year.

Statistical Analyses

Descriptive analysis

To examine demographic characteristics of our study sample, we computed cell sizes and percentages for categorical variables and means and standard deviations for continuous variables. We descriptively examined bias-based bullying by computing cell sizes and proportions of the study sample who reported experiencing at least one incident of bullying for each of our six bias-based bullying outcomes overall, and by school year and period (pre-period: 2013–2014 and 2014–2015 school years; early-post: 2015–2016 and 2016–2017 school years; and later-post: 2017–2018 and 2018–2019 school years). To examine unadjusted differences in proportions of each type of bullying by period, we conducted chi-squared tests. Calculating Pearson’s correlation coefficients, we also examined associations between our six bias-based bullying outcomes.

Regression analysis

We employed logistic regression and calculated odds ratios to compare differences in the odds of bias-based bullying from the 2-year period before Trump announced his candidacy for U.S. President (reference period: 2013–2014 and 2014–2015) to the early-post (2015–2016 and 2016–2017) and later-post (2017–2018 and 2018–2019) 2-year periods after he announced his candidacy. In all regression models, we controlled for the covariates described above. For all demographic covariates, missing values were included in analysis; we also conducted a sensitivity analysis with the missing values removed. In addition, we ran two sets of stratified analyses, one in which we stratified models by county voting status (counties where Trump did and did not win a plurality of the vote) and another in which we stratified models by students’ grade level (middle school [grades 6–8] and high school [grades 9–13]). To address potential heteroskedasticity, in all regression models we estimated robust standard errors (Huber, 1967); statistical significance was determined at the traditional 5% alpha level. All analyses were performed using Stata version 14.2 (StataCorp, 2015). This study was reviewed and approved by the Providence St. Joseph Health Institutional Review Board.

Results

Characteristics of the Study Sample

Sample characteristics are presented in Table 1. Students in the study sample were approximately evenly split between male and female sex, and the majority identified as straight and not transgender. Students were racially/ethnically diverse, and the majority of students received passing grades and had moderate- to high-school connectedness. A plurality of students reported living at home with one or more parents/guardians, while less than half reported that their parents graduated college.

Table 1.

Characteristics of the Study Sample.

Characteristic	Frequency	Percent
Characteristic	(N = 2,817,487)	(100.00)
Sex
Male	1,350,075	47.92
Female	1,362,699	48.37
Not reported	104,713	3.72
Sexual orientation
Straight	2,114,917	75.06
Gay/lesbian/bisexual	187,225	6.65
Not sure	151,375	5.37
Other	18,013	0.64
Not reported	345,957	12.28
Transgender status
Not transgender	2,682,193	95.20
Transgender	37,664	1.34
Not sure	18,076	0.64
Not reported	79,554	2.82
Race/ethnicity
Non-Hispanic White	604,655	21.46
Non-Hispanic Black	102,537	3.64
Non-Hispanic Asian	303,266	10.76
Non-Hispanic AIAN	31,359	1.11
Non-Hispanic NHPI	39,346	1.40
Non-Hispanic multiracial	286,322	10.16
Hispanic	1,422,302	50.48
Not reported	27,700	0.98
Percent of county voted Republican (mean (SD))	—	33.93 (11.55)
Majority (≥50%) of county voted Republican	268,690	9.54
Grade level
6th grade	23,330	0.83
7th grade	869,555	30.86
8th grade	66,564	2.36
9th grade	844,046	29.96
10th grade	111,296	3.95
11th grade	716,769	25.44
12th grade	87,138	3.09
13th grade	76,327	2.71
Not reported	22,462	0.80
Grades received in school
Mostly A’s	572,070	20.30
A’s and B’s	934,709	33.18
Mostly B’s	233,406	8.28
B’s and C’s	586,330	20.81
Mostly C’s	147,367	5.23
C’s and D’s	218,059	7.74
Mostly D’s	44,489	1.58
Mostly F’s	63,115	2.24
Not reported	17,942	0.64
School connectedness
Low	296,913	10.54
Moderate	1,142,398	40.55
High	1,378,176	48.92
Living arrangement
Home w/1 + parents/guardians	2,266,648	80.45
Other relative’s home	50,836	1.80
Home w/2 + families	102,907	3.65
Friend’s home	9,310	0.33
Foster/group home, waiting placement	10,459	0.37
Hotel or motel	6,230	0.22
Transitional/temporary housing	13,182	0.47
Other living arrangement	68,332	2.43
Not reported	289,583	10.28
Parent education level
Did not finish high school	381,333	13.53
Graduated high school	441,011	15.65
Some college	348,885	12.38
Graduated college	1,080,036	38.33
Don’t know/not reported	566,222	20.10
Academic year
2013–2014	501,905	17.81
2014–2015	314,597	11.17
2015–2016	515,384	18.29
2016–2017	434,961	15.44
2017–2018	590,025	20.94
2018–2019	460,615	16.35

AIAN = American Indian/Alaskan Native; NHPI = Native Hawaiian/Pacific Islander.

Bias-Based Bullying, Overall, and by Period and School Year

Overall, about one quarter of students reported experiencing any type of bias-based bullying in a given school year during the study period (Table 2). The most commonly reported type of bias-based bullying was that which is based on race, ethnicity, or national origin, followed by sexual orientation, gender, religion, and disability. This finding was held when examining bias-based bullying by period (pre-period, early-post, and late-post). Compared with the pre-period, the prevalence of any bias-based bullying and each type of bias-based bullying was slightly less in the early- and late-post periods. While students may have experienced multiple types of bias-based bullying during the study period, none of these types strongly correlated with each other (Supplemental Appendix Table 1).

Table 2.

Prevalence of Bias-Based Bullying Overall and by Type of Bullying.

Type of Bullying		By Period			p Value
	Overall	Pre-Period	Early-Post	Late-Post
	N (%)	N (%)	N (%)	N (%)
Any bias-based bullying	691,692 (24.55)	214,320 (26.25)	231,060 (24.31)	246,312 (23.44)	<0.001
Race/ethnicity/national origin	422,159 (14.98)	134,010 (16.49)	140,645 (14.83)	147,504 (14.06)	<0.001
Religion	210,075 (7.46)	71,252 (8.79)	71,103 (7.53)	67,720 (6.47)	<0.001
Gender	224,032 (7.95)	67,419 (8.32)	78,561 (8.31)	78,052 (7.50)	<0.001
Sexual orientation	248,500 (8.82)	74,959 (9.24)	80,311 (8.48)	93,230 (8.90)	<0.001
Disability	135,929 (4.82)	43,775 (5.40)	43,381 (4.59)	48,773 (4.67)	<0.001

Notes. For any bias-based bullying, statistics (frequencies, percentages, p values) are for the full study sample (N = 2,817,487). For the five specific types of bias-based bullying, statistics are for the sample that responded to each item. Overall, 0.27% of the sample (n = 7,468) did not respond to the item on race/ethnicity/national origin bullying, 0.59% (n = 16,697) religion-based bullying, 0.75% (n = 21,106) gender-based bullying, 0.45% (n = 12,618) sexual orientation-based bullying, and 0.64% (18,117) disability-based bullying. Chi-squared tests were used to examine unadjusted differences in proportions of each type of bullying by period.

From the 2013–2014 to 2018–2019 school year, the proportion of students who reported experiencing each type of bias-based bullying decreased, albeit by small amounts (data not shown). Bullying based on race, ethnicity, or national origin decreased by the largest amount (from 16.7% in 2013–2014 to 13.3% in 2018–2019), followed by bullying based on religion (from 9.0% in 2013–2014 to 6.1% in 2018–2019), gender (from 8.1% in 2013–2014 to 7.0% in 2018–2019), and disability (from 5.6% in 2013–2014 to 4.6% in 2018–2019). Bullying based on sexual orientation decreased the least during the study period (from 9.4% in 2013–2014 to 8.9% in 2018–2019).

For two types of bias-based bullying, decreases primarily occurred prior to the Trump U.S. Presidential candidacy announcement in June 2015, and either remained the same or increased afterward. For bullying based on sexual orientation, the proportion of students who reported experiencing bullying was 9.4% in the 2013–2014 school year and dropped to 8.4% in the 2015–2016 school year. After that, the rate began to increase and reached 8.9% in the 2018–2019 school year. For bullying based on disability, the proportion went from 5.6% in the 2013–2014 school year to 4.6% in the 2015–2016 school year, then hovered between 4.6% and 4.7% in the two remaining school years.

Trump Candidacy Announcement and Odds of Bias-Based Bullying

After controlling for covariates, the Trump Presidential candidacy announcement was inconsistently associated with differences in the odds of bias-based bullying among students (Table 3). For any bias-based bullying, odds of bullying in the first 2 years after the candidacy announcement were not significantly different than that of the 2 years prior to the candidacy announcement but were slightly higher during the second 2-year period after the candidacy announcement. For bullying based on race, ethnicity, or national origin and religion, there were no significant differences in the odds of bullying during either the first or second 2-year periods after the candidacy announcement. For gender-based bullying, there was a significant increase in the odds of bullying in the first 2-year period after the candidacy announcement, but a significant decrease in the second 2-year period after the candidacy announcement. Conversely, for sexual orientation- and disability-based bullying, there were decreases in the odds of bullying in the first 2-year period after the candidacy announcement but increases in the second 2-year period after the candidacy announcement. Sensitivity analyses conducted with missing values removed revealed only changes in effect sizes; no differences in the direction of effects were observed for any outcomes (Supplemental Appendix Tables 2 and 3).

Table 3.

Odds (Standard Error) of Bias-Based Bullying Associated with Trump’s Presidential Candidacy Announcement, by Type of Bullying.

	Any	Race, Ethnicity, and National Origin	Religion	Gender	Sexual Orientation	Disability
	OR (SE)	OR (SE)	OR (SE)	OR (SE)	OR (SE)	OR (SE)
Post-Trump’s presidential candidacy announcement periods (reference 2013–2014 and 2014–2015)
Period 1 (early-post), 2015–2016 and 2016–2017	0.99 (0.0074)	0.997 (0.0088)	0.998 (0.0119)	1.046*** (0.0126)	0.897*** (0.0108)	0.946*** (0.0141)
Period 2 (later-post), 2017–2018 and 2018–2019	1.031* (0.0131)	1.008 (0.0153)	0.97 (0.0200)	0.957* (0.0197)	1.066** (0.0217)	1.062* (0.0272)
Annual time trend	0.940*** (0.0028)	0.949*** (0.0033)	0.926*** (0.0044)	0.959*** (0.0045)	0.934*** (0.0044)	0.947*** (0.0056)
Percent of county voted Republican in 2016 presidential election	1.004*** (0.0001)	1.003*** (0.0002)	1.006*** (0.0002)	1.001** (0.0002)	1.005*** (0.0002)	1.002*** (0.0003)
Sex (reference male)
Female	1.136*** (0.0034)	0.970*** (0.0035)	0.985** (0.0048)	2.116*** (0.0107)	0.811*** (0.0040)	0.903*** (0.0056)
Not reported	0.938*** (0.0076)	0.847*** (0.0082)	0.930*** (0.0121)	1.514*** (0.0192)	0.848*** (0.0106)	0.979 (0.0148)
Sexual orientation (reference straight)
Gay/lesbian/bisexual	3.585*** (0.0193)	1.489*** (0.0097)	1.506*** (0.0129)	2.902*** (0.0210)	13.45*** (0.0863)	3.218*** (0.0287)
Not sure	1.531*** (0.0093)	1.109*** (0.0083)	1.377*** (0.0130)	1.966*** (0.0170)	3.209*** (0.0266)	2.194*** (0.0239)
Other	2.436*** (0.0388)	1.382*** (0.0266)	1.572*** (0.0369)	3.070*** (0.0600)	6.716*** (0.1200)	3.273*** (0.0754)
Not reported	1.168*** (0.0055)	0.995 (0.0057)	1.211*** (0.0089)	1.346*** (0.0100)	1.904*** (0.0138)	1.560*** (0.0142)
Transgender (reference no)
Yes	1.980*** (0.0249)	2.083*** (0.0258)	3.292*** (0.0440)	4.646*** (0.0614)	2.098*** (0.0306)	3.735*** (0.0527)
Not sure	1.814*** (0.0298)	1.442*** (0.0270)	1.985*** (0.0431)	2.480*** (0.0480)	2.319*** (0.0436)	2.517*** (0.0545)
Not reported	1.073*** (0.0099)	1.043*** (0.0118)	1.204*** (0.0173)	1.392*** (0.0190)	1.213*** (0.0157)	1.322*** (0.0214)
Race/ethnicity (reference Non-Hispanic White)
Non-Hispanic Black	1.234*** (0.0098)	2.401*** (0.0211)	0.863*** (0.0108)	0.804*** (0.0103)	0.807*** (0.0105)	0.800*** (0.0124)
Non-Hispanic Asian	1.228*** (0.0066)	2.447*** (0.0153)	1.057*** (0.0088)	0.723*** (0.0065)	0.773*** (0.0071)	0.681*** (0.0083)
Non-Hispanic AIAN	1.056*** (0.0143)	1.443*** (0.0234)	1.245*** (0.0233)	1 (0.0203)	0.964 (0.0195)	1.106*** (0.0257)
Non-Hispanic NHPI	1.325*** (0.0157)	2.161*** (0.0288)	1.112*** (0.0200)	1.033 (0.0191)	1.061** (0.0205)	0.988 (0.0231)
Non-Hispanic multiracial	1.166*** (0.0062)	1.780***(0.0116)	0.948*** (0.0079)	0.967*** (0.0080)	1.011 (0.0084)	0.982 (0.0102)
Hispanic	0.836*** (0.0034)	1.249*** (0.0066)	0.697*** (0.0045)	0.745*** (0.0048)	0.818*** (0.0053)	0.744*** (0.0060)
Not reported	0.949*** (0.0141)	1.479*** (0.0258)	1.122*** (0.0237)	0.737*** (0.0179)	0.744*** (0.0179)	0.766*** (0.0224)
Grade level (reference 7th grade)
6th grade	0.944*** (0.0151)	0.853*** (0.0171)	1.095*** (0.0280)	1.105*** (0.0269)	0.938** (0.0229)	1.033 (0.0336)
8th grade	0.960*** (0.0090)	1.015 (0.0113)	1.036* (0.0160)	0.942*** (0.0142)	0.913*** (0.0132)	1.087*** (0.0205)
9th grade	0.812*** (0.0033)	0.890*** (0.0044)	0.980** (0.0067)	0.815*** (0.0054)	0.675*** (0.0044)	0.998 (0.0086)
10th grade	0.674*** (0.0055)	0.754*** (0.0073)	0.920*** (0.0119)	0.741*** (0.0097)	0.534*** (0.0071)	0.955** (0.0150)
11th grade	0.659*** (0.0029)	0.740*** (0.0039)	0.918*** (0.0067)	0.745*** (0.0053)	0.467*** (0.0034)	0.873*** (0.0080)
12th grade	0.548*** (0.0052)	0.626*** (0.0071)	0.878*** (0.0127)	0.726*** (0.0105)	0.421*** (0.0066)	0.947** (0.0165)
13th grade	0.494*** (0.0051)	0.581*** (0.0071)	0.887*** (0.0135)	0.724*** (0.0114)	0.475*** (0.0074)	1.113*** (0.0185)
Not reported	0.991 (0.0155)	1.164*** (0.0195)	1.637*** (0.0311)	1.369*** (0.0286)	1.055* (0.0231)	1.956*** (0.0431)
Grades received in school (reference mostly A’s)
A’s and B’s	1.023*** (−0.00427)	1.020*** (−0.00509)	0.963*** (−0.00651)	0.890*** (−0.00587)	1.027*** (−0.00716)	1.136*** (−0.0107)
Mostly B’s	1.066*** (−0.00648)	1.054*** (−0.00763)	0.997 (−0.00975)	0.943*** (−0.00911)	1.107*** (−0.0109)	1.322*** (−0.0168)
B’s and C’s	1.007 (−0.00489)	0.993 (−0.00576)	0.950*** (−0.00746)	0.833*** (−0.00647)	1.077*** (−0.00851)	1.333*** (−0.0138)
Mostly C’s	1.002 (−0.00741)	0.987 (−0.00866)	0.957*** (−0.0113)	0.869*** (−0.0103)	1.128*** (−0.0131)	1.462*** (−0.0209)
C’s and D’s	1.040*** (−0.00666)	0.995 (−0.0076)	0.986 (−0.0101)	0.870*** (−0.0089)	1.221*** (−0.0121)	1.554*** (−0.0195)
Mostly D’s	1.048*** (−0.0125)	1.004 (−0.0141)	0.996 (−0.0184)	0.907*** (−0.0169)	1.237*** (−0.0217)	1.659*** (−0.034)
Mostly F’s	1.142*** (−0.0115)	1.180*** (−0.0131)	1.415*** (−0.019)	1.222*** (−0.0172)	1.496*** (−0.0211)	2.182*** (−0.0344)
Not reported	1.004 (−0.0185)	0.977 (−0.0215)	1.072* (−0.0292)	0.954 (−0.0272)	1.160*** (−0.0325)	1.488*** (−0.0488)
School connectedness (reference high)
Low	2.747*** (−0.0129)	2.893*** (−0.0155)	2.901*** (−0.02)	3.358*** (−0.0234)	3.128*** (−0.0218)	3.885*** (−0.0328)
Moderate	1.750*** (−0.00561)	1.770*** (−0.00686)	1.605*** (−0.00862)	1.857*** (−0.01)	1.789*** (−0.00946)	1.887*** (−0.0134)
Living arrangement (reference home w/1+ parents/guardians)
Other relative’s home	1.205*** (−0.0126)	1.213*** (−0.0146)	1.224*** (−0.0196)	1.266*** (−0.0196)	1.288*** (−0.0196)	1.437*** (−0.0255)
Home w/2+ families	1.125*** (v0.00862)	1.113*** (−0.00994)	1.268*** (−0.0149)	1.189*** (−0.0142)	1.110*** (−0.0131)	1.249*** (−0.0177)
Friend’s home	1.582*** (−0.0363)	1.623*** (−0.041)	2.205*** (−0.0606)	2.148*** (−0.0633)	2.050*** (−0.0628)	2.648*** (−0.0805)
Foster/group home, waiting placement	1.554*** (−0.0342)	1.646*** (−0.0392)	2.018*** (−0.0546)	1.837*** (−0.0528)	1.872*** (−0.0553)	2.350*** (−0.0706)
Hotel or motel	1.720*** (−0.0484)	1.904*** (−0.0551)	2.845*** (−0.0889)	2.500*** (−0.0882)	2.206*** (−0.0817)	2.980*** (−0.106)
Transitional/temporary	1.943*** (−0.0383)	2.372*** (−0.0458)	3.097*** (−0.0651)	2.826*** (−0.0669)	2.393*** (−0.0599)	3.225*** (−0.078)
Other living arrangement	1.074*** (−0.00994)	1.076*** (−0.0118)	1.216*** (−0.0173)	1.171*** (−0.0165)	1.057*** (−0.0148)	1.255*** (−0.0209)
Not reported	0.988 (−0.00615)	1.040*** (−0.00766)	1.045*** (−0.0103)	0.950*** (−0.00959)	0.874*** (−0.00866)	1.083*** (−0.0136)
Parent education level (reference did not graduate high school)
Graduated high school	1.003 (−0.00552)	0.977*** (−0.0063)	0.947*** (−0.00848)	1.041*** (−0.00922)	1.032*** (−0.00901)	0.981 (−0.0105)
Some college	1.218*** (−0.00702)	1.150*** (−0.00776)	1.081*** (−0.0101)	1.264*** (−0.0115)	1.237*** (−0.0113)	1.126*** (−0.0127)
Graduated college	1.164*** (−0.00594)	1.089*** (−0.00652)	1.156*** (−0.00942)	1.229*** (−0.0101)	1.123*** (−0.00916)	1.112*** (−0.0112)
Don’t know/not reported	0.817*** (−0.00439)	0.783*** (−0.00496)	0.854*** (−0.0074)	0.867*** (−0.00756)	0.852*** (−0.00722)	0.868*** (−0.00901)
Regression model N	2,817,487	2,810,019	2,800,790	2,796,381	2,804,869	2,799,370
Chi-squared value	180234.7	114199.2	83223.4	152988.9	261814.2	120785.7
Pseudo-R² value	6.42%	4.91%	5.31%	10.30%	17.30%	11.10%

Notes: Robust standard errors are in parenthesis.

AIAN = American Indian/Alaskan Native; NHPI = Native Hawaiian/Pacific Islander; OR = odds ratio; SE = standard error.

p < .05; **p < .01; ***p < .001.

In the analyses stratified by county voting status, results in blue counties (i.e., counties where Trump did not win a plurality of votes) were consistent with the overall findings (Supplemental Appendix Table 5). However, there were a few differences in results in red counties (i.e., counties where Trump won a plurality of votes); for example, red counties saw a decrease in the odds of any bias-based bullying in the first 2-year period and no significant difference during the second 2-year period (Supplemental Appendix Table 4). In the analyses stratified by grade level, results for high school-aged respondents were consistent with the overall findings while the odds of any bias-based bullying decreased during both time periods for middle-school-aged respondents (Supplemental Appendix Tables 6 and 7).

Covariates and Odds of Bias-Based Bullying

Other factors consistently associated with bullying included sexual orientation, transgender status, grade level in school, school connectedness, living arrangement, and proportion of the county that voted Republican. Female students had lower odds of each type of bias-based bullying compared with male students, except for gender-based bullying. Students who identified with a sexual orientation other than straight or who identified as transgender had higher odds of bullying for each type of bias-based bullying, and school connectedness was inversely associated with odds of bullying for each type of bullying. As grade level increased, the odds of each type of bullying generally decreased. Living in a home with one or more parents/guardians was associated with decreased odds of each type of bullying compared with other living arrangements. Factors that were associated with bias-based bullying but inconsistently included race/ethnicity, grades received in school, and parent education level.

Discussion

This study examined trends in bias-based bullying among California youth overall and by type of bias-based bullying and explored the extent to which Trump’s announcement of his candidacy for U.S. President in June 2015 impacted these bullying outcomes. Between 2013 and 2019, approximately one in four students reported experiencing at least one type of bias-based bullying, based on race, ethnicity, or national origin being the most commonly reported. The relatively high proportion of students who identified as part of a marginalized social group based on race/ethnicity, compared with other marginalized social identities (e.g., gender status or sexual orientation), may contribute to this finding. Nonetheless, the fact that nearly a quarter of students reported experiencing any kind of bias-based bullying is notable, given its strong association with poor outcomes, such as poor mental health and worse academic performance (Baams et al., 2017; Nishina et al., 2005; Russell et al., 2012). Previous studies on the prevalence of bias-based bullying in various states have ranged widely (Eisenberg et al., 2021; Mulvey et al., 2018; Russell et al., 2012). A national survey of youth between the ages of 12 and 18 years, found that of those who reported experiences of bullying, approximately 17% and 10% reported bias-based bullying on single or multiple social identities, respectively (Mulvey et al., 2018). Roughly, during the same time period, prevalence of general bullying was approximately 20% nationally and ranged from 18.5% to 23.5% in California (Centers for Disease Control and Prevention, 1991–2019).

We found that Trump’s announcement for candidacy was inconsistently associated with differences in the odds of bias-based bullying. In the two school years immediately following the announcement (fall 2015–spring 2017), there was an increase in the odds of bullying based on gender, and a decrease in the odds of bullying based on sexual orientation or disability. In contrast, in the later time period (fall 2017–spring 2019), there was an increase in the odds of any bias-based bullying, and bullying based on sexual orientation or disability. There was no significant change in the odds of bullying based on race, ethnicity, or national origin or based on religion during either time period. The decrease in the odds of certain types of bias-based bullying during the first time period may be due to the already declining rates of these types of bullying. This time period includes the early days of Trump’s presidential campaign, the Republican primaries and formal nomination, the general election in the fall of 2016, and Trump’s inauguration and first few months in office. While each of these events were significant and exemplified the type of rhetoric in question in the current study, they did not appear to impact bias-based bullying rates in general, with the exception of bullying based on gender which saw an increase in odds during this time period. Interestingly, this was the first presidential election in which a woman received the nomination from a major party. On the other hand, in the later time period examined in this study, which fell entirely during the time when Trump was in office, the increased odds of any bias-based bullying and of certain types could represent validation and acceptance of divisive behaviors once Trump’s position became official and more widely accepted.

Sexual orientation, transgender status, grade level in school, school connectedness, and living arrangement were consistently associated with bullying; these findings were largely consistent with previous research (Eisenberg et al., 2003; Hong & Espelage, 2012; Juvonen et al., 2003; Smith, 2016). In addition, political affiliation of communities in which schools were situated also appeared to contribute to bullying outcomes; in the overall analysis, counties in which a higher proportion of the electorate voted for Trump had slightly higher odds of bullying for any bias-based bullying and for all specific types of bias-based bullying. This finding is consistent with former studies that examined bullying in the Trump context and found more bias-based bullying and less teacher intervention during bullying in conservative localities (Hobaica et al., 2021; Huang & Cornell, 2019). On the other hand, our stratified analyses revealed that the odds of bias-based bullying in red counties decreased during the first 2-year period while they remained unchanged in blue counties. During the second 2-year period, the odds increased only in blue counties. The stratified analyses should be interpreted with caution as stratified models do not provide a direct comparison of bias-based bullying behavior in red versus blue counties, and because roughly only 10% of survey respondents came from schools in red counties. Importantly, it is also possible that schools in red counties differed from those in blue counties on unmeasured characteristics, such as school size and homogeneity of the student body. Furthermore, winning or losing a plurality of votes is unlikely to be an exact measure of the extent to which Trump’s behavior was filtered through the media and adults down to children themselves, providing modeling or signaling shifts in normative beliefs around what is considered acceptable behavior and the value placed on historically marginalized groups (Horton, 2021).

The findings of this study reveal a need to collect more granular data on bias-based bullying across statewide and national surveillance systems to facilitate tracking of trends over time in the same way that is possible for bullying in general. The need for additional data is especially relevant for understanding the influence of the political and social context on bullying trends and behaviors. As Huang and Cornell (2019) point out, it is unclear whether reports of increased bullying in the wake of the 2016 presidential election represent an actual increase in bullying prevalence or a shift in the form of bullying to one that is more readily recognized, given Trump’s behavior during his candidacy and presidency. Additional data would help to shed light on whether there are actual changes in frequency or shifts in reporting of bullying, allowing for better targeting of interventions. In addition, results of the current study demonstrate that social context may influence different forms of bias-based bullying differently, supporting collection of disaggregated data on bias-based bullying.

Importantly, there is a need for additional efforts to mitigate bias-based bullying, given its link to poor health and social outcomes. A situation in which bullying based on a protected category (e.g., race, religion, disability status) interferes with a public school’s legal responsibility to provide education could be considered a civil rights violation (U.S. Department of Health and Human Services, n.d.); thus, schools have both an ethical and legal obligation to address bias-based bullying. The current findings invite further study and attention to bias as a motivation for bullying in the design of anti-bullying programs and policies. While we found that bullying based on race, ethnicity, or national origin was the most common type of bias-based bullying, any efforts to address this kind of bias are complicated by growing nationwide resistance to the inclusion of anti-racism and equity-focused curricula in schools (Alvarez, 2021). Several states have gone as far as advancing legislation to ban educators from teaching about systemic racism (Alvarez, 2021). Previous researchers have identified steps that educators can take, to address bias-based bullying in the context of Trump (Pollock, 2017); for example, condemning incidents of hate and intimidation and countering misinformation by engaging with facts. However, the onus cannot be entirely on classroom teachers. Parents, school leaders, as well as school- and district-level policies must also reinforce and provide support for these efforts, and public health advocates and organizations should play a role in bias-based bullying prevention as well.

Limitations

This study has several limitations to consider, when interpreting the results. First, survey data were self-reported by students, and thus susceptible to social desirability and recall biases, particularly, given the 12-month period asked about on the survey. This limitation is especially relevant given that bullying is often underreported, particularly among racial minority and male students (Lai & Kao, 2018). In addition, while bullying is typically defined as a repeated behavior, our measure of bullying in this study only differentiated between experiencing bullying or not, regardless of the number of incidents. Though bullying based on disability status was included as an outcome in this study, students were not asked whether or not they had a disability; accordingly, statistical models do not account for the presence or absence of a disability. We used pooled cross-sectional data collected in one state, preventing any causal claims and potentially limiting the generalizability of findings to states with different sociopolitical contexts. However, California is a large state with geographic variations (e.g., large urban centers and many rural areas) and a large diverse population. The use of pooled cross-sectional data also limits our ability to account for whether students were surveyed more than once during the study period. In addition, while the CHKS is fielded across the entire state of California, school districts have some leeway in its administration. Specifically, administration in 7th and 9th grades is required to be eligible for a state subsidy, and administration in grade 11 is recommended. As a result, there is an unequal distribution of participation across grade levels; however, there is substantial participation in all grades. Next, Trump’s announcement of his candidacy was selected as the index date in the current analysis; however, relevant bullying rhetoric may have been present at varying levels before and after this announcement. Nonetheless, this announcement was seen as a key moment in the 2016 election cycle and included a well-known example of Trump’s racist rhetoric. In addition, because this date fell in the summer, we were able to examine full school years before and after the index date. Finally, data on sexual orientation, living arrangement, and parent education level were missing at greater rates than other demographic, education, and socioeconomic status variables in our data (12.28%, 10.28%, and 20.10% missing, respectively). Caution should be exercised when interpreting results related to sexual orientation, living arrangement, and parent education level, as missing categories for these three variables were largely statistically significant in regression models.

Conclusion

Overall, our findings underscore the need for a commitment to protecting students from bullying, regardless of their identity and the political attitudes of the community in which a school is located. Findings also highlight the importance of furthering our understanding of the ways in which popular and divisive campaigns, movements, and rhetoric directly or indirectly impact youth. Future research should aim to monitor bullying outcomes to the extent necessary to disentangle changes in prevalence and shifts in types of bullying. Finally, public health and education researchers and practitioners should draw on our growing understanding of the different dimensions of bullying in designing, implementing, and evaluating intervention approaches that address bias-based bullying. Addressing bias as it relates to bullying is essential given the growing polarization in the United States and the increasing salience of identity in the lead-up to and since the 2016 and 2020 elections.

Supplemental Material

sj-docx-1-jiv-10.1177_08862605231162650 – Supplemental material for Bias-Based Bullying in California Schools: The Impact of the 2016 Election Cycle

Supplemental material, sj-docx-1-jiv-10.1177_08862605231162650 for Bias-Based Bullying in California Schools: The Impact of the 2016 Election Cycle by Monique Gill and Diana Govier in Journal of Interpersonal Violence

Footnotes

Acknowledgements

We would like to acknowledge Kimberly Phillips, former Research Associate at CORE, for her contributions to this work. Dr. Gill confirms that she has listed everyone who contributed significantly to the work.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interests with respect to the authorship and/or publication of this article.

Funding

The author(s) received no financial support for the research and/or authorship of this article.

ORCID iDs

Monique Gill

Diana Govier

Supplemental Material

Supplemental material for this article is available online.

Author Biographies

Monique Gill, PhD, MPH, is a Research Scientist at the Center for Outcomes Research and Education. Her research focuses broadly on the social determinants of health, and she is passionate about advancing social change through research and evaluation approaches that are equitable and community engaged.

Diana Govier, PhD, MPH is a Health Economist with the Veterans Health Administration. She dedicated to understanding the conditions that promote improved individual and community well-being, her research explores impacts of policy and systems change on health and social outcomes.

References

Alvarez

(2021). Teaching with an Anti-Racist Lens. NEA Today. https://www.nea.org/advocating-for-change/new-from-nea/teaching-anti-racist-lens

Austin

O’Malley

Izu

(2011). Making Sense of School Climate: Using the California School Climate, Health, and Learning (Cal-SCHLS) Survey System to Inform Your School Improvement Efforts. WestEd.

Baams

Talmage

C. A.

Russell

S. T.

(2017). Economic costs of bias-based bullying. School Psychology Quarterly, 32(3), 422–433. https://doi.org/10.1037/spq0000211

Benedict

F. T.

Vivier

P. M.

Gjelsvik

(2015). Mental health and bullying in the United States among children aged 6 to 17 years. Journal of Interpersonal Violence, 30(5), 782–795. https://doi.org/10.1177/0886260514536279

Bronfenbrenner

(1977). Toward an experimental ecology of human development. American Pscyhologist, 32(7), 513–531.

Bronfenbrenner

(1979). The Ecology of Human Development: Experiments by Nature and Design. Harvard University Press.

Carroll

(2016, June 9). “You were born in a Taco Bell”: Trump’s rhetoric fuels school bullies across US. The Guardian.

Centers for Disease Control and Prevention (CDC). (1991-2019). High School Youth Risk Behavior Survey Data. http://yrbs-explorer.services.cdc.gov/

Costello

M. B.

(2016). The Trump effect: The impact of the presidential campaign on our nation’s schools. Southern Poverty Law Center.

10.

Eisenberg

M. E.

Gower

A. L.

Brown

Nam

Y. S.

Ramirez

M. R.

(2021). School-based diversity education activities and bias-based bullying among secondary school students. Journal of Interpersonal Violence, 37(17–18), NP15992–NP16012. https://doi.org/10.1177/08862605211025016

11.

Eisenberg

M. E.

Neurnark-Sztainer

Perry

C. L.

(2003). Peer Harassment, School connectedness, and academic achievement. Journal of School Health, 73, 311–316.

12.

Farooqui

J. F.

Kaushik

(2021). Understanding islamophobia through the eyes of American Muslim children: Religious bullying and school social work interventions. Child and Family Social Work, 26(3), 454–466. https://doi.org/10.1111/cfs.12828

13.

Feinberg

Branton

Martinez-Ebers

(2019, March 22). Counties that hosted a 2016 Trump rally saw a 226 percent increase in hate crimes. The Washington Post.

14.

Hamburger

M. E.

Basile

K. C.

Vivolo

A. M.

(2011). Measuring bullying victimization, perpretration, and bystander experiences: A compendium of assessment tools. Centers for Disease Control and Prevention, National Center for Injury Prevention and Control. https://www.cdc.gov/violenceprevention/pdf/bullycompendium-a.pdf

15.

Hatzenbuehler

M. L.

Shen

Vandewater

E. A.

Russell

S. T.

(2019). Proposition 8 and homophobic bullying in California. Pediatrics, 143(6), e20182116. https://doi.org/10.1542/peds.2018-2116

16.

Hobaica

Kwon

Reiter

S. R.

Aguilar-Bonnette

Scott

W. D.

Wessel

Strand

P. S.

(2021). Bullying in schools and LGBTQ+ youth mental health: Relations with voting for Trump. Analyses of Social Issues and Public Policy, 21(1), 960–979. https://doi.org/10.1111/asap.12258

17.

Hong

J. S.

Espelage

D. L.

(2012). A review of research on bullying and peer victimization in school: An ecological system analysis. Aggression and Violent Behavior, 17(4), 311–322. https://doi.org/10.1016/j.avb.2012.03.003

18.

Horton

(2021). Building walls: Trump election rhetoric, bullying, and harassment in US schools. Confero: Essays on Education, Philosophy and Politics, 8(1), 7–32.

19.

Huang

F. L.

Cornell

D. G.

(2019). School teasing and bullying after the presidential election. Educational Researcher, 48(2), 69–83. https://doi.org/10.3102/0013189X18820291

20.

Huber

P. J.

(1967). The behavior of maximum likelihood estimates under nonstandard conditions. Proceedings of the Fifth Berkeley Symposium on Mathematical Statistics and Probability, 5(1), 221–233.

21.

Johnson

(2017). Trickle-down bullying and the truly great American response: Can responsible rhetoric in judicial advocacy and decision-making help heal the divisiveness of the trump presidency? Journal of Gender, Social Policy & The Law, 25(4), 445–508.

22.

Juvonen

Graham

Schuster

M. A.

(2003). Bullying among young adolescents: The strong, the weak, and the troubled. Pediatrics, 112, 1231–1237.

23.

Juvonen

Wang

Espinoza

(2011). Bullying experiences and compromised academic performance across middle school grades. Journal of Early Adolescence, 31(1), 152–173. https://doi.org/10.1177/0272431610379415

24.

Kennedy

R. S.

(2021). Bullying trends in the United States: A meta-regression. Trauma, Violence, and Abuse, 22(4), 914–927.

25.

Lai

Kao

(2018). Hit, robbed, and put down (but not Bullied): Underreporting of bullying by minority and male students. Journal of Youth and Adolescence, 47(3), 619–635. https://doi.org/10.1007/s10964-017-0748-7

26.

Levin

Grisham

Reitzel

J. D.

(2018). Hate crimes in Los Angeles 2017-2018: A comparative analysis with other major cities. California State University.

27.

Lian

Xie

(2020). Trends and sex disparities in school bullying victimization among U.S. youth, 2011–2019. BMC Public Health, 20(1), 1–6. https://doi.org/10.1186/s12889-020-09677-3

28.

Loeb

(2016, August 27). Donald Trump and the Crisis of Bullying. Huffington Post.

29.

Mahecha

Hanson

(2020). Measurement structure of the California School Climate Health, and Learning Surveys: Student, staff, and parent surveys. WestEd. https://calschls.org/docs/measurementstructurecalschls_final.pdf

30.

McDougall

Vaillancourt

(2015). Long-term adult outcomes of peer victimization in childhood and adolescence: Pathways to adjustment and maladjustment. American Psychologist, 70(4), 300–310. https://doi.org/10.1037/a0039174

31.

MIT Election Data and Science Lab. (2018). County Presidential Election Returns 2000-2020. In Harvard Dataverse (DRAFT VERS, Vol. 9). Harvard Dataverse. https://doi.org/doi:10.7910/DVN/VOQCHQ

32.

Mulvey

K. L.

Hoffman

A. J.

Gönültaş

Hope

E. C.

Cooper

S. M.

(2018). Understanding experiences with bullying and bias-based bullying: What matters and for whom? Psychology of Violence, 8(6), 702–711. https://doi.org/10.1037/vio0000206

33.

Newman

P. A.

Fantus

(2015). A social ecology of bias-based bullying of sexual and gender minority youth: Toward a conceptualization of conversion bullying. Journal of Gay & Lesbian Social Services, 27(1), 46–63. https://doi.org/10.1080/10538720.2015.988315

34.

Nishina

Juvonen

Witkow

M. R.

(2005). Sticks and stones may break my bones, but names will make me feel sick: The psychosocial, somatic, and scholastic consequences of peer harassment. Journal of Clinical Child & Adolescent Psychology, 34(1), 37–48. https://doi.org/10.1207/s15374424jccp3401_4

35.

Pollock

(2017). Three challenges for teachers in the era of trump. Educational Studies, 53(4), 426–427. https://doi.org/10.1080/00131946.2017.1286590

36.

Rushin

Edwards

G. S.

(2018). The effect of president trump’s election on hate crimes. SSRN Electronic Journal. Advance online publication. https://doi.org/10.2139/ssrn.3102652

37.

Russell

S. T.

Sinclair

K. O.

Poteat

Koenig

B. W.

(2012). Adolescent health and harassment based on discriminatory bias. American Journal of Public Health, 102(3), 493–495.

38.

Smith

P. K.

(2016). Bullying: Definition, types, causes, consequences and intervention. Social and Personality Psychology Compass, 10(9), 519–532. https://doi.org/10.1111/spc3.12266

39.

StataCorp. (2015). Stata Statistical Software: Release 14. StataCorp, LLC.

40.

U.S. Department of Health and Human Services. (n.d.). StopBullying.gov-Federal Laws. https://www.stopbullying.gov/resources/laws/federal

41.

Valdivia

Clark-Ibáñez

Schacht

Duran

Mendoza

(2021). Heightened uncertainty and determination: The “Trump Effect”and college aspirations for undocumented students. Journal of College Access, 6(2), 28–43. https://scholarworks.wmich.edu/jca/vol6/iss2/4

42.

Warren-Gordon

Rhineberger

(2021). The “Trump effect” on hate crime reporting: Media coverage before and after the 2016 presidential election. Journal of Ethnicity in Criminal Justice, 19(1), 25–45. https://doi.org/10.1080/15377938.2021.1895944

43.

WestEd. (2022). CalSCHLS-About the Surveys. https://calschls.org/about/the-surveys/

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.23 MB