High Expectations Across Multiple Domains,Peer Norms,and Physical Dating Violence Among California Adolescents

Abstract

The purpose of this study is, first, to assess whether high expectation messages (from school, home, and community), and peer norms, were associated with physical dating violence victimization (PDV) among a representative sample of California middle and high school students, and second, to assess whether these associations differed by gender and grade level and/or were mediated by self-efficacy. Data from 7th-, 9th-, and 11th-grade respondents of the 2008-2010 California Healthy Kids Survey (CHKS) were analyzed (N = 85,198). CHKS is an anonymous, school-based cross-sectional survey. Logistic regression was used to calculate adjusted odds ratios (ORs) and 95% confidence intervals (CIs) for expectations in each domain (school, community, and home), peers norms, and their cumulative effects on physical dating violence victimization. We examined interactions for expectations and for peer norms by gender and grade level, and tested the mediation effect of self-efficacy. Ten percent of students reported experiencing physical dating violence victimization in the past year. Students who reported high overall expectations (in multiple domains) had significantly lower odds of experiencing dating violence (OR = 0.24, CI = [0.20, 0.28]) compared with those who reported very low expectations. This association held across all expectation domains and peer norms when tested in separate models and also when tested together in a single model. High expectations in the home domain and peer norms showed the lowest odds. Associations between high expectations and dating violence were similar across gender and grade levels. Self-efficacy partially mediated the associations between high expectations and dating violence. Suggestions for future research are presented.

Keywords

dating violence domestic violence youth violence predicting domestic violence

Adolescent dating violence is a serious public health problem. The term dating violence describes physical or sexual violence, threats of violence, or emotional abuse that occurs between two people currently or formerly in a romantic relationship (Centers for Disease Control and Prevention [CDC], 2012). Dating violence is prevalent during adolescence (CDC, 2006, 2010; Foshee et al., 1996; Halpern, Oslak, Young, Martin, & Kupper, 2001; Miller et al., 2011; Smith, White, & Holland, 2003). Approximately 25% to 33% of adolescents report experiencing verbal, physical, emotional, or sexual abuse from a dating partner each year (CDC, 2006; Foshee et al., 1996; Halpern et al., 2001) and about 10% experience physical dating violence specifically (Eaton et al., 2010; Silverman, Raj, & Clements, 2004). However, prevalence estimates vary due to different definitions of dating violence, methodology, and populations studied.

Dating violence is associated with negative health outcomes and unhealthy risk behaviors. High school students who are victims of physical dating violence are more likely to report sexual intercourse and sexual risk behaviors (Halpern et al., 2001; Roberts, Auinger, & Klein, 2005; Silverman et al., 2004; Teitelman, Ratcliffe, Morales-Aleman, & Sullivan, 2008), depression and attempted suicide, heavy drinking, and physical fighting (Eaton, Davis, Barrios, Brener, & Noonan, 2007; Halpern et al., 2001; D. E. Howard, Beck, Kerr, & Shattuck, 2005), and poor school performance (CDC, 2013; Chiodo et al., 2012). Furthermore, adolescents who experience dating violence are more likely to experience partner violence as adults (Smith et al., 2003) and long-term negative health consequences, which increases the odds of hypertension, heart disease, chronic neck and back pain, and psychological distress (Vives-Cases, Ruiz-Cantero, Escriba-Aguir, & Miralles, 2011), as well as limitations in social functioning (McCaw, Golding, Farley, & Minkoff, 2007).

Although past research has focused on the risk characteristics associated with dating violence, there is a dearth of research on protective factors. Understanding protective factors could facilitate the development of more effective prevention programs, as described in the Model of Resiliency (Benard, 1991). Resilience factor theory is a youth development model with an emphasis on strengths and enhancement of individual and environmental protective factors that help individuals to succeed despite adversity (Kumpfer, 2002). While traditionally, resiliency was defined by Ann Masten in the context of an adverse event (Masten, 2001), this concept varies across disciplines and has been applied more broadly in public health. We use the term resilience here to refer to factors that promote an individual’s chances for success in general, not solely in the face of adversarial circumstances. One resilience factor that may be particularly important for youths’ success is high expectation messages (WestEd Health and Human Development Program for the California Department of Education, 2010). High expectation messages are the consistent communication of direct and indirect messages from others that an individual can succeed. These expectations are believed to shape behavioral and health outcomes for youth by influencing their sense of support and self-efficacy (Constantine, Benard, & Diaz, 1999). A second important protective factor may be peer influence through positive peer role modeling, or peer norms. Youth who are surrounded by peers engaged in positive activities are thought to make healthier choices (WestEd Health and Human Development Program for the California Department of Education, 2008-2009).

To our knowledge, the protective influence of high expectation messages on youths’ dating violence has not been examined. Therefore, the current study assessed associations between high expectation messages in multiple domains (i.e., school, community, home) and peer norms on dating violence. Related research on parenting behaviors provides indirect support that high expectation messages promote positive intimate relationships among youth. Studies have demonstrated that parental monitoring protects against dating violence (Foshee, Linder, MacDougall, & Bangdiwala, 2001; Foshee et al., 2011; D. Howard, Qiu, & Boekeloo, 2003), which may in part be explained by the communication of high expectations by parents in addition to parental supervision. High expectation messages could also explain the associations of religious involvement, community norms, and peer norms with dating violence (Chiodo et al., 2012; Foshee et al., 2001; Foshee et al., 2011; D. Howard et al., 2003).

Findings in the peer literature also suggest that peer norms may affect dating violence. Longitudinal research suggests that having friends who are victims of dating violence predicted dating violence perpetration by females; for males, holding attitudes that are accepting of dating violence norms predicted dating violence perpetration (Foshee et al., 2001). Males who report peer pressure not to hit girls cite this as a reason for not perpetrating dating violence (Reeves & Orpinas, 2012). These studies provide support that peer norms may influence dating violence.

High expectation messages may reduce the likelihood of dating violence by promoting the development of a greater sense of self-efficacy. Self-efficacy is the belief in one’s ability to succeed, and is a critical component of developing one’s identity and sense of self (Bandura, 1977; WestEd Health and Human Development Program for the California Department of Education, 2008-2009). Self-efficacy is derived from personal accomplishments, vicarious experience, verbal persuasion, and physiological states (Bandura, 1977). High expectations from others would likely affect self-efficacy through verbal persuasion. The Model of Resiliency suggests that a sense of autonomy and self-efficacy is related to “healthy expectancies” and is a powerful predictor of positive outcomes (Benard, 1991). Youth who consistently receive high expectation messages are more likely to develop strong self-efficacy (Benard, 1991; WestEd, 1995). Self-efficacy has been linked to dating violence in previous research (Chiodo et al., 2012). Girls with lower grades, poorer self-efficacy, and lower community involvement were more likely to be in mutually violent relationships. A strong sense of self appears to be protective against dating violence for females (D. E. Howard et al., 2005). As such, self-efficacy may serve as a mediator (i.e., causal mechanism) between high expectation messages and dating violence.

While research indicates that expectations and peer norms may be associated with dating violence, certain domains (school, home, and community) or peer norms might be uniquely important but have not yet been studied concurrently. Adolescents who receive high expectation messages in certain domains or who are influenced by positive peer norms may be more likely to internalize these messages and thereby hold high expectations for themselves, their relationships, and their dating partners. In turn, they might be less likely to use or accept the use of violence.

The purpose of this study was to assess associations between physical dating violence victimization (PDV) and high expectation messages from school, home, and community, as well as peer norms among a representative sample of California middle and high school students. The secondary purpose was to assess whether these associations differed by gender and grade level, and whether the associations were mediated by self-efficacy. We hypothesized that students who experienced higher expectation messages across more domains and more positive peer norms would exhibit lower odds of experiencing PDV than those who did not. We also hypothesized that the protective effect of high expectations could be partially explained by self-efficacy. While past research on the association between expectations or norms and dating violence have been based on small studies, we used data from a large survey of California adolescents.

Method

Study Design and Participants

This study used the 2008-2010 CHKS, which includes 716,508 respondents from the 7th, 9th, and 11th grades. CHKS is an anonymous, school-based cross-sectional survey consisting of a core set of questions about health-related risk behavior and resilience, as well as behavior-specific supplemental modules (WestEd, 1995; Austin, Bates, & Duerr, 2011). Permission was obtained to use the database from the California Department of Education, via WestEd. These data were completely de-identified; therefore, human subject approval was not required.

This sample represents about 63% of California’s schools with enrollment for Grades 7, 9, and 11. Students who had parental or guardian consent were surveyed, and participation was voluntary, anonymous, and confidential. Districts varied regarding the use of active or passive consent procedures. CHKS data were weighted to district level enrollment within each grade, based on the latest data available from the California Basic Educational Data System. School was used as the primary sampling unit. Validity criteria were applied such that students were excluded for inconsistent patterns of responding, implausible reports of drug use, the endorsement of a fictitious drug, or failure to assent to having answered questions honestly.

For our purposes, students who were missing data (n = 57,681) or selected “does not apply” (n = 306,306) for the PDV item were excluded. Those who selected “does not apply” were not likely in a dating relationship in the previous 12 months. Items for high expectations and peer norms were administered at about 27% of schools because administering supplemental modules was optional, resulting in a final sample of 85,198 students who had complete expectation and peer norm data and representing about 17% of California schools.

Comparison of the students who were included to those who were excluded based on missing expectation or peer norm data revealed no statistically significant differences based on grade level, grades received in school, community expectations, or peer norms. However, there were differences in gender, race/ethnicity, school expectations, home expectations, self-efficacy, and PDV, with more females (p < .005) and non-Whites (p < .005) in the study sample compared with the excluded group, higher school expectations (p < .005), home expectations (p < .005), and self-efficacy (p < .005), and lower rates of PDV (p < .005) in the study sample (StataCorp, 2011).

Measures

PDV

PDV was assessed using one item: “During the past 12 months, did your boyfriend or girlfriend ever hit, slap, or physically hurt you on purpose?” Response options were “yes,” “no,” or “does not apply.”

High expectations

Nine items were used to assess high expectations (three items in each of the three domains: school, community, and home). All items were measured on a 4-point Likert-type scale with students indicating how true a statement was for them. Response options included “not at all true,” “a little true,” “pretty much true,” or “very much true.” Higher scores on the measures denote more positive endorsement of high expectations. High expectations at school were assessed using the following three items: “At my school, there is a teacher or adult who: (1) tells me when I do a good job, (2) always wants me to do my best, (3) believes that I will be a success.” These statements represent three separate items, and students could endorse all, some, or none of them. High expectations in the community were assessed with the same three items, but the following item anchor was used: “Outside of my home and school, there is an adult who . . . .” High expectations at home were measured using the following three items: “In my home, there is a parent or some other adult who: (1) expects me to follow the rules, (2) who always wants me to do my best, (3) who believes that I will be a success.” Results from Polychoric Principal Components Analysis (PCA); Kolenikov and Angeles (2004) indicated just one major underlying dimension for each domain. (Polychoric PCA is better for analyzing ordered categorical variables than is PCA, which is used for continuous variables.) Based on the polychoric PCA, an index was created for each domain. Because the simple summed scores of the items were highly correlated with the PCA scores, we used them for ease of understanding. The relationship between the log odds of PDV and these indices were not linear, so categories were created and coded as indicator variables so that no functional form assumptions were made. We created four categories of expectations within each domain: very low (average score for items < 2), low (≥2 and <3), moderate (≥3 and <4), and high (=4). We also created an Overall High Expectations score that averaged all nine items from the three domains and categorized them in the same way.

Peer norms

Three items measured peer norms, using the same 4-point Likert-type scale as described for High Expectations. Peer norms were assessed with the three items: “My friends (1) get into a lot of trouble,. (2) try to do what is right, (3) do well in school.” The first item was reverse coded, and mean scores for the three peer norm items were categorized in the same way that high expectations were categorized. Higher scores on the peer norm measure denote more positive peer behaviors. Results from Polychoric PCA indicated only one underlying dimension.

Self-efficacy

Three items were used to assess self-efficacy. All items were measured on a 4-point Likert-type scale, with the same response options as presented for High Expectations. Self-efficacy was assessed using the following three items: (1) “I can work out my problems.” (2) “I can do most things if I try.” (3) “There are many things that I do well.” These statements represent three separate items, and students could endorse all, some, or none of them. Results from Polychoric PCA indicated only one underlying dimension. Because of the functional form, we created four categories of self-efficacy: very low, low, moderate, and high.

Covariates

Student demographic characteristics included gender, grade level, and race/ethnicity. Race/ethnicity was based on the question, “How do you describe yourself? (Mark all that apply)” and was categorized as White, Latino, Black, Asian, Pacific Islander, Native American, Other, and Multiracial. Students who marked only one group were assigned to that category. Students who marked both White and one non-White race/ethnicity were assigned to that non-White category. Students who marked both Asian and Pacific Islander were assigned to the Pacific Islander category because a quarter of students who marked Pacific Islander also marked Asian. All other students who marked more than one non-White race/ethnicity were assigned to the Multiracial category. Grades earned in school was also included as a covariate, assessed using an 8-point scale (“Mostly As” to “Mostly Fs”) based on responses to the question, “During the past 12 months, how would you describe the grades you mostly received in school?”

Data Analysis

Analyses were conducted using STATA version 12.0 and survey weights. We examined high expectations in each domain, separately, as well as a composite measure of all three. Univariate and bivariate analyses were conducted to determine which covariates to include in multivariate analyses.

Logistic regression was used to calculate adjusted ORs and 95% CIs. First, we examined the unique effects for high expectations in each domain (school, home, and community) and peer norms on PDV in a single model, each adjusting for the other, to determine their unique effects. Second, we examined the associations between PDV and high expectations and peer norms, separately, and also cumulative (overall) expectations. For all multivariate models, we adjusted for gender, race/ethnicity, grade level, and self-reported grades. Third, we examined interactions for each expectation/peer norms domain by grade level and gender. Finally, we tested the mediation effect of self-efficacy using Baron and Kenny’s (1986) method. We assessed the bivariate associations between (1) exposures (high expectations and peer norms) and outcome (PDV), (2) exposures and mediator (self-efficacy), and (3) mediator and outcome. We then examined whether adding self-efficacy to each model reduced associations between exposures and PDV significantly. For these mediation analyses, the study sample was further restricted to only those students with complete self-efficacy data (n = 83,147).

Results

Bivariate Associations

Demographics are presented in Table 1. Approximately 10% of students reported experiencing PDV in the past 12 months. Males were more likely to report experiencing PDV than females (10.9% vs. 8.9%). Based on bivariate analyses, we included the following covariates in our multivariate models: gender, grade level, race/ethnicity, and grades.

Table 1.

Sociodemographics of Study Population and Association With Physical Dating Violence, California Healthy Kids Survey 2008-2010.

Characteristic	Study Population (%)	Physical Dating Violence (Past 12 Months)
All	n = 85,198	n = 8,409 (9.85%)
Grade level
7th	31.1	8.1***
9th	34.5	10.0***
11th	34.4	11.2***
Gender
Male	46.4	10.9***
Female	53.7	8.9***
Race/ethnicity
White	20.6	9.4***
Latino	46.5	9.1***
Black	6.1	11.9***
Asian	8.7	8.8***
Pacific Islander	3.0	10.4***
Native American	2.0	16.1***
Other	6.2	10.9***
Multiracial	7.0	12.5***
Grades
Mostly As	15.0	8.8***
Mostly As and Bs	30.0	7.6***
Mostly Bs	9.2	9.9***
Mostly Bs and Cs	23.0	9.5***
Mostly Cs	6.6	11.3***
Mostly Cs and Ds	10.3	11.6***
Mostly Ds	2.4	15.5***
Mostly Fs	3.5	23.1***
Self-efficacy
Very low	3.8	25.6***
Low	17.2	11.8***
Medium	42.6	8.5***
High	36.4	8.6***

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

Multivariate Results

Results from multivariable logistic regression analyses are presented in Tables 2 and 3. When the three expectation domains (home, school, community) and peer norms were run in a single model, all exhibited significant unique contributions (Table 2). The home domain exhibited the strongest association; students who reported high expectations in the home domain had 0.38 times the odds of PDV (95% CI [0.32, 0.45]) of those with very low expectations at home. Students with positive peer norms showed similarly reduced odds of PDV (OR = 0.38; 95% CI [0.32, 0.44]).

Table 2.

Odds Ratios for Physical Dating Violence From Multiple Logistic Regression Model Including All Domains, Controlling for Gender, Grade Level, Race/Ethnicity, and Grades, California Healthy Kids Survey 2008-2010.

Protective Factor	Odds Ratio	95% CI
School expectations
Very low	Ref
Low	0.74	[0.67, 0.82***]
Moderate	0.66	[0.60, 0.73***]
High	0.67	[0.60, 0.75***]
Home expectations
Very low	Ref
Low	0.59	[0.50, 0.70***]
Moderate	0.43	[0.36, 0.50***]
High	0.38	[0.32, 0.45***]
Community expectations
Very low	Ref
Low	0.87	[0.78, 0.98*]
Moderate	0.81	[0.73, 0.90***]
High	0.81	[0.73-0.91***]
Peer norms
Very low	Ref
Low	0.62	[0.55, 0.70***]
Moderate	0.47	[0.41, 0.53***]
High	0.38	[0.32, 0.44***]

Note. CI = confidence interval.

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

Table 3.

Adjusted Odds Ratios for Physical Dating Violence From Separate Logistic Regression Models for Each Expectation Domain and for Peer Norms, California Healthy Kids Survey 2008-2010.

Protective Factor^a	Odds Ratio	95% CI	Odds Ratio	95% CI
Overall expectations
Very low	Ref		Ref
Low	0.51	[0.44, 0.59***]	0.63	[0.54, 0.74***]
Moderate	0.32	[0.28, 0.36***]	0.43	[0.37, 0.50***]
High	0.24	[0.20, 0.28***]	0.33	[0.27, 0.39***]
Self-efficacy
High (vs. very low)			0.45	[0.39, 0.53***]
School expectations
Very low	Ref		Ref
Low	0.59	[0.54, 0.65***]	0.67	[0.61, 0.74***]
Moderate	0.46	[0.42, 0.51***]	0.55	[0.50, 0.62***]
High	0.43	[0.39, 0.48***]	0.53	[0.48, 0.60***]
Self-efficacy
High (vs. very low)			0.36	[0.31, 0.43***]
Home expectations
Very low	Ref		Ref
Low	0.49	[0.42, 0.58***]	0.60	[0.52, 0.70***]
Moderate	0.31	[0.27, 0.36***]	0.41	[0.36, 0.47***]
High	0.25	[0.21, 0.29***]	0.33	[0.29, 0.39***]
Self-efficacy
High (vs. very low)			0.51	[0.45, 0.58***]
Community expectations
Very low	Ref		Ref
Low	0.70	[0.63, 0.79***]	0.77	[0.68, 0.86***]
Moderate	0.55	[0.50, 0.60***]	0.64	[0.58, 0.71***]
High	0.49	[0.45, 0.55***]	0.59	[0.53, 0.66***]
Self-efficacy
High (vs. very low)			0.36	[0.31, 0.43***]
Peer norms
Very low	Ref		Ref
Low	0.52	[0.46, 0.59***]	0.58	[0.52, 0.65***]
Moderate	0.34	[0.30, 0.39***]	0.41	[0.36, 0.46***]
High	0.27	[0.23, 0.31***]	0.31	[0.27, 0.36***]
Self-efficacy
High (vs. very low)			0.40	[0.35, 0.46***]

Each expectation and norm domain was run in a separate model. All models controlled for gender, grade level, race/ethnicity, and grades.

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

Table 3 presents the results of separate models for each domain. Students with high overall expectations had 76% decreased odds (OR = 0.24; 95% CI [0.20, 0.28]) of being a victim of dating violence compared with students with very low overall expectations. Within all four individual domains, separately, higher expectations or peer norms were associated with lower odds of PDV. We also tested the relationships with continuous independent variables, and it did not make a difference in the overall test of significance.

Interaction With Grade Level and Gender

There was no evidence of interaction between grade level and expectations nor gender and expectations across the home, school, and community expectation domains. However, there were significant interactions for peer norms by grade level, as well as peer norms by gender, predicting PDV. Logistic regression results, stratified by gender and grade level, showed significant differences in the associations between peer norms and PDV depending on gender and grade level. The association between peer norms and dating violence was slightly stronger for girls than boys, particularly in 11th grade (OR = 0.28 for females with positive vs. negative peer norms, compared with OR = 0.37 for males with positive vs. negative peer norms).

Self-Efficacy as a Mediator

Mediation results suggested that self-efficacy mediated the association between expectations and dating violence. Self-efficacy was significantly associated with dating violence (Table 1) and high expectations in each domain and peer norms. High expectations and peer norms continued to be associated with dating violence when self-efficacy was included in the respective models (Table 3). However, with self-efficacy included, the strength of these associations decreased in all domains. These findings indicate that self-efficacy partially mediated the associations between high expectations and dating violence.

Discussion

Findings from this study support the hypothesis that high expectations are protective against teen PDV in a large representative sample of California middle and high school students. High expectations in all three domains, school, home, and community, and positive peer norms were associated with PDV when tested in separate models, and also when tested together in a single model. Moreover, the association between high expectations and PDV was particularly strong when looking at the overall effect of high expectations across all three domains, but high expectations in the home domain and peer norms provided the greatest unique contributions. Associations between high expectations and dating violence were similar across gender and grade levels. Although not the focus of this study, higher grades earned in school was associated with lower levels of PDV. Grades were also strongly associated with expectations across all domains.

Our findings highlight the importance of high expectation messages across many aspects of an adolescent’s life. High expectations from adults at school, at home, and in the community all contributed uniquely to the association with dating violence. This suggests that adolescents benefit from having multiple supportive adults in their lives, and that no single adult or domain determines a young person’s chances for success.

Parents Remain Important in Adolescence

Although high expectations in all domains were important, the results of this study emphasize the critical role of parents and adults in the home environment. Research suggests that developing youth begin to seek autonomy from their parents during adolescence (Friedman, 2008). Therefore, it might be surprising that expectations from adults at home had the largest influence on dating violence in this study. This finding is consistent with previous studies that have highlighted the protective effect of parental monitoring on dating violence (Foshee et al., 2001; Foshee et al., 2011; D. Howard et al., 2003). Despite their growing independence, adolescents appear affected by what their parents expect of them and how supported they feel. It could be the long-term and cumulative effect of these relationships and expectations, based on experiences that start earlier in life, that makes the home environment particularly influential.

Peer Norms

In addition, this study underscores the importance of positive peer norms during middle school and high school. While previous research has studied the associations between peer relationship norms, risk behaviors, and dating violence (Foshee et al., 2001; D. Howard et al., 2003; Reeves & Orpinas, 2012), we examined positive behavioral norms, such as peers doing well in school. Findings demonstrated that although these norms were not directly related to dating relationships, they were associated with dating violence. The strong influence of peers is not surprising given that adolescence is a developmental period characterized by increasingly more time spent with peers.

Gender

Contrary to our expectations, we did not find gender differences in the associations between high expectations and dating violence. We did, however, find that boys reported more PDV than girls, which is consistent with past studies (Foshee et al., 1996; O’keefe & Treister, 1998; Wolfe et al., 2001). Although male and female adolescents generally report similar rates of victimization and perpetration of dating violence (CDC, 2006; Foshee et al., 1996; Hickman, Jaycox, & Aronoff, 2004; Molidor & Tolman, 1998), there are differences in severity, types of violence, and reasons for using violence (Foshee et al., 1996; Foshee et al., 2011; Hickman et al., 2004; Molidor & Tolman, 1998; Noonan & Charles, 2009; O’keefe & Treister, 1998; Watson, Cascardi, Avery-Leaf, & O’Leary, 2001; Wolfe et al., 2001; Wolitzky-Taylor et al., 2008). Females report higher rates of injury caused by a dating partner and sexual victimization. Although males and females report anger as a primary reason for using violence, more females report using violence as self-defense, while more males report using violence to exert control over their partners (O’keefe & Treister, 1998; Watson et al., 2001). Our single-item measure of PDV victimization was likely unable to capture these subtleties in differences between males and females with regard to dating violence because the question was too general and did not ask about type of violence, severity, or reasons for using violence.

Self-Efficacy

This study provided support for our conceptual model that self-efficacy may partially mediate the effects of expectations (home, school, and community) on dating violence. It is plausible that adolescents who have consistently been told they can succeed will develop a stronger sense of self-efficacy, and that in turn that self-efficacy will enable them to develop healthy relationships and advocate for their own safety. This finding is consistent with previous research documenting the association between self-efficacy and dating violence (Chiodo et al., 2012; D. E. Howard et al., 2005). However, although self-efficacy shifted the ORs, this finding should be interpreted with caution given that other unmeasured variables might contribute to the relationship between expectations and dating violence.

Limitations

Our cross-sectional data limited causal inference. While we believe that high expectations, positive peer norms, and self-efficacy are protective against dating violence, it is equally possible that students who have been victims of dating violence may subsequently report or perceive lower expectations, poorer norms, and less self-efficacy. Moreover, our measure of PDV was limited to one item that focused on physical violence; other forms of dating violence were not assessed, nor the severity or chronicity of violence. There was also no measure of violence perpetration nor history of abuse or trauma. In addition, because the supplemental module with the expectation and peer norms items was optional for schools, there may be selection bias, which would limit external validity of the results. Notably, most of the differences between the students included in the study sample versus those excluded suggest that the study sample represented a more advantaged group, with overall higher levels of school expectations, home expectations, and self-efficacy, and lower rates of PDV. The highest risk students may not be captured in the study sample, and therefore, we do not know whether the association between expectations and PDV found in this study would be true for this higher risk group. Furthermore, because CHKS is administered in schools, youth who have dropped out of school or students who were absent the day the survey was administered were not captured in this study. In addition, we did not control for participant socioeconomic status because this was not assessed in CHKS but could be an important confounding factor. In addition, for schools that used active (rather than passive) parental consent procedures, students with uninvolved parents may have been less likely to be consented for the study. Finally, the study was based solely on self-reported data. However, only students who passed validity criteria were used in the analysis.

Despite limitations, this study has notable strengths, including its very large sample of middle and high school students in the California public schools. The sample size allowed for adequate power to examine differences by gender and grade level. In addition, the examination of high expectations in multiple domains represents a unique contribution to the literature.

Future Directions

Further research is needed to better understand how high expectations may foster resilience among adolescents who are faced with multiple challenges (e.g., losing a parent, learning difficulties, community violence), and thereby promote positive and healthy romantic relationships. Furthermore, future research should consider whether low expectations from adults in one domain may be partially overcome by strong communication of positive expectations from adults in other domains. Finally, while peer norms provide an estimate of peer expectations, further research looking specifically at peer expectations would be valuable. Adolescents’ perceptions of their peers’ expectations might be particularly influential and differ from peer behavioral norms, especially as they relate to PDV.

Footnotes

Acknowledgements

Nancy Bagnato and Mina White of the California Department of Public Health encouraged and supported our work through discussions.

Authors’ Note

T. Kiku Annon, research associate at WestEd, facilitated our access to the CHKS dataset. This study was presented as a poster at the National Conference of Family Medicine Residents and Medical Students on July 26-28, 2012, and at the APHA Annual Meeting on October 27-31, 2012.

Declaration of Conflicting Interests

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

Funding

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

Author Biographies

Anna Loeb completed a master’s in public health in maternal and child health at the University of California, Berkeley, in May 2012, and will be graduating from medical school at the University of California, San Francisco, in May 2013.

Julianna Deardorff, PhD, is an assistant professor of maternal and child health in the School of Public Health at the University of California, Berkeley, and an adolescent clinical psychologist. Her research focuses on adolescent sexuality, pubertal development, and promotion of positive mental health among children and adolescents.

Maureen Lahiff, PhD, is a lecturer of biostatistics in the School of Public Health at the University of California, Berkeley. Her research interests include multivariate methods, time series and longitudinal data, and acculturation and health.

References

Austin

Bates

Duerr

(2011). Guidebook to the California Healthy Kids Survey Part I: Administration (2011-2012 ed.). San Francisco, CA: WestEd. Retrieved from http://chks.wested.org/resources/chks_guidebook_1_admin.pdf

Bandura

(1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychological Review, 84, 191-215.

Baron

R. M.

Kenny

D. A.

(1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 1173-1182.

Benard

(1991). Fostering resiliency in kids: Protective factors in the family, school and community. Portland, OR: Western Center for Drug-Free Schools and Communities.

Centers for Disease Control and Prevention (2013). Teen dating violence: Intimate partner violence: Violence prevention: Injury. Retrieved January 10, 2012, from http://www.cdc.gov/ViolencePrevention/intimatepartnerviolence/teen_dating_violence.html

Centers for Disease Control and Prevention. (2006). Physical dating violence among high school students—United States, 2003. Morbidity and Mortality Weekly Report, 55(19), 532-535.

Centers for Disease Control and Prevention. (2012). Understanding teen dating violence fact sheet 2012. Retrieved from http://www.cdc.gov/violenceprevention/pdf/teendatingviolence2012-a.pdf

Centers for Disease Control and Prevention. (2012). Understanding intimate partner violence fact sheet 2012. Retrieved from http://www.cdc.gov/ViolencePrevention/pdf/IPV_factsheet-a.pdf

Chiodo

Crooks

C. V.

Wolfe

D. A.

McIsaac

Hughes

Jaffe

P. G.

(2012). Longitudinal prediction and concurrent functioning of adolescent girls demonstrating various profiles of dating violence and victimization. Prevention Science: The Official Journal of the Society for Prevention Research, 13, 350-359. doi:10.1007/s11121-011-0236-3

10.

Constantine

N. A.

Benard

Diaz

(1999, May). Measuring protective factors and resilience traits in youth: The healthy kids resilience assessment. Paper Presented at the Seventh Annual Meeting of the Society for Prevention Research, New Orleans, LA.

11.

Eaton

D. K.

Davis

K. S.

Barrios

Brener

N. D.

Noonan

R. K.

(2007). Associations of dating violence victimization with lifetime participation, co-occurrence, and early initiation of risk behaviors among U.S. high school students. Journal of Interpersonal Violence, 22, 585-602. doi:10.1177/0886260506298831

12.

Eaton

D. K.

Kann

Kinchen

Shanklin

Ross

Hawkins

. . .Chyen

(2010). Youth risk behavior surveillance—United States, 2009. Morbidity and Mortality Weekly Report: Surveillance Summaries, 59(5), 1-142.

13.

Foshee

V. A.

Linder

MacDougall

J. E.

Bangdiwala

(2001). Gender differences in the longitudinal predictors of adolescent dating violence. Preventive Medicine, 32, 128-141. doi:10.1006/pmed.2000.0793

14.

Foshee

V. A.

Linder

G. F.

Bauman

K. E.

Langwick

S. A.

Arriaga

X. B.

Heath

J. L.

. . .Bangdiwala

(1996). The safe dates project: Theoretical basis, evaluation design, and selected baseline findings. American Journal of Preventive Medicine, 12(5 Suppl.), 39-47.

15.

Foshee

V. A.

Reyes

H. L.

Ennett

S. T.

Suchindran

Mathias

J. P.

Karriker-Jaffe

K. J.

. . .Benefield

T. S.

(2011). Risk and protective factors distinguishing profiles of adolescent peer and dating violence perpetration. The Journal of Adolescent Health: Official Publication of the Society for Adolescent Medicine, 48, 344-350. doi:10.1016/j.jadohealth.2010.07.030

16.

Friedman

L. S.

(2008). Adolescents. In Feldman

M. D.

Christensen

J. F.

(Eds.), Behavioral medicine: A guide for clinical practice (3rd ed., Chapter 10, pp. 82-89). New York, NY: McGraw-Hill.

17.

Halpern

C. T.

Oslak

S. G.

Young

M. L.

Martin

S. L.

Kupper

L. L.

(2001). Partner violence among adolescents in opposite-sex romantic relationships: Findings from the national longitudinal study of adolescent health. American Journal of Public Health, 91, 1679-1685.

18.

Hickman

L. J.

Jaycox

L. H.

Aronoff

(2004). Dating violence among adolescents: Prevalence, gender distribution, and prevention program effectiveness. Trauma, Violence, & Abuse, 5, 123-142. doi:10.1177/1524838003262332

19.

Howard

Qiu

Boekeloo

(2003). Personal and social contextual correlates of adolescent dating violence. The Journal of Adolescent Health: Official Publication of the Society for Adolescent Medicine, 33, 9-17.

20.

Howard

D. E.

Beck

Kerr

M. H.

Shattuck

(2005). Psychosocial correlates of dating violence victimization among Latino youth. Adolescence, 40, 319-331.

21.

Kolenikov

Angeles

(2004). The use of discrete data in principal component analysis with applications to socio-economic indices. Unpublished manuscript.

22.

Kumpfer

K. L.

(1999). Factors and processes contributing to resilience. In Glantz

M. D.

Johnson

J. L.

(Eds.), Resilience and development (pp. 179-224). New York, NY: Kluwer Academic/Plenum Publishers.

23.

Masten

A. S.

(2001). Ordinary magic: Resilience processes in development. American Psychologist, 56, 227-238.

24.

McCaw

Golding

J. M.

Farley

Minkoff

J. R.

(2007). Domestic violence and abuse, health status, and social functioning. Women & Health, 45, 1-23. doi:10.1300/J013v45n02_01

25.

Miller

Breslau

Chung

W. J.

Green

J. G.

McLaughlin

K. A.

Kessler

R. C.

(2011). Adverse childhood experiences and risk of physical violence in adolescent dating relationships. Journal of Epidemiology & Community Health, 65, 1006-1013. doi:10.1136/jech.2009.105429

26.

Molidor

Tolman

R. M.

(1998). Gender and contextual factors in adolescent dating violence. Violence Against Women, 4, 180-194.

27.

Noonan

R. K.

Charles

(2009). Developing teen dating violence prevention strategies: Formative research with middle school youth. Violence Against Women, 15, 1087-1105. doi:10.1177/1077801209340761

28.

O’keefe

Treister

(1998). Victims of dating violence among high school students. Are the predictors different for males and females? Violence Against Women, 4, 195-223.

29.

Reeves

P. M.

Orpinas

(2012). Dating norms and dating violence among ninth graders in northeast Georgia: Reports from student surveys and focus groups. Journal of Interpersonal Violence, 27, 1677-1698. doi:10.1177/0886260511430386

30.

Roberts

T. A.

Auinger

Klein

J. D.

(2005). Intimate partner abuse and the reproductive health of sexually active female adolescents. The Journal of Adolescent Health: Official Publication of the Society for Adolescent Medicine, 36, 380-385. doi:10.1016/j.jadohealth.2004.06.005

31.

Silverman

J. G.

Raj

Clements

(2004). Dating violence and associated sexual risk and pregnancy among adolescent girls in the United States. Pediatrics, 114, e220-e225.

32.

Smith

P. H.

White

J. W.

Holland

L. J.

(2003). A longitudinal perspective on dating violence among adolescent and college-age women. American Journal of Public Health, 93, 1104-1109.

33.

StataCorp. (2011). Stata survey data reference manual. College Station, TA: Stata Press.

34.

Teitelman

A. M.

Ratcliffe

S. J.

Morales-Aleman

M. M.

Sullivan

C. M.

(2008). Sexual relationship power, intimate partner violence, and condom use among minority urban girls. Journal of Interpersonal Violence, 23, 1694-1712. doi:10.1177/0886260508314331

35.

Vives-Cases

Ruiz-Cantero

M. T.

Escriba-Aguir

Miralles

J. J.

(2011). The effect of intimate partner violence and other forms of violence against women on health. Journal of Public Health (Oxford, England), 33, 15-21. doi:10.1093/pubmed/fdq101

36.

Watson

J. M.

Cascardi

Avery-Leaf

O’Leary

K. D.

(2001). High school students’ responses to dating aggression. Violence and Victims, 16, 339-348.

37.

WestEd (1995). California healthy kids survey. Retrieved from http://www.wested.org/project/california-healthy-kids-survey-chks/

38.

WestEd Health and Human Development Program for the California Department of Education. (2010). California school district secondary school survey results fall 2009/Spring 2010: Core module A. Retrieved from http://chks.wested.org/resources/a-text_0910.pdf

39.

WestEd Health and Human Development Program for the California Department of Education. (2008-2009). Resilience and youth development module B (CHKS Report). Retrieved from http://chks.wested.org/resources/mshs-resilienceyd-1314.pdf

40.

Wolfe

D. A.

Scott

Reitzel-Jaffe

Wekerle

Grasley

Straatman

A. L.

(2001). Development and validation of the conflict in adolescent dating relationships inventory. Psychological Assessment, 13, 277-293.

41.

Wolitzky-Taylor

K. B.

Ruggiero

K. J.

Danielson

C. K.

Resnick

H. S.

Hanson

R. F.

Smith

D. W.

. . .Kilpatrick

D. G.

(2008). Prevalence and correlates of dating violence in a national sample of adolescents. Journal of the American Academy of Child & Adolescent Psychiatry, 47, 755-762. doi:10.1097/CHI.0b013e318172ef5f