DEP Paper

Ksyangbo

2.pdf

Home >Applied Sciences homework help >DEP Paper

Violence and Victims, Volume 28, Number 6, 2013

Bystander Education Training for Campus Sexual Assault Prevention:

An Initial Meta-Analysis

Jennifer Katz, PhD SUNY College at Geneseo

Jessica Moore, PhD University of Rochester Medical Center

The present meta-analysis evaluated the effectiveness of bystander education programs for preventing sexual assault in college communities. Undergraduates trained in bystander education for sexual assault were expected to report more favorable attitudes, behavioral proclivities, and actual behaviors relative to untrained controls. Data from 12 studies of college students (N 5 2,926) were used to calculate 32 effect sizes. Results suggested moderate effects of bystander education on both bystander efficacy and intentions to help others at risk. Smaller but significant effects were observed regarding self-reported bystander helping behaviors, (lower) rape-supportive attitudes, and (lower) rape proclivity, but not perpetration. These results provide initial support for the effectiveness of in-person bystander education training. Nonetheless, future longitudinal research evaluating behav- ioral outcomes and sexual assault incidence is needed.

Keywords: bystander education; sexual assault; rape; prevention

S exual assault is a common problem on college campuses in North America. Many researchers define sexual assault as involving attempted or completed sexual pen- etration in the absence of affirmative consent. Between 20% and 25% of women are

estimated to experience a sexual assault during college (Fisher, Cullen, & Turner, 2000). College men also experience sexual assaults. According to a report commissioned by the National Institutes of Justice (NIJ), 19% of women and 6% of men reported experiencing attempted or completed rape specifically since starting college (Krebs, Lindquist, Warner, Fisher, & Martin, 2007).

Institutions of higher education are challenged in their attempts to prevent campus sexual assault. Traditionally, educators have sought to reduce women’s risk for assaults by providing education and self-defense training (rape avoidance or risk reduction programs). In a narrative review of the rape avoidance literature, Ullman (2007) concluded that partic- ipating in self-defense and other rape avoidance programs inconsistently led to decreased rates of victimization. Although helpful, risk reduction programs alone are insufficient to end campus sexual assault (Gidycz et al., in press). Other educators have sought to reduce men’s rape-supportive attitudes and perpetration behaviors. Unfortunately, there have

Bystander Education Meta-Analysis 1055

been relatively few evaluations of such programs, and the available results are inconsis- tent (Lonsway et al., 2009). Even more concerning, among men identified as “high risk,” sexual assault education may actually predict increased perpetration (Stephens & George, 2009), perhaps because of a backlash effect.

Increasingly, community-based prevention efforts involving bystander education are recommended in the primary prevention of sexual violence (e.g., American College Health Association [ACHA], 2011). Bystanders are third party witnesses to the problem of sexual assault; they are neither perpetrators nor victims. Those third parties who intervene in response to risk for harm are responsive bystanders. The bystander education approach to sexual assault prevention encourages responsive bystander behaviors to “spread” respon- sibility for safety to members of the broader community.

Multiple in-person bystander education programs for sexual assault and interpersonal violence prevention have emerged over the past few decades. A partial list includes Bringing in the Bystander (Banyard, Plante, & Moynihan, 2004), InterACT (Rich, 2010), SCREAM peer education (McMahon, Postmus, Warrener, & Koenick, in press), and the Mentors in Violence Prevention (MVP) Program (Katz, Heisterkamp, & Fleming, 2011). Programs aimed specifically at either men or women include the Men’s Project (Gidycz, Orchowski, & Berkowitz, 2011), the Men’s Program, and the Women’s Program (Foubert, 2011).

Although specific components and audiences of these bystander education programs vary, they share common methods and goals. In terms of methods, participants are approached as potential allies or helpers, which may reduce defensiveness or backlash. In addition, bystander programs typically involve single-sex groups facilitated by trained peers who serve as positive role models. In terms of goals, bystander programs seek to promote prosocial attitudes and behaviors related to both sexual assault and helping others. Participants are educated about prevalence rates, indicators of high risk situations, and how they, as bystanders, can promote safety. Sexual assault awareness may engage community members by challenging the acceptability of sexual assault and helping par- ticipants to notice high risk situations and respond constructively. Such programs may also empower participants as capable of helping others and meaningfully contributing to creating an inclusive, safe community. In addition to promoting awareness and empower- ment, bystander programs teach participants how to remain safe while also promoting safety more generally. For example, in response to either actual or potential risk, program attendees are encouraged to interrupt the situation in some way to prevent a completed assault. Likewise, after hearing sexually degrading comments or victim disclosures, pro- gram attendees are encouraged to express support for victims and people as humans who deserve respect. Awareness, empowerment, and skills imparted by educational programs have the potential to help specific victims while also promoting norms for community safety more generally.

The goal of the present meta-analysis was to evaluate the existing literature on bystander education programs for campus sexual assault prevention. To what degree do these bystander programs specifically benefit participants and promote campus safety? In what ways are these programs successful, and in what ways are they not? To date, no systematic or meta-analytic evaluations of the effects of bystander education programs for sexual assault prevention exist. The effectiveness of these programs across campuses, programs, and researchers merits evaluation. Furthermore, this initial evaluation also may help identify important directions for program development and future research.

Previous meta-analyses have examined outcomes associated with general (not bystander-based) sexual assault education programs for college students. For example,

1056 Katz and Moore

Flores and Harlaub (1998) analyzed 15 effect sizes from 11 studies of education for college men; the authors concluded that various programs yield positive but short term changes in rape-supportive attitudes. Subsequently, Brecklin and Forde (2001) analyzed 153 effect sizes from 22 published and 23 unpublished studies with either pre–post or treatment-control comparisons of rape-supportive attitudes. These authors concluded that, overall, education fosters prosocial attitudes about rape. However, effect sizes were negatively associated with the length of time between the training and the posttraining assessment, and men in mixed-gender groups reported less attitude change than men in same-gender groups.

A more recent meta-analysis of sexual assault education programs provides more refined and rigorous information about how college students are affected by general sexual assault education. Anderson and Whiston (2005) separately analyzed seven different out- comes across 69 studies: (a) rape-supportive attitudes, (b) rape empathy, (c) rape-related attitudes (e.g., gender stereotypes), (d) rape knowledge, (e) behavioral intent to rape, (f) awareness behavior (e.g., interest in rape activism), and (g) sexual assault incidence. These authors employed more rigorous inclusion criteria than in past meta-analyses, including a requirement for a control condition to avoid potential biases associated with precomparisons to postcomparisons. This analysis provided strong evidence for the use of sexual assault education. On average, five of the seven outcomes (all but rape empathy and awareness behavior) showed significant effects. Several moderator variables also were identified: for example, more focused, longer programs had a greater effect on rape-supportive attitudes and rape-related attitudes and professional educators were more effective than peers.

The current meta-analysis focused specifically on bystander education programs for college students. Guided by Anderson and Whiston (2005), we restricted our analysis to effect sizes based on comparisons of treatment versus control conditions and we exam- ined multiple outcomes separately. Outcomes evaluated in the present meta-analysis were selected based on both conceptual and pragmatic considerations. Conceptually, bystander education for sexual assault attempts to teach students about bystander behavior and about sexual assault, and so both bystander-related and rape-related outcomes were evaluated. Pragmatically, we could only analyze outcomes assessed in multiple previous evaluations of bystander education. In essence, six different meta-analyses were conducted, one for each different outcome.

The first outcome was bystander efficacy, defined as one’s perceived competence in help- fully responding to sexual assault risk. For example, participants might be asked how confi- dent they are that they would be able to “Walk a friend who has had too much to drink home from a party” (Banyard, 2008, p. 90). The second outcome was rape-supportive attitudes such as acceptance of rape myths: common but generally false attitudes and beliefs that deny, min- imize, or justify men’s use of sexual aggression against women. A sample rape myth is “If a woman doesn’t physically fight back, you can’t really say that it was rape” (Payne, Lonsway, & Fitzgerald, 1999, p. 49). The third outcome was intent to help as a bystander. For example, participants might be asked how willing they would be or how likely they would be in the future to “think through the pros and cons of different ways I might help if I see an instance of sexual violence” (Banyard & Moynihan, 2011, p. 300). The fourth outcome was rape pro- clivity, assessed by asking how reinforcing sexual aggression is or by how likely one would be to rape someone if they knew they wouldn’t be caught or punished (Malamuth, 1991). The fifth outcome was actual bystander helping behaviors enacted since the training. Examples include “I stopped and checked in with my friend who looked very intoxicated when being

Bystander Education Meta-Analysis 1057

taken upstairs at a party” and “said something when I heard someone say ‘she deserved to be raped’” (Banyard & Moynihan, 2011, p. 300). The final outcome was perpetration behaviors since the training, typically assessed by asking male participants about their use of behavior- ally specific acts adapted from the Sexual Experiences Survey (Koss & Oros, 1982).

The primary aim of the present research was to estimate the degree to which bystander education programs promote favorable outcomes among trained relative to untrained participants. We expected students who attended bystander education training sessions would report more prosocial attitudes, behavioral proclivities, and actual behaviors as compared to controls. A secondary aim was to examine the magnitude of training effects on various outcomes, both bystander-related (efficacy, intent to help others, and bystander helping behavior) and rape-related (rape-supportive attitudes, rape proclivity, and perpetra- tion behavior). We expected to find stronger effects on bystander-related outcomes overall as compared to rape-related outcomes. Finally, we explored possible moderators of the effects of bystander education programs on these outcomes.

METHOD

We adopted strategies suggested by Cooper (2010) to systematically search the avail- able research literature on bystander education programs for sexual assault prevention. Initially, we conducted computerized literature searches of the following databases using multiple combinations of various search terms including “bystander” or “wit- ness,” “prevention” or “training,” and “sexual assault” or “rape”: (a) PsycInfo, (b) ERIC (Educational Resources Information Center), (c) Dissertation Abstracts Online, (d) Sociological Abstracts, and (e) MEDLINE. Descriptions of program content were exam- ined for explicit focus on approaching participants as allies and who could help others at risk for or following sexual assault. Next, we examined the reference sections of rel- evant articles to identify and locate other studies that might be appropriate for inclusion. General internet searches were conducted to identify unpublished electronically available reports on the internet; search terms included researcher’s names as well as the names of programs (e.g., MVP program, Step Up!). We also contacted study authors to request copies of unpublished or in press work not yet available online. In the case of dissertation studies, we conducted online searches to locate individuals and request copies. Finally, we searched by hand through important journals in community psychology, prevention, and interpersonal violence.

A bystander education program for sexual assault prevention was defined as involving an in-person training session in which participants are approached as allies in prevention. At least part of the education explicitly focused on ways in which participants could help reduce others’ risk for sexual assault and/or respond to others who may be or who were victims. Studies eligible for inclusion in the meta-analysis involved those in which North American college students (a) attended a bystander education program for sexual assault prevention, (b) provided quantitative data on at least one of our six study outcome vari- ables both pretraining and posttraining (i.e., qualitative studies, studies with posttraining data only, or both were excluded), and (c) whose participation in the program could be compared to control students not trained in bystander education.

Out of a possible 41 identified studies, 12 studies (one unpublished) met our inclusion criteria. Studies were excluded for one or more of the following reasons: participants were not in college, bystander education took the form of a speech or a poster, no quantitative

1058 Katz and Moore

pretraining and posttraining data were available, there was no control or comparison group, or the study analyzed data from a sample already included in the meta-analysis. Because not all studies included assessments of all outcomes, different numbers of studies were included for each potential outcome variable. In all, 32 effect sizes were evaluated. Only one study (Banyard, Moynihan, & Plante, 2007) employed a multiple training ses- sion condition. Accordingly, the present meta-analysis examines only the effects of a single training session based on pre- and post-scores. Most studies assessed outcomes at a single time point. When multiple outcome assessments were conducted, only the first outcome was coded. The following study characteristics were coded as potential modera- tor variables: year of the study, sample characteristics (average age, percent male, Greek affiliation), program presenters (professional or peer), length of the training program (in minutes), and length of time between training and outcome assessment (in days).

RESULTS

Studies included in the meta-analysis are listed in Table 1. There were 1,452 college stu- dents trained in bystander education, and 1,474 served as untrained controls. On average, study participants were 19.34 years old (SD 5 0.56, range 18.50–20.33). Across the stud- ies, 53.8% of participants were in their first year of college, and 66.7% of the participants in the studies were men. Peers conducted the trainings in 86.0% of the studies. Every study reported that bystander education was conducted in same-sex groups. Only three studies included mixed sex samples. On average, attitudes and behavioral intention outcomes were assessed 20 days after the training; behaviors were assessed at 2 or more months (M 5 112 days, SD 5 68.58, range 56–224). Only nine programs provided information about the length of the training; on average, trainings were 140 min long (SD 5 122.78, range 60–420 min). Studies were conducted from 1997 to 2011, with an average of 2006.71 (SD 5 4.84 years).

Comprehensive Meta-Analysis Version 2.0 (CMA) Software was used to calculate standardized mean differences between the posttraining education group and the control group. Within each study, separate effect sizes were calculated for each of the six different outcome variables available. Next, using CMA Software, a random effects meta-analysis was calculated for each outcome. A random effects meta-analysis is used when the studies differ in terms of participant characteristics as well as the content and implementation of the training programs. Each study outcome was weighted by an estimate of its precision; outcomes were then combined to create a summary effect size: a standardized mean dif- ference. The summary effect is an estimate of the mean of the true effects of training in bystander education, with a null hypothesis that the standardized mean difference between participants in training versus control groups is zero (Borenstein, Hedges, Higgins, & Rothstein, 2009).

Results are reported in Table 2. As can be seen, effect sizes ranged from .579 for measures of intent to help others to 2.167 for intent to perpetrate. The null hypothesis of a true effect of zero was rejected for five of the six outcome variables. Compared to controls, students who attended bystander education programs reported more bystander efficacy, intent to help others, and actual bystander behaviors, and less rape myth accep- tance and rape proclivity. In contrast, trained students were no less likely to report per- petration behaviors.

Bystander Education Meta-Analysis 1059

Orwin’s fail safe N statistic for each significant outcome is reported in Table 2. This statistic is an index of publication bias; it shows the number of additional studies with trivial effects needed to reduce the effect size enough to accept the null hypothesis of no significant training effect (Borenstein et al., 2009). The criterion for a “trivial” standard difference in means was set at .05. As shown in Table 2, 44, 50, and 54 studies would be needed to nullify the significant effects of bystander education programs on participants’ bystander efficacy, rape- supportive attitudes, or intent to help as a bystander, respectively. Publication bias may be more likely for the other study outcomes; 11 studies with trivial effects would nullify the observed significant education effects on bystander helping behavior, and only 10 studies with trivial effects would nullify observed effects on reduced rape proclivity.

Based on Cohen’s reference values, a standardized mean difference of about .50 is con- sidered a moderate effect size, and a standardized mean difference of about .20 is small (Cooper, 2010). Using these guidelines, moderate effect sizes were obtained regarding measures of bystander efficacy (.49, 95% CI 5 .31 to .66) and intent to help others as a bystander (.58, 95% CI 5 .38 to .78). In contrast, smaller effect sizes were obtained for rape myth acceptance (2.28, 95% CI 5 2.20 to 2.36), rape proclivity (2.17, 95% CI 5 2.03 to 2.30), and for actual bystander behavior (.23, 95% CI 5 .04 to .41). Overall, the available evidence suggests that bystander education programs may be more effective in promoting bystander efficacy and intent to help others than in promoting actual bystander helping behavior. Furthermore, bystander programs exert moderate effects on bystander- related attitudes and behavioral intentions (bystander efficacy, intent to help) but small effects on rape-related attitudes and behavioral intentions (rape-supportive attitudes, rape proclivity).

Tests of homogeneity of the summary effects are also reported in Table 2. The Q sta- tistic addresses the null hypothesis; the true dispersion of effects is zero. As can be seen, none of the Q tests reached statistical significance, although a p 5 .05 trend emerged for bystander efficacy. Unlike Q, Tau2 and I2 are not sensitive to the number of avail- able studies. Tau2 reflects an absolute amount of true (rather than random) heterogeneity across studies, whereas I2 reflects the proportion of true, potentially explainable variance (Borenstein et al., 2009). Only bystander efficacy, intent to help as a bystander, and actual perpetration showed both (a) nonzero values for Tau2 and (b) values of I2 indicating that variability might be explained by different characteristics of the studies, participants, or interventions. However, there were only few studies for each outcome, particularly for per- petration. As such, exploratory moderator analyses were conducted for bystander efficacy and intent to help as a bystander, although these analyses must be viewed as preliminary given the small number of studies available.

Separate subgroup and meta-regression analyses were conducted to identify predictors of posttraining bystander efficacy and intent to help. Meta-regression results suggested that the average age of participants was negatively associated with program effects on bystander efficacy. More specifically, bystander programs offered to younger participants showed stronger effect sizes (B 5 2.392, SE 5 .146, Z 5 22.69, p 5 .007), although the total model only approached significance, Q(5) 5 11.03, p 5 .051. Meta-regression results also showed that the percentage of male participants was positively related to intent to help others (B 5 .005, SE 5 .002, Z 5 2.35, p 5 .018), although again, the total model did not reach statistical significance, Q(4) 5 6.72, p 5 .15. No other moderator variables were sig- nificantly associated with effect sizes for either bystander efficacy or intent to help others.

1060 Katz and Moore

T A

B L

E 1

. S

tu d

ie s

In cl

u d

ed i

n M

et a-

A n

al ys

A ut

ho rs

/Y ea

r S

am pl

e P

ro gr

am P

re se

nt er

s T

im in

g of

O ut

co m

e M

et a-

A na

ly ti

c O

ut co

m es

M oy

ni ha

n, B

an ya

rd ,

A rn

ol d,

E ck

st ei

n, &

ta pl

et on

( 20

11 )

N 5

5 6

so ro

ri ty

w om

en M

ea n

ag e:

1 9.

0 ye

ar s

P er

ce nt

m al

e: 0

B ri

n g in

g i

n t

h e

B ys

ta n d er

0 m

in )

P ro

fe ss

io na

l st

af f

5– 6

w ee

ks p

os tt

ra in

in g

B ys

ta nd

er e

ff ic

ac y

B ys

ta nd

er i

nt en

M oy

ni ha

n, B

an ya

rd ,

A rn

ol d,

E ck

st ei

n, &

ta pl

et on

( 20

10 )

N 5

9 8

at hl

et es

M ea

n ag

e: 1

9. 4

ye ar

s P

er ce

nt m

al e:

5 7

B ri

n g in

g i

n t

h e

B ys

ta n d er

h r)

P ro

fe ss

io na

l st

af f

1 w

ee k

po st

tr ai

ni ng

B ys

ta nd

er e

ff ic

ac y

R ap

e at

ti tu

de s

B ys

ta nd

er i

nt en

t B

ys ta

nd er

b eh

av io

B an

ya rd

, M

oy ni

ha n,

P la

nt e

(2 00

7) N

5 2

37 M

ea n

ag e:

1 9.

4 ye

ar s

P er

ce nt

m al

e: 4

B ri

n g in

g i

n t

h e

B ys

ta n d er

0 m

in )

— A

tt it

ud es

/i nt

en ti

on s:

m ed

ia te

b eh

av io

rs :

2 m

on th

B ys

ta nd

er e

ff ic

ac y

R ap

e at

ti tu

de s

B ys

ta nd

er i

nt en

t B

ys ta

nd er

b eh

av io

F ou

be rt

, L

an gh

in ri

ch se

n- R

oh li

ng ,

B ra

sf ie

ld ,

H il

l (2

01 0)

N 5

2 79

f ir

st y

ea r

w om

en M

ea n

ag e:

1 8.

9 ye

ar s

P er

ce nt

m al

e: 0

T h e

W o m

en ’s

ro g ra

m P

ee rs

Im m

ed ia

te B

ys ta

nd er

e ff

ic ac

y R

ap e

at ti

tu de

s B

ys ta

nd er

i nt

en t

G id

yc z,

O rc

ho w

sk i,

B er

ko w

it z

(2 01

1) N

5 4

53 f

ir st

y ea

r m

en M

ea n

ag e:

1 8.

5 ye

ar s

P er

ce nt

m al

e: 1

T h e

M en

’s

P ro

je ct

0 m

in )

P ee

rs 4

m on

th s

po st

tr ai

ni ng

R ap

e at

ti tu

de s

R ap

e pr

oc li

vi ty

B ys

ta nd

er b

eh av

io r

P er

pe tr

at io

L an

gh in

ri ch

se n-

R oh

li ng

, F

ou be

rt ,

B ra

sf ie

ld ,

H il

l, &

he ll

ey -T

re m

bl ay

( 20

11 )

N 5

1 77

f ir

st y

ea r

m en

M ea

n ag

e: 1

8. 9

ye ar

s P

er ce

nt m

al e:

1 00

T h e

M en

’s

P ro

g ra

m P

ee rs

Im m

ed ia

te B

ys ta

nd er

e ff

ic ac

y R

ap e

at ti

tu de

s B

ys ta

nd er

i nt

en t

Bystander Education Meta-Analysis 1061

F ou

be rt

& M

ar ri

ot t

(1 99

7) N

5 7

7 fr

at er

ni ty

p le

dg es

M ea

n ag

e: 1

8. 8

ye ar

s P

er ce

nt m

al e:

1 00

T h e

M en

’s

P ro

g ra

m P

ee rs

Im m

ed ia

te R

ap e

at ti

tu de

s R

ap e

pr oc

li vi

F ou

be rt

& M

cE w

an (

19 98

) N

5 1

05 f

ra te

rn it

y m

en M

ea n

ag e:

1 9.

9 ye

ar s

P er

ce nt

m al

e: 1

T h e

M en

’s

P ro

g ra

(6 0

m in

)

P ee

rs Im

m ed

ia te

R ap

e at

ti tu

de s

R ap

e pr

oc li

vi ty

F ou

be rt

( 20

00 )

N 5

8 8

fr at

er ni

ty m

en M

ea n

ag e:

2 0.

3 ye

ar s

P er

ce nt

m al

e: 1

T h e

M en

’s

P ro

g ra

m P

ee rs

A tt

it ud

es &

p ro

cl iv

it y:

m ed

ia te

b eh

av io

2 m

on th

R ap

e at

ti tu

de s

R ap

e pr

oc li

vi ty

P er

pe tr

at io

F ou

be rt

& N

ew be

rr y

(2 00

6) N

5 1

62 f

ra te

rn it

y m

en M

ea n

ag e:

— P

er ce

nt m

al e:

1 00

T h e

M en

’s

P ro

g ra

m P

ee rs

Im m

ed ia

te R

ap e

at ti

tu de

s R

ap e

pr oc

li vi

F ou

be rt

, N

ew be

rr y,

T at

um (

20 07

) N

5 4

76 f

ir st

y ea

r m

en M

ea n

ag e:

— P

er ce

nt m

al e:

1 00

T h e

M en

’s

P ro

g ra

m P

ee rs

A tt

it ud

es :

im m

ed ia

be ha

vi or

: 7

m on

th s

R ap

e at

ti tu

de s

P er

pe tr

at io

C is

sn er

( 20

09 )

N 5

7 16

M ea

n ag

e: 1

9. 7

ye ar

s P

er ce

nt m

al e:

4 7.

M en

to rs

i n

V io

le n ce

ro g ra

m (

7 hr

)

P ee

rs Im

m ed

ia te

B ys

ta nd

er e

ff ic

ac y

R ap

e at

ti tu

de s

N o te

. S

am pl

e si

ze s

ar e

ba se

d on

t he

n um

be r

of p

ar ti

ci pa

nt s

w ho

p ro

vi de

d fu

ll p

re te

st a

nd p

os tt

es t

da ta

i n

bo th

s in

gl e

se ss

io n

tr ai

ni ng

a nd

nt ro

l co

nd it

io ns

1062 Katz and Moore

T A

B L

E 2

. S

am p

le S

iz e

(k ),

M ea

n W

ei gh

te d

E ff

ec t

S iz

e (d

), 9

5% C

on fi

d en

ce I

n te

rv al

, O

rw in

’s F

ai l

S af

e N

, an

d H

om og

en ei

ty T

es ts

r S

ix O

u tc

om es

O ut

co m

e C

at eg

or y

k d

95 %

C I

F ai

l S

af e

N H

om og

en ei

ty T

es t

T au

2 I2

A tt

it ud

ys ta

nd er

e ff

ic ac

y 6

.4 86

** *

.3 10

/. 66

1 44

Q (5

) 5

1 1.

03 ,

p 5

. 05

.0 24

54 .6

ap e-

su pp

or ti

ve a

tt it

ud es

11 2

.2 76

** *

2 .1

96 /2

.3 57

50 Q

(1 0)

5 9

.4 9,

p 5

. 45

.0 00

0 .0

B eh

av io

ra l

in te

nt io

te nt

t o

he lp

5 .5

79 **

* .3

83 /.

77 6

54 Q

(4 )

5 6

.7 3,

p 5

. 15

.0 20

40 .5

ap e

pr oc

li vi

ty 4

2 .1

67 *

2 .0

30 /2

.3 05

10 Q

(3 )

5 1

.9 6,

p 5

. 58

.0 00

0 .0

A ct

ua l

be ha

vi or

ys ta

nd er

h el

pi ng

b eh

av io

r 3

.2 27

* .0

40 /.

41 3

11 Q

(2 )

5 2

.8 0,

p 5

. 25

.0 08

28 .5

er pe

tr at

io n

3 2

.2 82

.0 85

/2 .6

49 —

Q (2

) 5

5 .2

1, p

5 .

07 .0

63 61

.3 4

N o te

. k

5 n

um be

r of

e ff

ec t

si ze

s; d

5 m

ea n

st an

da rd

iz ed

d if

fe re

nc e

ac ro

ss s

tu di

es ;

Q , T

au 2 a

nd I

2 5

t es

ts o

f ho

m og

en ei

ty a

nd h

et er

og en

ei ty

. *p

, .

05 .

** p ,

. 01

. **

*p ,

. 00

Bystander Education Meta-Analysis 1063

DISCUSSION

The primary aim of this meta-analysis was to examine the effectiveness of in-person bystander education for campus sexual assault prevention. We expected that students who attended bystander education programs would report more prosocial attitudes, behavioral proclivities, and behaviors compared to untrained controls. Six separate quantitative outcomes were evalu- ated, and favorable results were observed for five of these outcomes. Across studies, students trained in bystander education reported increased bystander efficacy, intent to help others, and actual bystander behavior. Trained students also reported less rape myth acceptance and rape proclivity. Unfortunately, there was no evidence for an effect on perpetration behaviors.

The effects of bystander education programs varied across different types of outcomes. Moderate-sized effects were observed for measures of bystander efficacy and intent to help others. In contrast, smaller effects were found for actual bystander behavior as well as (lower) rape myth acceptance and rape proclivity. The observed patterns of effect size strength suggest bystander education programs may primarily impact outcomes associated with bystander behavior and secondarily impact outcomes associated with rape/sexual assault. Possibly, bystander-related attitudes, behavioral proclivities, and behaviors are more amenable to change than rape-related attitudes, proclivities, and behaviors. That is, many students may feel empowered to help others and willing to enact a helping role after attending a single training program. In contrast, rape-related attitudes and proclivities may be more entrenched and thus less malleable. In the one study that examined rape- supportive attitudes after one versus three sessions of bystander education, students who attended three sessions reported significantly more favorable outcomes (Banyard et al., 2007). Multiple sessions may be needed to change rape-related outcomes.

Other factors also might help explain differences in effect sizes across bystander-related versus rape-related outcomes. Bystander education programs for sexual assault preven- tion offer a positive, empowering solution to a difficult, controversial problem. Part of the training, namely the solution-oriented part, may be more engaging and less challenging for students than the parts about sexual assault. Possibly, program content focused on positive solutions (i.e., bystander behavior) is more appealing than content focused on a negative problem (i.e., sexual assault). Likewise, learning about bystander behavior entails a focus on behavioral involvement (e.g., get consent, support victims), whereas learning about sexual assault often entails a focus on prohibitions (e.g., don’t presume consent, don’t blame the victim). Although essential to sexual assault prevention efforts, educational con- tent focused on sexual assault could be less engaging and more challenging than content focused on positive solutions. Nonetheless, talking about sexual assault during bystander education programs decreases the silence surrounding sexual assault, moving a typically private problem to the public sphere, and challenging social tendencies to minimize and tolerate sexual assaults as “just sex” (Gavey, 2005).

Exploratory analyses investigated possible moderators of the effects of bystander edu- cation programs on bystander efficacy and intent to help. Results suggested a negative relationship between average participant age and bystander efficacy; studies with younger participants showed larger effects. This finding indicates that younger college students might be particularly empowered by attending bystander training programs. Although preliminary, this finding suggests that it may be particularly advantageous to implement bystander education programs for sexual assault with younger students, such as during college orientation or early in the first year. Younger students, who also tend to be newer to college, are entering emerging adulthood, a developmental stage known for exploring

1064 Katz and Moore

new behaviors and identities (Arnett, 2000). Perhaps the initial shift to emerging adult- hood fosters particular responsiveness to messages about one’s ability to help others and contribute to a safer college environment.

Moderator analyses also showed a positive relationship between participant gender and intent to help others; studies with a larger percentage of male participants promoted stronger effects on intent to help. Although this finding is preliminary and warrants rep- lication, it is encouraging. As bystanders, men may face unique social pressures related to prescriptive masculine norms; for example, men may fear that if they help others at risk for sexual assault, they will be ridiculed as “weak” (Carlson, 2008). Such fears may help explain why, in untrained samples of college students, men overall are less likely than women to say they are willing to engage in bystander helping behaviors (Banyard, 2008; Banyard & Moynihan, 2011). Bystander education programs provide college men with an opportunity for camaraderie with prosocial, “like-minded men” (McMahon & Dick, 2011). Such interactions also may help college men develop more accurate, proso- cial expectations of their male peers. College men commonly underestimate their peers’ willingness to intervene to prevent sexual assault, and this (mis)perception of other men’s willingness to intervene decreases men’s own willingness to intervene (Fabiano, Perkins, Berkowitz, Linkenbach, & Stark, 2003). Bystander training programs facilitated by or attended by men can demonstrate that masculinity is not incompatible with helping others.

In several respects, the current meta-analytic results based on in-person bystander edu- cation are consistent with previous meta-analyses regarding the effects of general sexual assault education. Previous studies concluded that general programs have a significant, posi- tive effect on reducing college students’ rape-supportive attitudes (Brecklin & Forde, 2001; Flores & Hartlaub, 1998), and this conclusion was upheld even in a rigorous meta-analysis of only controlled studies (Anderson & Whiston, 2005). Similarly, the present results showed a small but significant effect of bystander education programs specifically on rape-supportive attitudes. Anderson and Whiston (2005) also examined other outcomes, including behavioral intent to rape, and found a significant effect of general sexual assault education. Again, the present results showed a small but significant effect of bystander education programs on decreased rape proclivity. These bystander program effects on rape proclivity, however, must be viewed as preliminary given that few bystander education studies assessed rape proclivity as an outcome. Convergence with the general literature is encouraging.

Results of the present meta-analysis on bystander education programs also are consis- tent with previous meta-analytic results suggesting stronger effects on both attitudes and behavioral intentions than actual behaviors. Anderson and Whiston (2005) found no effect of general sexual assault education on sexual assault incidence. Although incidence was not assessed in any of the bystander education studies, the present meta-analysis found no effect of bystander education on reduced perpetration behaviors. Furthermore, the present analysis found moderate effect sizes for bystander efficacy and intent to help others, but small effect sizes for bystander behavior. This study provides further evidence that educational programs tend to have a stronger impact on attitudes and behavioral intentions than actual behaviors.

Limitations of this study should be noted. First, relatively few studies met our inclu- sion criteria. Systematic research on bystander education is relatively new. In addition, numerous bystander education programs have been implemented but not evaluated using quantitative methods, in controlled studies, or both. Peer education theater, for example, represents a promising approach to engaging students in developing behavioral skills nec- essary to intervene in high risk situations (McMahon et al., in press; Rich, 2010), and con- trolled studies are needed. Although relatively few studies were included in this analysis, the number of studies is comparable to an early meta-analysis of general sexual education

Bystander Education Meta-Analysis 1065

programs (Flores & Hartlaub, 1998). Furthermore, our focus on quantitative, controlled studies decreases potential biases associated with precomparisons versus postcomparisons, providing more rigorous estimates of the effectiveness of bystander education programs.

As the field develops, future meta-analytic investigations will be needed. Such work may address another limitation of the present work: limited power to detect moderator variables. Although some evidence for moderating effects on bystander-related outcomes was obtained, these effects should be viewed as preliminary. No moderating effects on rape- related outcomes (i.e., rape-supportive attitudes, rape proclivity) were observed, which was unexpected given previous meta-analyses regarding moderator effects related to the timing of the outcome assessment (Brecklin & Forde, 2001), type of presenter, or type of audience (Anderson & Whiston, 2005). The current analysis had limited statistical power to detect such differences, if such differences exist. It is also possible that moderators of bystander education programs differ from moderators of general sexual assault education programs. Before these questions can be answered, additional studies are needed in which researchers vary the timing of outcome, peer versus professional presenters or types of audiences, and assess bystander-related and rape-related outcomes following bystander education.

Additional areas for future evaluation include the systematic evaluation of multiple session training programs (e.g., Banyard et al., 2007), booster sessions (e.g., Gidycz et al., 2011), and longitudinal outcomes, including behaviors over time. In the current analy- sis, only 25% of the studies included assessments of actual bystander helping behavior. Although we found that bystander education programs promote actual bystander behavior, these results are based on only a few studies. Moreover, to date, no research has addressed whether and to what degree bystander helping behaviors lead to fewer incidents of sexual assault. This is an important question for research: how effective is bystander helping behavior in actual prevention? No studies assessed incidence of rape or sexual assault, and we found no effect of bystander education on perpetration. Without assessments of sexual assault incidence following bystander education, it remains unclear to what degree bystander education truly prevents sexual assaults.

In conclusion, initial meta-analytic results suggest that bystander education programs are effective, particularly in promoting bystander efficacy and intent to help others at risk. Bystander education programs also reduce rape-supportive attitudes and rape proclivity, although to a lesser degree. It is unclear whether and to what degree bystander education decreases the incidence of sexual assault. Future research is needed to compare the effects of educational programs focusing primarily on bystanders versus the effects of combin- ing bystander education with other types of education (e.g., risk reduction). Research on other educational approaches, such as bystander-themed posters (e.g., Potter, Moynihan, & Stapleton, 2011), also is needed. Although the present results are encouraging, researchers and educators must continue to develop empirically supported approaches to decrease, and ultimately eliminate, campus sexual assault.

REFERENCES

References marked with an asterisk indicate studies included in the meta-analysis.

American College Health Association. (2011, December). Position statement on preventing sexual violence on college and university campuses. Hanover, MD: Author. Retrieved from http://www .acha.org/Publications/docs/ACHA_Statement_Preventing_Sexual_Violence_Dec2011.pdf

Anderson, L. A., & Whiston, S. C. (2005). Sexual assault education programs: A meta-analytic examination of their effectiveness. Psychology of Women Quarterly, 29, 374–388. http://dx.doi .org/10.1111/j.1471-6402.2005.00237.x

1066 Katz and Moore

Arnett, J. J. (2000). Emerging adulthood: A theory of development from the late teens through the twenties. American Psychologist, 55, 469–480. http://dx.doi.org/10.1037/0003-066x.55.5.469

Banyard, V. L. (2008). Measurement and correlates of prosocial bystander behavior: The case of interpersonal violence. Violence and Victims, 23, 83–97. http://dx.doi.org/10.1891/0886- 6708.23.1.83

Banyard, V. L., & Moynihan, M. M. (2011). Variation in bystander behavior related to sexual and intimate partner violence prevention: Correlates in a sample of college students. Psychology of Violence, 1, 287–301. http://dx.doi.org/10.1037/a0023544

*Banyard, V. L., Moynihan, M. M., & Plante, E. G. (2007). Sexual violence prevention through bystander education: An experimental evaluation. Journal of Community Psychology, 35, 463–481. http://dx.doi.org/10.1002/jcop.20159

Banyard, V. L., Plante, E. G., & Moynihan, M. M. (2004). Bystander education: Bringing a broader community perspective to sexual violence prevention. Journal of Community Psychology, 32, 61–79. http://dx.doi.org/10.1002/jcop.10078

Borenstein, M., Hedges, L. V., Higgins, J. P. T., & Rothstein, H. R. (2009). Introduction to meta- analysis. West Sussez, United Kingdom: Wiley.

Brecklin, L. R., & Forde, D. R. (2001). A meta-analysis of rape education programs. Violence and Victims, 16, 303–321.

Carlson, M. (2008). I’d rather go along and be considered a man: Masculinity and bystander inter- vention. The Journal of Men’s Studies, 16, 3–17. http://dx.doi.org/10.3149/jms.1601.3

*Cissner, A. B. (2009). Evaluating the Mentors in Violence prevention program: Preventing gender violence on a college campus (Q184H060084). New York, NY: Center for Court Innovation.

Cooper, H. M. (2010). Research synthesis and meta-analysis: A step by step approach. Los Angeles, CA: Sage.

Fabiano, P. M., Perkins, W., Berkowitz, A., Linkenbach, J., & Stark, C. (2003). Engaging men as social justice allies in ending violence against women: Evidence for a social norms approach. Journal of American College Health, 52, 105–112. http://dx.doi.org/10.1080/07448480309595732

Fisher, B. S., Cullen, F. T., & Turner, M. (2000). The sexual victimization of college women. Washington, DC: U.S. Department of Justice.

Flores, S. A., & Hartlaub, M. G. (1998). Reducing rape-myth acceptance in male college stu- dents: A meta-analysis of intervention studies. Journal of College Student Development, 39, 438–448.

*Foubert, J. D. (2000). The longitudinal effects of a rape-prevention program on fraternity men’s attitudes, behavioral intent, and behavior. Journal of American College Health, 48, 158–163. http://dx.doi.org/10.1080/07448480009595691

Foubert, J. D. (2011). The Men’s and Women’s Programs: Ending rape through peer education. New York, NY: Routledge.

*Foubert, J. D., Langhinrichsen-Rohling, J., Brasfield, H., & Hill, B. (2010). Effects of a rape aware- ness program on college women: Increasing bystander efficacy and willingness to intervene. Journal of Community Psychology, 38, 813–827. http://dx.doi.org/10.1002/jcop.20397

*Foubert, J. D., & Marriott, K. A. (1997). Effects of a sexual assault peer education program on men’s belief in rape myths. Sex Roles, 36, 259–268. http://dx.doi.org/10.1007/bf02766271

*Foubert, J. D., & McEwen, M. K. (1998). An all-male rape prevention peer education program: Decreasing fraternity men’s behavioral intent to rape. Journal of College Student Development, 39, 548–556.

*Foubert, J. D., & Newberry, J. T. (2006). Effects of two versions of an empathy-based rape preven- tion program on fraternity men’s survivor empathy, attitudes, and behavioral intent to commit rape or sexual assault. Journal of College Student Development, 47, 133–148. http://dx.doi .org/10.1353/csd.2006.0016

*Foubert, J. D., Newberry, J. T., & Tatum, J. L. (2007). Behavior differences seven months later: Effects of a rape prevention program. NASPA Journal, 44, 728–749.

Gavey, N. (2005). Just sex? The cultural scaffolding of rape. New York, NY: Routledge.

Bystander Education Meta-Analysis 1067

*Gidycz, C. A., Orchowski, L. M., & Berkowitz, A. D. (2011). Preventing sexual aggression among college men: An evaluation of a social norms and bystander intervention program. Violence Against Women, 17, 720–742. http://dx.doi.org/10.1177/1077801211409727

Gidycz, C. A., Orchowski, L. M., Probst, D., Edwards, K., Murphy, M., & Tansill, E. (in press). Concurrent administration of sexual assault prevention and risk reduction programming: Outcomes for women. Violence Against Women.

Katz, J., Heisterkamp, H. A., & Fleming, W. M. (2011). The social justice roots of the mentors in violence prevention model and its application in a high school setting. Violence Against Women, 17, 684–702. http://dx.doi.org/10.1177/1077801211409725

Koss, M. P., & Oros, C. J. (1982). Sexual Experiences Survey: A research instrument investigat- ing sexual aggression and victimization. Journal of Consulting and Clinical Psychology, 50, 455–457. http://dx.doi.org/10.1037/0022-006x.50.3.455

Krebs, C. P., Lindquist, C. H., Warner, T. D., Fisher, B. S., & Martin, S. L. (2007). The campus sexual assault (CSA) study. (NCJ 221153). Washington, DC: National Institute of Justice.

*Langhinrichsen-Rohling, J., Foubert, J. D., Brasfield, H. M., Hill, B., & Shelley-Tremblay, S. (2011). The Men’s Program: Does it impact college men’s self-reported bystander effi- cacy and willingness to intervene? Violence Against Women, 17, 743–759. http://dx.doi .org/10.1177/1077801211409728

Lonsway, K. A., Banyard, V. L., Berkowitz, A., Gidycz, C. A., Katz, J., Koss, M. P., Schewe, P. A., Ullman, S. (2009). Rape prevention and risk reduction: A review of the research literature for practitioners. VAW Net, 1–20.

Malamuth, N. M. (1991). Rape proclivity among males. Journal of Social Issues, 37, 138–157. McMahon, S., & Dick, A. (2011). “Being in a room with like-minded men”: An exploratory study

of men’s participation in a bystander intervention program to prevent intimate partner violence. The Journal of Men’s Studies, 19, 3–18. http://dx.doi.org/10.3149/jms.1901.3

McMahon, S., Postmus, J. L., Warrener, C., & Koenick, R. A. (in press). Utilizing peer education theater for the primary prevention of sexual violence on college campuses. Journal of College Student Development.

*Moynihan, M. M., Banyard, V. L., Arnold, J. S., Eckstein, R. P., & Stapleton, J. G. (2010). Engaging intercollegiate athletes in preventing and intervening in sexual and intimate partner violence. Journal of American College Health, 59, 197–204. http://dx.doi.org/10.1080/07448481.2010 .502195

*Moynihan, M. M., Banyard, V. L., Arnold, J. S., Eckstein, R. P., & Stapleton, J. G. (2011). Sisterhood may be powerful for reducing sexual and intimate partner violence: An evaluation of the Bringing in the Bystander in-person program with sorority members. Violence Against Women, 17, 703–719. http://dx.doi.org/10.1177/1077801211409726

Payne, D. L., Lonsway, K. A., & Fitzgerald, L. F. (1999). Rape myth acceptance: Exploration of its structure and its measurement using the Illinois Rape Myth Acceptance Scale. Journal of Research in Personality, 33, 27–68. http://dx.doi.org/10.1006/jrpe.1998.2238

Potter, S. J., Moynihan, M. M., & Stapleton, J. G. (2011). Using social self-identification in social marketing materials aimed at reducing violence against women on campus. Journal of Interpersonal Violence, 26, 971–990.

Rich, M. D. (2010). The interACT model: Considering rape prevention from a performance activism and social justice perspective. Feminism & Psychology, 20, 511–528. http://dx.doi .org/10.1177/0959353510371366

Stephens, K. A., & George, W. H. (2009). Rape prevention with college men: Evaluating risk status. Journal of Interpersonal Violence, 24, 996–1013. http://dx.doi.org/10.1177/088626050831936

Ullman, S. E. (2007). A 10-year update of “review and critique of empirical studies of rape avoidance.” Criminal Justice and Behavior, 34, 411–429. http://dx.doi.org/10.1177/0093854806297117

Correspondence regarding this article should be directed to Jennifer Katz, PhD, SUNY College at Geneseo, Department of Psychology, Geneseo, NY 14454. E-mail: [email protected]

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.