Understanding Differences in Sexting Behaviors Across Gender,Relationship Status,and Sexual Identity,and the Role of Expectancies in Sexting

Abstract

Sexting, or the exchange of sexually explicit material via Internet social-networking site or mobile phone, is an increasingly prevalent behavior. The study sought to (1) identify expectancies regarding sexting behaviors, (2) examine how demographics (i.e., gender, sexual identity, relationship status) might be differentially related to sexting expectancies and behaviors, and (3) examine whether these concurrent relationships are consistent with a theoretical causal model in which sexting expectancies influence sexting behaviors. The sample consisted of 278 undergraduate students (mean age=21.0 years, SD=4.56; 53.8% female; 76.3% caucasian). Factor analyses supported the validity and reliability of the Sextpectancies Measure (α=0.85–0.93 across subscales) and indicated two expectancy domains each for both sending and receiving sexts: positive expectancies (sexual-related and affect-related) and negative expectancies. Males reported stronger positive expectancies (F=4.64, p=0.03) while females reported stronger negative expectancies (F=6.11, p=0.01) about receiving sexts. There were also differences across relationship status regarding negative expectancies (F=2.25, p=0.05 for sending; F=4.24, p=0.002 for receiving). There were also significant effects of positive (F=45.98, p<0.001 for sending, F=22.42, p<0.001 for receiving) and negative expectancies (F=36.65, p=0.02 sending, F=14.41, p<0.001 receiving) on sexting behaviors (η² from 0.04–0.13). College students reported both positive and negative sextpectancies, although sextpectancies and sexting varied significantly across gender, race, sexual identity, and relationship status. Concurrent relationships were consistent with the causal model of sextpectancies influencing sexting behaviors, and this study serves as the first test of this model, which could inform future prevention strategies to mitigate sexting risks.

Introduction

Sexting, defined as the exchange of sexually charged material (picture or text) via mobile phone or social-networking site, is one of many behaviors that demonstrates the importance of digital media in social communication and everyday life.^1,2 Much of our understanding about sexting comes from media and anecdotal reports about the dangers of sexting. Negative beliefs about sexting include the potential for legal ramifications;^3–7 the risk of sexts being shared with others;^7,8 and the resulting cyberbullying,³ social humiliation, and psychological distress.^7,8

Despite these risks, sexting is a prevalent behavior, suggesting that there are potentially a number of positive outcomes from sexting.^9,10 Although estimates of sexting vary across studies,⁹ it is estimated that between 13%¹¹ to 44% (ages 18–24)^12,13 of young adults (ages 18–26) and anywhere from 2.5% (ages 10–17)⁵ to 20% of adolescents (ages 13–19)¹¹ have sent or received nude or seminude photos of themselves. Evidence suggests some expectations of positive sexting outcomes, such as increasing the chances of “hooking up”¹¹ (i.e., unplanned coital or noncoital sexual activity between uncommitted individuals)¹⁴ or initiating sexual activity;^12,15,16 however, the field has yet to fully understand expected sexting outcomes, and the influence of these expectations on sexting activity. Endorsement of expected outcomes could differ by gender, relationship status, and sexual identity, given evidence that (1) males and females have different societal expectations concerning sexual-related behaviors,¹⁷ (2) females may be more pressured into sexting by male peers,^5,16 and (3) sexting for those with differing relationship statuses^18,19 or sexual identities²⁰ could have different outcomes. More needs to be understood about the reasons for sexting and how these reasons influence participation in sexting behaviors for different individuals.⁹

Utilizing an expectancy theory perspective, we hypothesized that expectancies about sexting outcomes influence the likelihood of engaging in sexting behaviors. Expectancies refer to individuals' beliefs or perceptions of what the outcome of a certain behavior might be,²¹ and have been used to understand attitudes about sexual-related Internet use, among other behaviors.²² Positive expectancies are important and consistent predictors of engagement in specific risk behaviors.^23–26 For example, alcohol expectancies predict later alcohol use,²⁷ gambling expectancies predict later gambling behaviors,²⁸ and eating expectancies predict binge-eating behaviors,²⁹ although many other factors are involved in these risk processes.²⁵ One recent study suggested three possible domains of attitudes toward sexting: the risks associated with sexting, the fun and carefree nature of sexting, and relational expectations about sexting.¹⁸ However, these were developed using a small pool of items, and research has yet to measure the full range of sexting expectancies or examine how these expectancies influence decisions to sext.

This study is the first cross-sectional test of how sexting expectancies influence sexting behaviors. Success of this cross-sectional study would suggest the viability of examining this theory in future longitudinal design, thus we view this as an important first step in the development of this prospective model. We hypothesized that there would be two overarching domains of sexting expectancies: positive expectancies, encompassing expectations of sexual-related behaviors as well as positive feelings associated with sexting, and negative expectancies, entailing negative feelings or outcomes that could result from sexting.¹⁸ We also wanted to explore whether sexting behaviors and expectancies would differ by gender, sexual identity, relationship status, or race in the sample, due to mixed findings on demographic differences in sexting. ^{12–13,18,19,21} Finally, we hypothesized that positive sexting expectancies would be positively related to engagement in sexting behaviors, while negative expectancies would be negatively related to engagement in sexting behaviors.

Methods

Participants

The original sample consisted of 611 undergraduate students enrolled at a large, public university in Midwestern United States. Due to the disproportional number of females in the original sample, we randomly selected 139 females from the sample using SPSS software, resulting in a total sample of N=278 that was used for data analyses. Males and females did not significantly differ across other demographic variables. Mean age of the sample was 21.0 years (SD=4.56, range 18–43); 53.8% female, and 76.3% Caucasian. Heterosexuals made up 91.4% of the sample (3.9% homosexual, 3.6% bisexual, 1.1% “other”), and 42.7% were in a serious relationship (38.0% single, 9.7% dating, 5.0% married, 4.7% cohabitating). Participants were recruited through an Internet database available to all students enrolled in introductory psychology courses. Participants had to be at least 18 years old to participate and received course credit for their participation. The following measures and informed consent were completed via an online survey database in accordance with Indiana University Institutional Review Board (IRB) approval.

Measures

The Sexting Behaviors Scale (SBS).^30–31

The SBS consists of 11 items: Ten items are measured on a five-point Likert scale from 1 (never) to 5 (frequently) and assess the frequency of sending and receiving texts and pictures via mobile phone or social-networking site (SNS). One item assesses lifetime number of sexting partners. The scale had an internal consistency of α=0.81 in the current study and showed evidence of convergent and discriminant validity in the developmental sample, showing significant relationships between SBS scores and impulsivity (r's range from 0.05–0.22), sexual behaviors (r=0.37), problematic mobile phone use (r=0.20), and Internet use (r=0.17; p<0.001 for all samples).³¹

Sextpectancies pilot measure

In order to understand and identify common beliefs and expectations about sexting outcomes, we created a pilot measure of sexting expectancies. Pilot items were taken from four sources: (1) Existing expectancy measures, such as the A.E. Max for alcohol,³² and the Sex-Related Alcohol Expectancies Measure;²⁴ (2) existing qualitative and quantitative data about sexting outcomes and reasons for sexting;^5,16 (3) the sexting attitudes proposed previously;¹⁸ and (4) qualitative pilot interviews²⁷ with currently sexting young adults. Based on these findings, initial development of a measure of sexting expectancies was formulated on the idea that people's beliefs about sexting outcomes would be centered on expected outcomes for both the self and for other people³³ in order to encompass the two sources of learning: one's own experiences and socially-learned experiences of those around the person.³⁴ We also thought that expectancies could differ based on sending sext messages and receiving sext messages, as these are unique behavioral outcomes. Thus, the pilot measure consisted of items assessing expectancies of both sending sexts (51 items) and receiving sexts (30 items), measured on a Likert scale from 1 (not true at all) to 4 (extremely true). We used item stems, Receiving sexts makes one…and Sexting makes one…, similar to those used in the creation of the A.E. Max (Alcohol makes one…).³² In addition to these quantitative items, four qualitative open-ended questions were given to assess for other content not covered in the measure (including Sexting makes one/me…; I sext/do not sext because…). We presented qualitative items first to avoid bias from other items.

Demographics measure

Participants also self-reported on age, gender, sexual identity (heterosexual, homosexual, bisexual, other), race, as well as relationship status—single, dating (involved with individual[s] but not in an exclusive, committed, or monogamous relationship), in a serious relationship (exclusive, committed, and monogamous), cohabitating with a partner, or married.

Data analyses

All analyses were conducted using SPSS Version 20.³⁵ In order to determine the factor structure of the Sextpectancies measure, qualitative data collected was reviewed to assess for common responses not represented in the pilot measure. Based on the data, one item was added to the measure based on high response rates (21 participants endorsed): Sexting makes one excited. No other unique qualitative responses were found.

Next, as a first step in examining the factor structure and validity of the sextpectancies measure, and to determine items that should be retained on the final scale, exploratory principal axis factor analyses were conducted using oblimin rotation followed by parallel analyses to determine the number of meaningful factors to retain.^36–37 We ran two analyses—one exploratory factor analysis (EFA) on sending sext items and one EFA on receiving sext items—and restricted each to two factors. Items were retained on a factor if they had a loading of at least 0.40 and no cross-loadings of more than 0.25.

In order to examine differences in frequencies of specific sexting behaviors and sexting expectancies across demographic groups, we computed analyses of variance (ANOVA) and post-hoc planned comparisons using independent samples t-tests. ANOVAs were also conducted to examine study hypotheses as follows: (1) gender, relationship status, and sexual identity as predictors of positive and negative sending and receiving expectancies (four separate ANOVAs were conducted); (2) gender, sexual identity, relationship status, and Sextpectancies subscales as predictors of sexting behaviors (two separate ANOVAs were run for sending and receiving scales). We calculated eta-squared values to estimate effect sizes and interpreted them based on recommendations of values for small (≥0.01), medium (≥0.06), and large (≥0.14) effect sizes.^38–39

Results

Prevalence of sexting and differences in sexting frequencies by demographics

Overall, sexting prevalence in the current sample was high: 80.3% reported receiving and 67.4% reported sending sext text messages, while fewer reported sending (46.6%) and receiving (64.2%) sext pictures. However, individuals are not sexting frequently; individuals reported “rarely” (less than three times a month) sending and receiving sext texts [M(SD)=2.15 (1.1) and M(SD)=2.38(1.0), respectively] or sext pictures [M(SD)=1.64(0.82) and M(SD)=1.87(0.83), respectively] via mobile phone.

Gender, relationship status, and sexual identity were significant predictors of specific sexting behaviors (Table 1). Follow-up t-tests indicated that males reported receiving pictures via SNS (t=2.99, p=0.003) and mobile phone (t=2.62, p=.01) more frequently, sending more pictures via SNS (t=3.00, p=.003), and sexting with more partners (t=2.04, p=0.04). For sexual identity, those who identified as homosexual or bisexual reported receiving pictures via mobile phone more frequently compared to those who identified as “other” (t=3.67, p=.03, and t=2.21, p=.03, respectively). Based on relationship status, single individuals sent sexts via mobile phone less frequently than those who were dating, in a serious relationship, and cohabitating (t=−2.86, p=0.01; t=−3.28, p=0.001; t=−3.05, p=0.002); and those who were cohabitating sent more mobile sexts than those who were married (t=2.13, p=0.03).

Table 1.

ANOVA for Sexting Behaviors by Age, Sex, Sexual Orientation, Race, and Relationship Status

	F (p value)
	Received sext mobile	Received pic mobile	Received pic SNS	Sent pic mobile	Sent sext mobile	Sent pic SNS	Post pic SNS	Number sext partners
Mean (SD) Total sample	2.38 (1.0)	1.87 (0.83)	1.42 (0.74)	1.64 (0.82)	2.15 (1.0)	1.16 (0.54)	1.11 (0.47)	2.16 (4.1)
Age (df=20)	0.68 (.84)	0.72 (0.80)	0.82 (0.70)	0.56 (0.94)	0.82 (0.69)	0.42 (0.98)	0.23 (1.0)	0.77 (0.75)
Gender^a (df=1)	0.34 (0.56)	7.90 (0.005)	9.03 (0.003)	1.93 (0.17)	0.02 (0.88)	7.68 (0.01)	2.92 (0.08)	6.98 (0.01)
Sexual identity^b (df=3)	0.58 (0.63)	1.78 (0.15)	3.83 (0.01)	0.57 (0.63)	0.87 (0.46)	0.77 (0.51)	1.55 (0.20)	0.35 (0.79)
Race^c (df=4)	0.35 (0.85)	0.76 (0.55)	1.63 (0.17)	0.93 (0.45)	0.42 (0.80)	1.62 (0.17)	1.86 (0.12)	0.37 (0.83)
Relationship status^d (df=4)	1.43 (0.23)	0.80 (0.53)	0.53 (0.72)	1.96 (0.10)	4.06 (0.003)	1.10 (0.36)	0.28 (0.89)	2.26 (0.06)

Significant values are bolded.

Gender coded as 1=male, 2=female.

Sexual identity coded as 1=heterosexual, 2=bisexual, 3=homosexual, 4=other.

Race coded as 1=Caucasian, 2=African American, 3=Asian American, 4=Hispanic American, 5=other.

Relationship status coded as 1=single, 2=dating (nonexclusive and nonmonogamous relationship), 3=serious relationship (committed and monogamous relationship), 4=cohabitating, 5=married.

Development and factor structure of the sextpectancies measure

Examination of eigenvalues, the scree plot, and the results of the parallel analyses suggested two domains of sending sext expectancies and two domains of receiving sext expectancies (Table 2).

Table 2.

Sextpectancies Measure Factor Loadings

Expectancy subscales (and subscale items)	Factor one loading	Factor two loading
F1: Sending Positive Expectancy Scale^a
Sexting makes one adventurous.	0.60	−0.07
Sexting makes one feel attractive.	0.66	−0.05
Sexting makes one excited.	0.67	−0.05
Sexting makes one playful.	0.61	−0.14
Sexting makes one affectionate.	0.49	−0.15
Sexting makes one intimate.	0.60	−0.14
Sexting makes relationships more interesting.	0.68	−0.18
Sexting makes one aroused.	0.63	0.06
Sexting makes one horny.	0.68	0.06
Sexting makes more likely to “hook up.”	0.45	0.20
Sexting makes one more likely to have sex.	0.48	0.16
Sexting makes it easier to flirt.	0.56	−0.09
Sexting makes one more likeable.	0.44	−0.01
Sexting makes one feel sexy.	0.73	−0.07
Sexting makes one more attracted to others.	0.61	−0.04
F1: Sending Negative Expectancy Scale^b
Sexting makes one feel awkward.	−0.19	0.61
Sexting makes one feel embarrassed.	−0.18	0.62
Sexting makes one feel dirty.	0.13	0.61
Sexting lowers ones self-esteem.	−0.05	0.66
Sexting makes one vulnerable.	0.16	0.54
Sexting makes one desperate.	0.02	0.65
Sexting makes one immature.	−0.08	0.66
Sexting makes one feel foolish.	−0.02	0.73
Sexting makes one feel guilty.	−0.04	0.70
Sexting makes one feel ashamed.	−0.09	0.76
Sexting makes one feel inappropriate.	−0.02	0.76
F1: Receiving Negative Expectancy Scale^c
Receiving sexts makes one feel disgusted.	0.77	−0.09
Receiving sexts makes one feel ashamed.	0.81	−0.06
Receiving sexts makes one feel guilty.	0.78	−0.02
Receiving sexts lowers one's self-esteem.	0.58	−0.07
Receiving sexts makes one avoid the sender.	0.70	−0.15
Receiving sexts makes one feel vulnerable.	0.60	0.10
Receiving sexts makes one feel awkward.	0.71	−0.13
Receiving sexts makes one feel uncomfortable.	0.70	−0.16
Receiving sexts makes one feel insulted.	0.68	−0.14
Receiving sexts makes one feel dirty.	0.74	0.06
Receiving sexts makes one feel embarrassed.	0.83	0.01
F2: Receiving Positive Expectancy Scale^d
Receiving sexts makes one happy.	−0.18	0.67
Receiving sexts makes one attracted to the sender.	−0.07	0.62
Receiving sexts makes one feel sexy.	−0.03	0.71
Receiving sexts gives one confidence.	−0.06	0.68
Receiving sexts makes one feel excited.	−0.01	0.75
Receiving sexts makes one feel admired.	−0.12	0.67
Receiving sexts raises one's self-esteem.	−0.06	0.69
Receiving sexts makes on horny.	−0.11	0.69
Receiving sexts makes one feel wanted.	−0.08	0.68
Receiving sexts makes one feel more attractive.	−0.08	0.72
Receiving sexts makes one want to have sex.	−0.10	0.62

α=0.91, overall reliability for the sending positive expectancy scale.

α=0.85, overall reliability for the sending negative expectancy scale.

α=0.93, overall reliability for the receiving positive expectancy scale.

α=0.92, overall reliability for the receiving negative expectancy scale.

The following factors of sending expectancies were: positive expectancies (e.g., Sexting makes one feel sexy and Sexting makes one excited; eigenvalue=7.78; 27.78% of explained variance) and negative expectancies (e.g., Sexting makes one embarrassed; eigenvalue=4.55; 16.25% of explained variance). As hypothesized, items pertaining to both the fun flirtatious nature of sexting, as well as sexual expectations associated with sexting, both loaded on Factor 1.

For receiving expectancies, there was a negative expectancy domain (e.g., Receiving sexts makes one feel uncomfortable; eigenvalue=9.37; 37.48% of explained variance) and a positive expectancy domain (e.g., Receiving sexts gives one confidence; eigenvalue=4.14; 16.55% of explained variance) (Table 2). In general, the Sextpectancy subscales showed good internal consistency (subscales ranged from α=0.89 to 0.93).

Sextpectancies and differences by demographics

Overall, there were common negative and positive beliefs about sexting, and these expectancies were similarly endorsed for both sending sexts [positive M(SD)=2.38(0.66); negative M(SD)=2.25(79)] and receiving sexts [positive M(SD)=2.38(.77); negative M(SD)=2.01(0.84)], suggesting that individuals hold these beliefs as “somewhat” to “quite a bit” true. When demographic variables were entered as predictors of sextpectancies in an ANOVA (Table 3), (1) there were no significant predictors of positive or negative sending expectancies; (2) gender (F=4.64, p=0.03) and race (F=3.01, p=0.02) significantly predicted positive receiving expectancies; and (3) gender (F=6.11, p=0.01) and relationship status (F=4.24, p=0.002) predicted negative receiving expectancies. Follow-up contrasts suggested (1) males reported stronger positive expectancies about receiving sexts (t=2.07, p=0.04), while females reported stronger negative expectancies (t=–2.14, p=0.03) about receiving sexts; (2) Hispanics reported higher positive receiving expectancies than those who identified as “other” (t=2.08, p=0.05); and (3) single individuals had stronger negative receiving expectancies than those who were dating, single, cohabitating, or married (t=2.13, p=0.03; t=2.86, p=0.01; t=2.86, p=0.01; and t=2.29, p=0.02, respectively).

Table 3.

ANOVA of Sexting Expectancies Predicted by Demographics

	Sending expectancies F (p value)		Receiving expectancies F (p value)
Demographic variable	Positive	Negative	Positive	Negative
Age (df=1)	10.42 (0.11)	10.51 (0.08)	0.87 (0.63)	10.55 (0.07)
Gender^a (df=1)	0.85 (0.36)	20.94 (0.09)	40.64 (0.03)	60.11 (0.01)
Sexual identity^b (df=3)	20.12 (0.10)	0.62 (0.61)	0.27 (0.85)	0.54 (0.66)
Race^c (df=4)	20.04 (0.09)	10.57 (0.19)	30.01 (0.02)	0.93 (0.45)
Relationship status^d (df=4)	10.79 (0.13)	20.25 (0.07)	20.10 (0.08)	40.24 (0.002)

Significant values are bolded.

Gender coded as 1=male, 2=female.

Sexual identity coded as 1=heterosexual, 2=bisexual, 3=homosexual, 4=other.

Race coded as 1=Caucasian, 2=African American, 3=Asian American, 4=Hispanic American, 5=other.

Relationship status coded as 1=single, 2=dating (nonexclusive and nonmonogamous relationship), 3=serious relationship (committed and monogamous relationship), 4=cohabitating, 5=married.

Prediction of sexting by sextpectancies and demographics

An ANOVA (Table 4) indicated that, for sending expectancies, positive and negative expectancies were both significantly related to sexting behaviors (F=48.37, p<0.001 and F=31.74, p=0.01, respectively). Calculated eta-squared values revealed medium to large effect sizes for both positive and negative expectancies (η²=0.12 and η²=0.09, respectively). For receiving expectancies, both positive and negative expectancies were associated with sexting and had small-to-medium effect sizes (F=27.60, p<0.001, η²=0.07; and F=11.12, p=0.001, η²=0.04, respectively). Higher levels of positive expectancies were associated with more frequent sexting (t=6.90, p<0.001 for sending; t=4.84, p<0.001 for receiving), and higher levels of negative expectancies were associated with lower rates of sexting (t=−5.96, p<0.001 for sending; t=−3.85, p<0.001 for receiving). None of the demographic variables were significant predictors of overall sexting behaviors.

Table 4.

Prediction of Sexting by Sextpectancies and Demographics

DV: Sexting behaviors	F	df	p	η²
Gender	0.32	1	0.57	0.00
Sexual identity	10.88	3	0.13	0.01
Relationship status	10.61	4	0.17	0.00
Positive sending expectancies	450.98	1	<0.001	0.12
Negative sending expectancies	360.65	1	<0.001	0.09
Gender	0.04	1	0.84	0.00
Sexual identity	10.88	3	0.14	0.01
Relationship status	0.95	4	0.43	0.00
Positive receiving expectancies	220.42	1	<0.001	0.07
Negative receiving expectancies	140.41	1	<0.001	0.04

N=278, positive and negative expectancies scales were taken from the Sextpectancies Measure. Gender, sexual identity, and relationship status were taken from self-report demographics. SPSS results presented as partial eta-squared values; thus, classical eta-squared values were computed by η²⁼ SS_factor/SS_total.

Discussion

In this study, we began to examine the validity of the Sextpectancies Measure in order to identify common expectancies about sexting behaviors. Exploratory factor analyses indicated that there were two overall expectancy factors each for sending and receiving sexts: positive expectancies, including references to both the fun and sexual-related aspects of sexting, and negative expectancies. These are the first empirical findings concerning existing domains of sexting expectancies and should be replicated using confirmatory factor analysis (CFA), both in an independent sample of college students and in other samples of interest, including adolescents, who might endorse different sexting expectancies. In general, the current study findings are consistent with other reported reasons for sexting, such as doing so to be fun and flirtatious^11,18 or “to explore or experiment with sexuality.”¹⁶

These findings are the first test of a temporal, causal model in which sexting expectancies are thought to influence one's decision to engage in sexting behaviors. Findings are consistent with this model and suggest the viability of future research to examine this relationship both prospectively and experimentally. This is the first data to suggest that there are positive and negative domains of sexting expectancies that could potentially be important for future engagement in sexting behaviors. We found that positive sending and receiving expectancies were positively associated with sexting, while negative sending and receiving expectancies were negatively associated with sexting. The causal direction of this model cannot be determined by this study; therefore, it is important to examine these factors prospectively in future work.

Replication of these findings in longitudinal designs would suggest viability of using clinical approaches targeting expectancies in order to prevent sexting behaviors. For example, there is evidence that altering or challenging one's alcohol-related expectancies can lead to changes in drinking behaviors.^40–41 Thus, expectancy challenge techniques could help reduce or prevent negative outcomes associated with sexting, such as unwanted sexual behavior.³¹

Prevalence rates of sexting vary across studies,^9–11,16,42 and although sexting was prevalent in the study sample, we found that most individuals are sexting only occasionally or rarely. These frequency rates are similar to what has been shown in previous samples of young adults,¹⁹ suggesting that while many have experimented with sexting at least once, most individuals do not sext regularly. Further investigation is needed to determine to what extent frequency of sexting might be related to negative sexting consequences, to what extent the Internet's disinhibitory effects may prompt risky sexting,⁴³ and how frequency of sexting might be related to personality-based dispositions of cautiousness or impulsivity. Although individuals who sext more often might be at a higher risk for negative outcomes simply due to probability, this does not mean that those sexting infrequently are not prone to risk.

Moreover, there are mixed findings across studies as to whether there are differences in sexting across gender,¹³ relationship status,^18,19 and sexual identity;^15,42 however, this is the first study to examine how these differences in sexting may be due to differences in beliefs about sexting outcomes. With respect to sexting expectancies, females reported stronger negative expectancies about receiving sexts. Also, single individuals had significantly stronger negative expectancies about sending and receiving sexts compared to others who were in some romantic relationship. The current data also examined differences in sexting expectancies based on sexual identity; however, these data are limited by power concerns due to the largely heterosexual sample, thus further examination in a larger sample is necessary to examine the robustness of such findings. Additionally, although there were some differences across demographics on specific sexting behaviors, demographics were not significantly associated with overall sexting behaviors when expectancies were considered. Thus, it could be that differences in sexting behaviors are driven by sexting expectancies, and that differences across groups in expectancies are more important in explaining differences in behavior than individual demographic characteristics. This should be examined in a prospective design.

Differences in sexting expectancies also suggest that sexting may be riskier for some individuals more than others. For one, based on single individuals' stronger endorsement of negative expectancies, sexting may be more risky than those who are in a committed relationship; however, this correlational relationship cannot be proof of causation. Therefore, future prospective designs could examine differential negative outcomes across demographic groups, as well as whether or not certain sexting behaviors, such as sending pictures vs. sending messages, might be riskier than others.

The current study had some limitations. The study focused only on a population of college students, and thus the nature of sexting among adolescents or older adults is still unknown. Additionally, the predominantly Caucasian and heterosexual sample potentially limits generalizability, and the cross-sectional nature of the study limits causality conclusions. Also, there are a number of limitations to using online self-reports (i.e., low response rates);⁴⁴ however, despite limitations, online surveys do have data to support their validity and are an adequate design for preliminary examination of a causal model.⁴⁴ Additionally, although the current study provides initial evidence for the reliability and validity of the Sexting Behaviors Scale and the Sextpectancies Measure, continued research should focus on examining, extending, and improving the validity of the these measures, especially as they might be used for different purposes (screening rather than research) or with different samples (e.g., adolescents). Despite limitations, this study adds to the literature on the nature of sexual activity and sexual communication through digital media and how one's beliefs about sexting may influence their decision to sext.^2,45

Footnotes

Author Disclosure Statement

No competing financial interests exist.

References

Chalfen

. ‘It's only a picture’: sexting, ‘smutty’ snapshots and felony charges. Visual Studies, 2009; 24:258–268.

Weiss

, Samenow

. Smart phones, social networking, sexting and problematic sexual behaviors – A call for research. Sexual Addiction & Compulsivity, 2010; 17:241–246.

Diliberto

, Mattey

. Sexting: Just how much of a danger is it and what can school nurses do about it? NASN School Nurse, 2009; 24:262–267.

Livingstone

. Taking risky opportunities in youthful content creation: Teenagers' use of social networking sites for intimacy, privacy and self-expression. New Media & Society, 2008; 10:393–411.

Manzo

. Administrators confront student ‘sexting.’ Education Week, 2009; 28:13–16.

Ringrose

, Eriksson Barajas

. Gendered risks and opportunities? Exploring teen girls' digital sexual identity in postfeminist media contexts. International Journal of Media and Cultural Politics, 2011; 7:121–138.

Wolak

, Finkelhor

, Mitchell

. Child pornography possessors: Trends in offender and case characteristics. Sexual Abuse, 2011; 23:122–142.

Jolicoeur

, Zedlewski

. Much ado about sexting. National Institute of Justice. 2010. NCJ, 230795:1–13.

Lounsbury

, Mitchell

, Finkelhor

. The true prevalence of “sexting.” Crimes Against Children Research Center, 2011; 1–9.

10.

Mitchell

, Finkelhor

, Jones

, Wolak

. Prevalence and characteristics of youth sexting: A national study. Pediatrics, 2012; 129:13–20.

11.

The National Campaign to Prevent Unplanned and Teen Pregnancy. Sex and tech: Results from a survey of teens and young adults. Sex and Tech, 2008; 1–20.

12.

Benotsch

, Snipes

, Martin

, Bull

. Sexting, substance use, and sexual risk behavior in young adults. Journal of Adolescent Health, 2012; 2–6.

13.

Gordon-Messer

, Bauermeister

, Grodzinski

, Zimmerman

. Sexting among young adults. Journal of Adolescent Health, 2012; 2–6.

14.

Stinson

. Hooking up in young adulthood: A review of factors influencing the sexual behavior of college students. Journal of College Student Psychotherapy, 2010; 24:98–115.

15.

Dake

, Price

, Maziarz

, Ward

. Prevalence and correlates of sexting behavior in adolescents. American Journal of Sexuality Education, 2012; 7:1–15.

16.

Lenhart

. Teens and sexting. Pew Internet & American Life Project. 2009; 1–16.

17.

Fielder

, Carey

. Predictors and consequences of sexual “hookups” among college students: A short-term prospective study. Archives of Sexual Behavior, 2009; 29:1105–1119.

18.

Weisskirch

, Delevi

. “Sexting” and adult romantic attachment. Computers in Human Behavior, 2010; 2:1697–1701.

19.

Drouin

, Landgraff

. Texting, sexting, and attachment in college students' romantic relationships. Computers in Human Behavior, 2012; 28:444–449.

20.

Rice

, Rhoades

, Winetrobe

et al. Sexually explicit cell phone messaging associated with sexual risk among adolescents. Pediatrics, 2012; 130:667–673.

21.

Jung

. 2010. Alcohol, other drugs, and behavior: Psychological research perspectives. 2nd. Los Angeles: SAGE.

22.

Goodson

, McCormick

, Evans

. Searching for sexually explicit materials on the Internet: An exploratory study of college students' behavior and attitudes. Archives of Sexual Behavior, 2001; 30.2:101–118.

23.

Fischer

, Smith

, Anderson

, Flory

. Expectancy influences the operation of personality on behavior. Psychology of Addictive Behaviors, 2003; 17:108–114.

24.

Dermen

, Cooper

. Sex-related alcohol expectancies among adolescents: I Scale development. Psychology of Addictive Behaviors, 1994; 8:152–160.

25.

Leigh

. In search of the seven dwarves: Issues of measurement and meaning in alcohol expectancy research. Psychological Bulletin, 1989; 105:361–373.

26.

Smith

, Toadvine

, Kennedy

. Women's perceptions of alcohol-related sexual disinhibition: Personality and sexually-related alcohol expectancies. International Journal of Sexual Health, 2009; 21:119–131.

27.

Carey

. Alcohol-related expectancies predict quantity and frequency of heavy drinking among college students. Psychology of Addictive Behaviors, 1995; 9.4:236–241.

28.

Walters

, Contri

. Outcome expectancies for gambling: Empirical modeling of a memory network in federal prison inmates. Journal of Gambling Studies, 1998; 14:173–191.

29.

Smith

, Simmons

, Flory

, Annus

, Hill

. Thinness and eating expectancies predict subsequent binge-eating and purging behavior among adolescent girls. Journal of Abnormal Psychology, 2007; 116:188–197.

30.

Dir

, Cyders

, Coskunpinar

. From the bar to the bed via mobile phone: A first test of the role of problematic alcohol use, sexting, and impulsivity-related traits in sexual hookups. Computers in Human Behavior, 2013; 29:1664–1670.

31.

Dir

, Steiner

, Coskunpinar

, Cyders

. Sexting behaviors scaleUnpublished technical report.

32.

Goldman

, Darkes

. Alcohol expectancy multiaxial assessment: A memory network-based approach. Psychological Assessment, 2004; 16:4–15.

33.

Cooper

, Shapiro

, Powers

. Motivations for sex and risky sexual behavior among adolescents and young adults: A functional perspective. Journal of Personality and Social Psychology, 1998; 75:1528–1558.

34.

Smith

, Fischer

, Fister

. Incremental validity principles in test construction. Psychological Assessment, 2003; 14:467–477.

35.

SPSS Inc. SPSS Version 20.0 for Windows 2011. Chicago: SPSS Inc.

36.

Fabrigar

, Wegener

, MacCallum

, Strahan

. Evaluating the use of exploratory factor analysis in psychological research. Psychological Methods, 1999; 4:272–299.

37.

Velicer

, Eaton

, Fava

. 2010. Construct explication through factor or component analysis: A review and evaluation of alternative procedures for determining the number of factors or components. Goffin

, Helmes

. Problems and solutions in human assessment: Honoring Douglas Jackson at Seventy. Boston: Kliwer, 41–71.

38.

Cohen

. 2010. Statistical power analysis for the behavioral sciences. Hillsdale, NJ: Erlbaum.

39.

Levine

, Hullett

. Eta squared, partial eta squared, and misreporting of effect size in communication research. Human Communication Research, 2002; 28.4:612–625.

40.

Darkes

, Goldman

. Expectancy challenge and drinking reduction: Experimental evidence for a mediational process. Journal of Consulting and Clinical Psychology, 1993; 61.2:344–353.

41.

Darkes

, Goldman

. Expectancy challenge and drinking reduction: Process and structure in the alcohol expectancy network. Experimental and Clinical Psychopharmacology, 1998; 6.1:64–76.

42.

Ferguson

. Sexting behaviors among young Hispanic women: Incidence and association with other high-risk sexual behaviors. Psychiatric Quarterly, 2011; 82:239–243.

43.

Suler

. The online disinhibition effect. Cyberpsychology & Behavior, 2004; 7.3:321–326.

44.

Evans

, Mathur

. The value of online surveys. Internet Research, 2005; 15:195–219.

45.

Moreno

, Swanson

, Royer

, Roberts

. Sexpectations: Male college students' views about displayed sexual references on females' social networking web sites. Journal of Pediatric and Adolescent Gynecology, 2010; 24:85–89.

Understanding Differences in Sexting Behaviors Across Gender,Relationship Status,and Sexual Identity,and the Role of Expectancies in Sexting

Abstract

Abstract

Introduction

Methods

Participants

Measures

The Sexting Behaviors Scale (SBS). 30–31

Sextpectancies pilot measure

Demographics measure

Data analyses

Results

Prevalence of sexting and differences in sexting frequencies by demographics

Development and factor structure of the sextpectancies measure

Sextpectancies and differences by demographics

Prediction of sexting by sextpectancies and demographics

Discussion

Footnotes

Author Disclosure Statement

References

The Sexting Behaviors Scale (SBS).^30–31