Further Understanding the Correlations Between Sexting and Mental Health: Considerations for Sex and Sexual Identity

Abstract

We explored how sexting (no sexting, sent only, received only, reciprocal) was associated with depression, anxiety, sleep problems, and compulsive sexual behaviors across sex and sexual identity groups. We also examined how substance use predicted sexting classifications. Data came from 2,160 college students living in the United States. Results indicated that 76.6 percent of the sample had engaged in sexting (primarily reciprocal). Participants who had engaged in sexting generally evidenced higher depression, anxiety, sleep problems, and compulsive sexual behaviors. Effect sizes were largest on compulsive sexual behavior indicators. Marijuana use was the only significant substance use predictor of reciprocal and “received only” sexting compared with no sexting. Illicit substance use (e.g., cocaine) was low base rate but descriptively was associated with sexting. Compulsive sexual behavior remained a robust positive correlate of sexting compared with non-sexting participants regardless of sex and sexual identity. Most other mental health indicators became nonsignificant correlates of sexting in nonheterosexual participants and were weak positive correlates in heterosexual participants. Marijuana use remained the only significant substance use predictor of reciprocal and “received only” sexting when adjusted for sex and sexual identity. We conclude that sexting is only weakly associated with depression, anxiety, and sleep problems but robustly associated with compulsive sexuality and marijuana use. These findings do not appear to vary meaningfully on the basis of sex or sexual identity, with the exception that the effect sizes between sexting and compulsive sexual behaviors were much stronger for females compared with males (regardless of sexual identity).

Introduction

Defined as sending a sexually explicit image of oneself over text messaging, the term “sexting” was first adopted by the Daily Telegraph in 2005 and became official English vernacular in 2009.^1,2 Findings suggest sexting to be a common practice among adolescents and emerging adults.^3,4 For example, Stasko and Geller reported data from 870 community participants in the United States indicating 88 percent of participants having ever sexted and 82 percent reporting having sexted in the past year.⁵ Meta-analytic findings of emerging adult samples indicate lower, but still notable, base rates with 38.3 percent reporting having sent a sext and 41.5 percent reporting having received a sext.³

Given the popularity of sexting, concerns have been raised regarding how it is associated with various outcomes, including sexual activity,^6–8 cyberbullying victimization,^9,10 and identity development.^11,12 Given such constructs often influence psychopathology, there is concern about how sexting might be associated with mental health outcomes.^1,13 Current findings suggest that the relationship between mental health constructs and sexting behaviors is mixed. For instance, some research demonstrates no relationship,^14,15 whereas other studies suggest a negative correlation between sexting and constructs such as depression.¹⁶ Conversely, researchers have also reported positive associations between sexting and mental health indicators, including substance use and internalizing problems.^13,17,18

These discrepancies suggest that sexting might be differentially correlated with mental health problems depending on individual differences. Basic identity indicators likely have some function, but research examining how identity influences the relationship between sexting and mental health is limited. Many identity variables might play a role. However, given the limited extant base, considerations for sex (as assigned at birth) and sexual identity are particularly important. These demographics were selected for this article as sexting is clearly a sexual activity, and like most sexual behaviors, interactions between parties are, in part, determined along sex and sexual identity lines. As such, understanding how the relationship between sexting and mental health indicators differs as a function of sex and sexual identity is a conceptually relevant area in need of further research.

Within this context, we sought to re-examine a variety of mental health constructs, such as depression, anxiety, and various substance use indicators, that have been explored in conjunction with sexting in previous studies.^{1,13,15,18–22} We also wanted to extend the literature by examining compulsive sexual behavior and illicit substance use (e.g., narcotic use), which have been less studied. Compulsive sexual behavior disorder (CSBD) was recently added to the International Classification of Diseases-11.²³ It is possible that sexting functions as a manifestation of CSBD in some cases, yet literature on CSBD in general is quite limited. Previous research has connected sexting to measures that are conceptually similar to CSBD, such as sexual pre-occupation.²⁴ Additionally, there have been calls in the literature for more research examining a potential connection between problematic sexual behaviors and sexting.²⁵

Conceptually, it is unlikely that sexting behaviors themselves are proximal causes of psychological distress or substance use. However, a better understanding of the connection between sexting and mental health/substance use, and how important demographic factors (sex and sexual identity) differentiate the effects of those correlates, may help researchers to better understand the causes of their shared relationships. Prior researchers have noted sexting to be correlated with a myriad of outcomes but no overarching theory connecting why such outcomes occur.²⁶

It is hoped that a broader study focusing more strongly on the effect sizes of the connections between sexting and various mental health/substance use indicators will help researchers in future theory development. As such, the current study is exploratory and does not operate from a specific framework. A single broad research question was posited: to what extent do the effect sizes between sexting classifications and mental health/substance use change as a function of sex and sexual identity? No hypotheses were generated for this study.

Methods

Participants/procedure

Participants were students from two universities in the United States: one in the Intermountain West and the other in the Deep South. Data gathering occurred between March 2020 and May 2021. Participants completed a survey that was part of a larger institutional review board (IRB)-approved project on sexual health. All participants consented before participation and were compensated with extra credit. Originally, 2,288 participants opened the survey. However, we did not analyze participants who did not complete at least 80 percent of our measures or those who evidenced extreme univariate outlier scores on any of our outcome variables (z's > 3.29). Upon data cleaning, 2,160 participants had data fit for analyses. For a full breakdown of demographics, see Table 1 (additional demographics are shown in Supplementary File S1). Demographics were all assessed via self-report.

Table 1.

Study Sample Characteristics Stratified by Sexting Behavior

Sexting classification
Variable	No sexting (n = 506)	Sent only (n = 44)	Received only (n = 374)	Reciprocal (n = 1,236)
Gender, n (percent)
Man	178 (35.2)	6 (13.6)	136 (36.4)	323 (26.1)
Woman	318 (62.8)	36 (81.8)	236 (63.1)	878 (71.0)
Transgender woman	1 (0.2)	0 (0.0)	1 (0.3)	3 (0.2)
Transgender man	0 (0.0)	0 (0.0)	0 (0.0)	3 (0.2)
Nonbinary	9 (1.8)	2 (4.5)	1 (0.3)	29 (2.3)
Sex, n (percent)
Male	179 (35.4)	6 (13.6)	137 (36.6)	332 (26.9)
Female	325 (64.2)	37 (84.1)	236 (63.1)	902 (73.0)
Declined to say	1 (0.2)	1 (2.3)	1 (0.0)	2 (0.2)
Sexual identity, n (percent)
Gay/lesbian	5 (1.0)	2 (4.5)	12 (3.2)	52 (4.2)
Bisexual	100 (19.8)	12 (27.3)	82 (21.9)	389 (31.5)
Questioning	14 (2.8)	1 (2.3)	9 (2.4)	11 (0.9)
Asexual	10 (2.0)	0 (0.0)	4 (1.1)	3 (0.2)
Pansexual	6 (1.2)	2 (4.5)	4 (1.1)	44 (3.6)
Heterosexual	371 (73.3)	27 (61.4)	263 (70.3)	737 (59.6)
Alcohol, n (percent)
Yes	126 (24.9)	25 (56.8)	167 (44.7)	773 (62.5)
No	380 (75.1)	19 (43.2)	206 (55.1)	461 (37.3)
Tobacco, n (percent)
Yes	129 (25.5)	25 (56.8)	171 (45.7)	778 (62.9)
No	377 (74.5)	19 (43.2)	203 (54.3)	455 (36.8)
Marijuana, n (percent)
Yes	44 (8.7)	12 (27.3)	93 (24.9)	475 (38.4)
No	459 (90.7)	32 (72.7)	280 (74.9)	759 (61.4)
Cocaine, n (percent)
Yes	1 (0.2)	2 (4.5)	10 (2.7)	39 (3.2)
No	504 (99.6)	42 (95.5)	364 (97.3)	1,195 (96.7)
Heroin, n (percent)
Yes	1 (0.2)	1 (2.3)	2 (0.5)	12 (1.0)
No	505 (99.8)	42 (95.5)	370 (98.9)	1,221 (98.8)
Club drugs, n (percent)
Yes	3 (0.6)	0 (0.0)	6 (1.6)	58 (4.7)
No	503 (99.4)	44 (100.0)	367 (98.1)	1,178 (95.3)
Painkiller, n (percent)
Yes	30 (5.9)	3 (6.8)	26 (7.0)	74 (6.0)
No	474 (93.7)	41 (93.2)	348 (93.0)	1,160 (93.9)
Methamphetamine, n (percent)
Yes	1 (0.2)	0 (0.0)	4 (1.1)	11 (0.9)
No	504 (99.6)	44 (100.0)	369 (98.7)	1,223 (98.9)
Sedative, n (percent)
Yes	12 (2.4)	1 (2.3)	20 (5.3)	65 (5.3)
No	490 (96.8)	43 (97.7)	352 (94.1)	1,169 (94.6)
Age, M (SD)	19.52 (3.57)	19.14 (1.39)	19.75 (3.84)	20.24 (4.41)

Note: Instances where cell numbers do not equal overall group numbers were associated with missing values.

M, mean; SD, standard deviation.

Measures

Sexting

Receiving a sext was measured by: “Have you ever received a nude or partially nude picture of a person you know (i.e., sexting) to your mobile device?” Sending a sext was measured by: “Have you ever sent a nude or partially nude picture of yourself (i.e., sexting) from your mobile device to another person's mobile device?”

Psychopathology

The Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5)²⁷ crosscutting measure was used to assess psychopathology. Each question was preceded by the prompt “During the past 2 weeks, how much (or how often) have you been bothered by the following problems?” Content areas of depression (two items, e.g., “Little interest or pleasure in doing things?” α = 0.81), anxiety (two items, e.g., “Feeling panic or being frightened?” α = 0.83), and sleep problems (single item: “Problems with sleep that affected your sleep quality overall”) were examined. Each item was scored 0 (“None at all”) to 4 (“Nearly every day”). Scales were then summed.

Substance use

The prompt “In the past 12 months, how often in a typical week have you consumed the following substances?” preceded each substance type: alcohol, tobacco (e.g., “Cigarettes, a cigar, or pipe, or using snuff or chewing tobacco?”), marijuana, cocaine (“Cocaine [powder or crack]”), heroin (“Heroin [snorted/injected]”), club drugs (“ecstasy/MDMA”), painkillers (e.g., “Oxycontin, Vicodin, Codeine”), methamphetamine (“speed, crystal”), and sedative or tranquilizers (e.g., “Valium, Librium, Xanax, sleeping pills”). All substances were assessed on a scale from 0 “None at all” to 4 “Nearly every day.” Because of severe normality problems with multiple substance distributions in the full sample, each substance use indicator was recoded to 0 “no use in the past 12 months” and 1 “use in the past 12 months.”

Compulsive sexual behaviors

Compulsive sexual behaviors were assessed using the 13-item Compulsive Sexual Behavior Inventory (CSBI-13).²⁸ Example item: “How often have you had sex or masturbated more than you wanted to?” (α = 0.87) scored from 0 “Never” to 4 “Very frequently.”

Analytic plan

We first examined how sexting categories were associated with mental health. We did this by constructing a multivariate analysis of the variance (MANOVA) with sexting classification (No Sexting, Received Only, Sent Only, Reciprocal) entered as the independent variable and indicators for depression, anxiety, sleep problems, and compulsive sexual behaviors entered as dependent variables. We followed these analyses with Bonferroni-corrected post hoc tests (p = 0.05/6 = 0.008).

To examine the connection between sexting and substance use, we constructed a multinomial logistic regression with binary indicators for past 12-month alcohol, tobacco, marijuana, cocaine, heroin, methamphetamine, club drug, and sedative use entered as predictors of different sexting categories (Sent Only, Received Only, or Reciprocal, with No Sexting as the reference group). Odds ratios (ORs) were calculated, but only interpreted as significant if the accompanying 95 percent confidence interval (CI) did not contain “1.”

To contextualize our findings, we repeated our analyses using a multigroup framework. Specifically, we attempted to repeat our initial MANOVA design but split across participants who were heterosexual and male assigned at birth (MAB), heterosexual and female assigned at birth (FAB), nonheterosexual and MAB, and nonheterosexual and FAB. Multivariate statistics were chosen in congruence with statistical assumption patterns. Wilks's Λ was used in cases where all MANOVA assumptions were met, Pillai's Trace was used when homogeneity of covariance matrices was violated, and Roy's Largest Root was used when sample was below desired power (only in the case of nonheterosexual MAB participants). We then attempted to repeat the multinomial logistic regression using the multigroup framework. Effect sizes were interpreted using established thresholds.²⁹

Results

In the total sample, depression, anxiety, sleep problems, and compulsive sexual behavior distributions were all normal. Initial base rates suggested that 76.6 percent (n = 1,654) of the sample had engaged in some form of sexting. Only 2 percent (n = 44) had sent a sext only, 17.3 percent (n = 374) had received a sext only, and 57.2 percent (n = 1,236) had reciprocally sexted. The first MANOVA indicated a significant multivariate effect, Pillai's Trace = 0.09, F(12, 6465) = 16.75, p < 0.001, η² = 0.03. Followup univariate tests suggested significant effects for each variable (see Table 2 for all estimates). Participants in the “Sent Only” and “Reciprocal” groups reported higher depression means compared with those in the “No Sexting” group, and participants in the “Reciprocal” group reported higher depression means than those in the “Received Only” group.

Table 2.

Followup Univariate Results of the Relationship Between Sexting Categories and Mental Health Indicators

Indicators, M (SD)	No sexting (n = 506)	Sent only (n = 44)	Received only (n = 374)	Reciprocal (n = 1,236)	F	d1	d2	d3	d4	d5	d6
Depression	2.15 (2.02)	3.23 (2.47)	2.34 (2.00)	2.85 (2.21)	16.09	0.48	—	0.33	—	—	0.24
Anxiety	3.40 (3.01)	4.55 (3.80)	3.33 (2.92)	4.12 (3.26)	10.39	—	—	0.23	—	—	0.26
Sleep problems	1.06 (1.21)	1.48 (1.47)	1.41 (1.32)	1.44 (1.36)	10.87	—	0.28	0.30	—	—	—
Compulsive sexual behavior	19.26 (7.00)	23.50 (7.33)	21.45 (7.48)	24.49 (8.16)	58.18	0.59	0.30	0.69	—	—	0.39

Note: Cohen's d's were calculated for between-group means that evidenced Bonferroni-corrected p's < 0.008 in followup post hoc tests. Cohen's d1 = “No Sexting” compared with “Sent Only,” d2 = “No Sexting” compared with “Received Only,” d3 = “No Sexting” compared with “Reciprocal,” d4 = “Sent Only” compared with “Received Only,” d5 = “Sent Only” compared with “Reciprocal,” d6 = “Received Only” compared with “Reciprocal.” All F's were significant at p < 0.001.

Participants in the “Reciprocal” group reported significantly higher anxiety means than those in the “No Sexting” and “Received Only” groups. Participants in the “Reciprocal” and “Received Only” groups reported significantly more sleep problems than those in the “No Sexting” group. Participants in the “Sent Only,” “Received Only,” and “Reciprocal” groups all reported significantly more compulsive sexual behaviors compared with those in the “No Sexting” group, with participants in the “Reciprocal” group also reporting significantly more compulsive sexual behaviors than those in the “Received Only” group.

Results from the first multinomial regression suggested that Hessian matrix problems likely associated with low base rate activity of some substances (e.g., methamphetamine) and the small sample of “Sent Only” participants (n = 44). Accordingly, we adjusted our outcome variable by removing “Sent Only” participants and did not include cocaine, heroin, club drug, sedative, or methamphetamine indicators as predictors. The full results are available in Table 3. Marijuana use was significantly associated with an increase in the odds of being categorized in the “Received Only” (OR = 2.65, 95 percent CI: 1.77–3.98) and “Reciprocal” (OR = 4.02, 95percent CI: 2.84–5.68) categories compared with those in the “No Sexting” group. Multiple other substances evidenced large odds increases (Table 3) associated with sexting behaviors; however, the CIs contained 1 for all other comparisons.

Table 3.

Multinomial Regression Model Predicting the Sexting Group (No Sexting as Referent)

Predictor variables by group	B	SE	Wald	Adjusted ORs	95 percent CI
“Received Only” compared with “No Sexting”
Alcohol use	−0.04	0.78	0.00	0.97	0.21–4.48
Tobacco use	0.71	0.78	0.83	2.03	0.44–9.38
Marijuana use	0.98	0.21	22.11^***	2.65	1.77–3.98
Painkiller (i.e., opioid) use	−0.03	0.29	0.01	0.97	0.55–1.70
“Reciprocal” compared with “No Sexting”
Alcohol use	1.05	0.74	2.05	2.87	0.68–12.14
Tobacco use	0.23	0.73	0.10	1.26	0.30–5.30
Marijuana use	1.39	0.18	62.05^***	4.02	2.84–5.68
Painkiller (i.e., opioid) use	−0.23	0.24	0.91	0.79	0.50–1.28

Note: Participants in the “Sent Only” group were not included in this analysis due to low frequency. Each substance reflects use in the past year (coded 0 = No use, 1 = Use in the past year).

***

p < 0.001.

CI, confidence interval; OR, odds ratio; SE, standard error.

We proceeded with our primary aim of examining how the sexting–mental health relationship changes when considering sex and sexual identity. Given statistical power problems with the “Sent Only” group, we only compared outcomes across the other conditions. In each case, there was a significant multivariate effect: heterosexual MAB: Wilks's Λ = 0.93, F(8, 992) = 4.89, p < 0.001, η² = 0.04; heterosexual FAB: Pillai's Trace = 0.13, F(8, 1718) = 14.64, p < 0.001, η² = 0.06; nonheterosexual MAB: Roy's Largest Root = 0.10, F(4, 140) = 3.56, p = 0.012, η² = 0.09; and nonheterosexual FAB: Wilks's Λ = 0.91, F(8, 1186) = 7.42, p < 0.001, η² = 0.05.

Followup univariate and Bonferroni-corrected least significant difference post hoc tests are available in Table 4 (for heterosexual MAB/FAB participants) and Table 5 (for nonheterosexual MAB and FAB participants). Anxiety was the only variable that did not evince a significant univariate effect. Depression was significant but only for heterosexual MAB and FAB participants, with participants in the “Reciprocal” group reporting significantly higher depression scores than those in the “No Sexting” group. Sleep problems also only varied in the heterosexual MAB and FAB groups, with participants in the “Reciprocal” and “Received Only” groups reporting more sleep problems than non-sexting participants.

Table 4.

Followup Univariate Results of the Relationship Between Sexting Categories and Mental Health Indicators for Heterosexual Males and Females

Heterosexual males
Indicators, M (SD)	No sexting (n = 140)	Received only (n = 117)	Reciprocal (n = 246)	F	d1	d2	d3
Depression	1.84 (1.88)	2.04 (1.99)	2.44 (2.20)	4.09^*	—	0.29	—
Anxiety	2.48 (2.65)	2.69 (2.76)	3.07 (2.71)	2.29	—	—	—
Sleep problems	0.89 (1.06)	1.40 (1.35)	1.22 (1.27)	5.76^**	0.42	0.28	—
Compulsive sexual behavior	21.43 (7.87)	22.76 (7.93)	25.43 (8.02)	12.35^***	—	0.50	0.34

Heterosexual females
Indicators, M (SD)	No sexting (n = 229)	Received only (n = 145)	Reciprocal (n = 490)	F	d1	d2	d3
Depression	1.84 (1.73)	2.11 (1.87)	2.47 (2.01)	8.85^***	—	0.32	—
Anxiety	3.47 (3.02)	3.28 (2.70)	3.71 (3.13)	1.28	—	—	—
Sleep problems	0.97 (1.20)	1.35 (1.28)	1.27 (1.33)	5.25^**	0.31	0.24	—
Compulsive sexual behavior	17.25 (5.49)	19.40 (6.38)	22.71 (7.76)	50.58^***	0.36	0.81	0.47

Note: Cohen's d's were calculated for between-group means that evidenced Bonferroni-corrected p's < 0.02 in followup least significant difference post hoc tests. Cohen's d1 = “No Sexting” compared with “Received Only,” d2 = “No Sexting” compared with “Reciprocal,” d3 = “Received Only” compared with “Reciprocal.”

p < 0.05, ^**p < 0.01, ^***p < 0.001.

Table 5.

Followup Univariate Results of the Relationship Between Sexting Categories and Mental Health Indicators for Nonheterosexual Males and Females

Nonheterosexual males
Indicators, M (SD)	No sexting (n = 39)	Received only (n = 20)	Reciprocal (n = 86)	F	d1	d2	d3
Depression	2.59 (2.25)	2.35 (1.73)	2.84 (2.24)	0.49	—	—	—
Anxiety	2.95 (2.77)	2.85 (2.60)	4.21 (3.34)	2.99	—	—	—
Sleep Problems	1.05 (1.08)	1.35 (1.31)	1.55 (1.26)	2.27	—	—	—
Compulsive sexual behavior	23.82 (8.81)	26.15 (8.09)	28.84 (8.36)	4.84^**	—	0.59	—

Nonheterosexual females
Indicators, M (SD)	No sexting (n = 96)	Received only (n = 91)	Reciprocal (n = 412)	F	d1	d2	d3
Depression	3.10 (2.40)	3.07 (2.10)	3.54 (2.27)	2.60	—	—	—
Anxiety	4.70 (3.14)	4.35 (3.31)	5.22 (3.39)	3.01	—	—	—
Sleep problems	1.49 (1.40)	1.54 (1.34)	1.77 (1.41)	2.15	—	—	—
Compulsive sexual behavior	18.77 (5.98)	21.89 (7.58)	25.22 (8.16)	29.70^***	0.46	0.90	0.42

p < 0.05, ^**p < 0.01, ^***p < 0.001.

Finally, compulsive sexual behaviors significantly varied across all groups. For each comparison, participants in the “Reciprocal” group reported significantly more compulsive sexual behaviors that those in the “No Sexting” group. With exception to nonheterosexual MAB participants, compulsive sexual behaviors were also higher in the “Reciprocal” group compared with the “Received Only” group, and significantly higher in the “Received Only” group compared with the “No Sexting” group in heterosexual/nonheterosexual FAB participants (Tables 4 and 5).

We then attempted to repeat our multinomial logistic regression using a multigroup framework. However, due to cell size issues, this was not possible for all identity groups/sexting classifications. Instead, we adjusted our analyses by repeating the initial multinomial regression but included sex assigned at birth (0 = male, 1 = female) and sexual identity (heterosexual as referent) as covariates along with the previously stated substance use predictors results are available on Table 6. Marijuana use remained associated with a significant increase in the odds of being classified in the “Received Only” (adjusted OR = 2.66, 95 percent CI: 1.76–4.00) and “Reciprocal” (adjusted OR = 3.89, 95 percent CI: 2.74–5.53) groups compared with participants in the “No Sexting” group while adjusting for sex and sexual identity. No other substance use indicators were significant correlates (see Supplementary File S2 for note on additional models).

Table 6.

Multinomial Regression Model Predicting the Sexting Group (No Sexting as Referent)

Predictor variables by group	B	SE	Wald	Adjusted ORs	95 percent CI
“Received Only” compared with “No Sexting”
Alcohol use	−0.08	0.78	0.03	0.93	0.20–4.30
Tobacco use	0.72	0.78	0.86	2.06	0.45–9.52
Marijuana use	0.98	0.21	21.77^***	2.66	1.76–4.00
Painkiller (i.e., opioid) use	−0.03	0.29	0.01	0.97	0.55–1.71
Covariates
Sex assigned at birth (male referent)	−0.03	0.15	0.04	0.97	0.73–1.30
Gay/lesbian identity	1.18	0.54	4.77^*	3.27	1.13–9.45
Bisexual identity	0.04	0.18	0.04	1.04	0.73–1.47
Questioning identity	−0.08	0.45	0.04	0.92	0.38–2.20
Asexual identity	−0.40	0.61	0.44	0.67	0.20–2.20
Pansexual identity	−0.18	0.66	0.08	0.83	0.23–3.02
“Reciprocal” compared with “No Sexting”
Alcohol use	1.07	0.75	2.02	2.90	0.67–12.62
Tobacco use	0.17	0.75	0.05	1.19	0.27–5.14
Marijuana use	1.36	0.18	57.52^***	3.89	2.74–5.53
Painkiller (i.e., opioid) use	−0.31	0.25	1.61	0.73	0.45–1.19
Covariates
Sex assigned at birth (male referent)	0.36	0.13	7.85^**	1.43	1.11–1.83
Gay/lesbian identity	1.70	0.48	12.30^***	5.46	2.11–14.08
Bisexual identity	0.45	0.14	9.99^**	1.56	1.18–2.06
Questioning identity	−1.11	0.46	5.70^*	0.33	0.13–0.82
Asexual identity	−1.47	0.69	4.54^*	0.23	0.06–0.89
Pansexual identity	1.02	0.46	4.87^*	2.77	1.12–6.85

Note: Participants in the “Sent Only” group were not included in this analysis due to low frequency. Each substance reflects use in the past year (coded 0 = No use, 1 = Use in the past year). Heterosexual is the referent group for all sexual identities.

p < 0.05, ^**p < 0.01, ^***p < 0.001.

Discussion

Our results are consistent with the literature suggesting sexting to be common.^3,5 Indeed, 76.6 percent of participants reported engaging in some form of sexting (primarily reciprocal). Sexting was positively associated with depression, anxiety, sleep problems, and compulsive sexual behaviors. These associations were most pronounced when comparing those who had reciprocally sexted with those who had never sexted. Overall, effect sizes varied considerably. The significant differences between depression, anxiety, and sleep problems across sexting classifications were all small and rarely reached a meaningful order of magnitude.²⁹ Alternatively, effect sizes were much bigger when considering compulsive sexual behaviors.

Importantly, observation of the raw means suggested that participants who had only ever sent (but not received) sexts reported more depression, anxiety, and sleep problems than the other groups. However, due to the small sample of “Sent Only” participants, significant findings were difficult to detect. Sending, but not receiving, may be considered a special interest topic for future sexting/mental health research. It is quite possible that sending, but receiving, sexts is associated with lack on consent, impulsivity, and/or poor social skills. All of which would suggest a low base rate and high psychological distress (as was observed). However, more research is necessary to explore the specific reasons for these observations in more detail.

Generally, substance use was not significantly associated with sexting, with some exceptions/considerations. Marijuana use was consistently linked to sexting, for both received only and reciprocal sexting compared with non-sexting participants. The effect sizes were quite robust and consistent with past research.¹³ Illicit substance use, such as cocaine and methamphetamine use, was difficult to model given the low base rate. However, observation of the raw percentages suggested a positive association between illicit substance use and sexting (particularly for reciprocal and sent only sexting compared with non-sexting participants). For example, only one person in the entire sample had tried cocaine, yet not sexted, compared with 41 who had tried cocaine and had sexted. Similar patterns were evident for heroin, methamphetamine, sedative, and club drug use.

When considering sex and sexual identity, general mental health differences (depression, anxiety, and sleep problems) across sexting classifications remained weak. Particularly, no significant general mental health differences were evident across sexting classifications in nonheterosexual participants. Concurrently, substance use findings did not change when adjusted for sex and sexual identity. Marijuana remained the only stable correlate. Consistent with the minority stress literature,^30,31 nonheterosexual participants evidenced higher mental health problems compared with heterosexual participants; however, this pattern was evident regardless of sexting classification.

Compulsive sexual behaviors were highest in reciprocal sexting participants compared with non-sexting participants, with the effect sizes being robust across groups. Regarding psychological distress variables, this was the only stable finding across groups, with compulsive sexual behavior means demonstrating a linear pattern (lowest in non-sexting, followed by received only, and highest in reciprocal sexting participants) regardless of sex and sexual identity. The effect sizes between non-sexting and reciprocal sexting participants on compulsive sexual behaviors were also notably larger in women (0.81 and 0.90) compared with men (0.50 and 0.59) regardless of sexual identity. We suggest these finding to be relevant for future researchers, as they suggest that sexting to be a possible manifestation of compulsive sexual behavior.

Limitations

The cross-sectional design precluded any causal inferences. Longitudinal designs are recommended to address this problem. Our sample was diverse but was drawn from a college student population. We are uncertain about how our results may generalize. We had to use homogeneous groupings of sexual minority identities under the “nonheterosexual” umbrella. Past research has shown that psychological distress varies significantly across minoritized sexual identities,³¹ as such future research looking at how the mental health–sexting relationship changes across specific minoritized sexual identities is needed. Sexting was also assessed via lifetime histories. While this approach produced meaningful results, it also provides an incomplete picture.

It is likely that many participants sext more often than others. A myriad of additional contextual variables may likely also influence the degree to which sexting is connected to psychological distress and substance use, such as who the sext came from/went too, the sexting frequency, and quality of relationship between sexting partners (among many other potential important variables). It is also possible that a restriction of range was evident on the DSM-5 crosscutting items, explaining why the effect sizes were much smaller when compared with the CSBI-13. Finally, our study is limited by general biases common in self-report psychological literature, such as social desirability biases. Our results need to be interpreted with these limitations in mind.

Clinical implications

Clinicians should be mindful of the possible connection between sexting, marijuana use, and compulsive sexual behavior. While sexting may not necessarily be a causal factor of negative mental health outcomes or substance use, some meaningful covariation clearly exists. Clinicians are encouraged to explore sexting within the context of clients struggling with CSBD. Prior research has suggested that other sexual behaviors, such as pornography use, masturbation, and partnered sex, might become problematic when used to dysfunctionally regulate emotions, when the behaviors are non-consistent with values, and/or when such behaviors are employed in inappropriate contexts.^23,32,33

It is likely that sexting will eventually be included with these behaviors as a possible manifestation of CSBD. This is not to imply that sexting is compulsive sexuality, but rather that those with CSBD are at-risk of using sexting dysfunctionally. Clinicians should approach sexting (and other sex-related topics) with an open mind when such issues are relevant to presentation. That is, even in cases of CSBD, it may not be that sexting is the underlying problem, but rather its expression. Careful assessment should be performed to identify potential causes of CSBD behaviors. Sexting reduction should only be targeted in therapy if there is clear conceptual relevance to the presenting problem.

Footnotes

Acknowledgments

The authors would like to thank Amelia Evans (counseling psychology PhD student at Texas Tech University) and Danielle Haakinson (undergraduate research fellow at Texas Tech University) for their assistance with this project. Also note that C.L.M. and M.R.T are “second authors” with the current ordering being issued by alphabetical order.

Authors' Contributions

N.C.B. is responsible for study design, analysis, data gathering, article writing, and project administrations. C.L.M. and M.R.T. is responsible for literature review, article editing, and table construction. S.G. is responsible for article writing and revision. S.W.K. is responsible for data gathering and article revision.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

Support for S.W.K. was provided by Kindbridge Research Institute. The funding agencies did not provide input or comment on the content of the article, and the content of the article reflects the contributions and thoughts of the authors and does not necessarily reflect the views of the funding agencies. Article fees were paid by Texas Tech University via N.C.B.'s developmental funds.

Supplementary Material

References

Gassó

, Klettke

, Agustina

, et al. Sexting, mental health, and victimization among adolescents: A literature review. Int J Environ Res Public Heal, 2019; 16:2364; doi: 10.3390/IJERPH16132364

Gaylord

. Sex, cells, and sorna: Applying sex offender registration laws to sexting cases. William Mary Law Rev, 2011; 52:1717–1746.

Mori

, Cooke

, Temple

, et al. The prevalence of sexting behaviors among emerging adults: A meta-analysis. Arch Sex Behav, 2020; 49:1103–1119; doi: 10.1007/S10508-020-01656-4/TABLES/5

Ševčíková

. Girls' and boys' experience with teen sexting in early and late adolescence. J Adolesc, 2016; 51:156–162; doi: 10.1016/j.adolescence.2016.06.007

Stasko

, Geller

. Reframing Sexting as a Positive Relationship Behavior. In: Rethinking Women's Sexual and Psychosocial Health through Novel Technology, Access, and Programming. (Geller PA, ed.). 123rd Annual Convention of the American Psychological Association; 2015. Available from: https://psycnet.apa.org/record/528002015-001?doi=1 [Last accessed: August 11, 2022].

Handschuh

, La Cross

, Smaldone

. Is sexting associated with sexual behaviors during adolescence? A systematic literature review and meta-analysis. J Midwifery Womens Health, 2019; 64:88–97; doi: 10.1111/jmwh.12923

Kosenko

, Luurs

, Binder

. Sexting and sexual behavior, 2011–2015: A critical review and meta-analysis of a growing literature. J Comput Commun, 2017; 22:141–160; doi: 10.1111/jcc4.12187

Mori

, Choi

, Temple

, et al. Patterns of sexting and sexual behaviors in youth: A latent class analysis. J Adolesc, 2021; 88:97–106; doi: 10.1016/j.adolescence.2021.01.010

Van Ouytsel

, Lu

, Ponnet

, et al. Longitudinal associations between sexting, cyberbullying, and bullying among adolescents: Cross-lagged panel analysis. J Adolesc, 2019; 73:36–41; doi: 10.1016/j.adolescence.2019.03.008

10.

Gámez-Guadix

, Mateos-Pérez

, Wachs

, et al. Assessing image-based sexual abuse: Measurement, prevalence, and temporal stability of sextortion and nonconsensual sexting (“revenge porn”) among adolescents. J Adolesc, 2022; 94:789–799; doi: 10.1002/jad.12064

11.

Gabriel

. Sexting, selfies and self-harm: Young people, social media and the performance of self-development. Media Int Aust, 2014; 151:104–112; doi: 10.1177/1329878X1415100114

12.

Jules

, Maynard

, Coulson

. Emerging adults' susceptibility to sexting influences on Facebook: Dangerous or developmental?. Int Psychol Bull, 2017; 21:35–43.

13.

Mori

, Temple

, Browne

, et al. Association of sexting with sexual behaviors and mental health among adolescents: A systematic review and meta-analysis. JAMA Pediatr, 2019; 173:770–779; doi: 10.1001/jamapediatrics.2019.1658

14.

Klettke

, Hallford

, Mellor

. Sexting prevalence and correlates: A systematic literature review. Clin Psychol Rev, 2014; 34:44–53; doi: 10.1016/j.cpr.2013.10.007

15.

Gordon-Messer

, Bauermeister

, Grodzinski

, et al. Sexting among young adults. J Adolesc Heal, 2013; 52:301–306; doi: 10.1016/j.jadohealth.2012.05.013

16.

Englander

Low risk associated with most teenage sexting: A study of 617 18-year-olds. MARC Res Reports. Published online July 1, 2012. Available from: https://vc.bridgew.edu/marc_reports/6 [Last accessed: August 12, 2022].

17.

Frankel

, Bass

, Patterson

, et al. Sexting, risk behavior, and mental health in adolescents: An examination of 2015 Pennsylvania Youth Risk Behavior Survey Data. J Sch Health, 2018; 88:190–199; doi: 10.1111/JOSH.12596

18.

Temple

, Le

, van den Berg

, et al. Brief report: Teen sexting and psychosocial health. J Adolesc, 2014; 37:33–36; doi: 10.1016/j.adolescence.2013.10.008

19.

Valiukas

, Pickering

, Hall

, et al. Sexting and mental health among young Australians attending a musical festival: A cross sext-ional study. Cyberpsychol Behav Soc Netw, 2019; 22:521–528; doi: 10.1089/cyber.2018.0671

20.

Benotsch

, Snipes

, Martin

, et al. Sexting, substance use, and sexual risk behavior in young adults. J Adolesc Heal, 2013; 52:307–313; doi: 10.1016/j.jadohealth.2012.06.011

21.

Dir

, Riley

, Cyders

, et al. Problematic alcohol use and sexting as risk factors for sexual assault among college women. J Am Coll Heal, 2018; 66:553–560; doi: 10.1080/07448481.2018.1432622

22.

Florimbio

, Brem

, Grigorian

, et al. An examination of sexting, sexual violence, and alcohol use among men arrested for domestic violence. Arch Sex Behav, 2019; 48:2381–2387; doi: 10.1007/S10508-019-1409-6/TABLES/4

23.

International Statistical Classification of Diseases and Related Health Problems. 11th ed. World Health Organization; 2022. Available from: https://icd.who.int [Last accessed: December 23, 2022].

24.

Clancy

, Howard

, Chong

, et al. Dream It, Do It? Associations between pornography use, risky sexual behaviour, sexual preoccupation and sexting behaviours among young Australian adults. Sexes, 2021; 2:433–444; doi: 10.3390/sexes2040034

25.

Weiss

, Samenow

. Smart phones, social networking, sexting and problematic sexual behaviors—A call for research. Sex Addict Compulsivity, 2010; 17:241–246; doi: 10.1080/10720162.2010.532079

26.

Courtice

, Shaughnessy

. Four problems in sexting research and their solutions. Sexes, 2021; 2:415–432; doi: 10.3390/sexes2040033

27.

Narrow

, Clarke

, Kuramoto

, et al. DSM-5 field trials in the United States and Canada, Part III: Development and reliability testing of a cross-cutting symptom assessment for DSM-5. Am J Psychiatry, 2013; 170:71–82.

28.

Coleman

, Swinburne Romine

, Dickenson

, et al. Compulsive Sexual Behavior Inventory-13. In: Handbook of Sexuality-Related Measures, 4th ed. (Milhausen RR, Skakaluk JK, Fisher TD, Davis CM, Yarber WL, eds.) Routledge: Oxfordshire; 2019.

29.

Ferguson

. An effect size primer: A guide for clinicians and researchers. Prof Psychol Res Pract, 2009; 40:532–538; doi: 10.1037/a0015808

30.

Meyer

. Prejudice, social stress, and mental health in lesbian, gay, and bisexual populations: Conceptual issues and research evidence. Psychol Bull, 2003; 129:674–697; doi: 10.1037/0033-2909.129.5.674

31.

Borgogna

, McDermott

, Aita

, et al. Anxiety and depression across gender and sexual minorities: Implications for transgender, gender nonconforming, pansexual, demisexual, asexual, queer, and questioning individuals. Psychol Sex Orientat Gend Divers, 2019; 6:54–63; doi: 10.1037/sgd0000306

32.

Kor

, Zilcha-Mano

, Fogel

, et al. Psychometric development of the Problematic Pornography Use Scale. Addict Behav, 2014; 39:861–868; doi: 10.1016/j.addbeh.2014.01.027

33.

Grubbs

, Hoagland

, Lee

, et al. Sexual addiction 25 years on: A systematic review of empirical literature on compulsive sexual behavior and an agenda for future research. Clin Psychol Rev, 2020; 82:101925; doi: 10.1016/j.cpr.2020.101925

Supplementary Material

Please find the following supplemental material available below.

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

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

0.00 MB

0.02 MB

0.01 MB