Transactional sex—the exchange of sexual acts for material benefits—poses risks to well-being among women students globally, and specifically in China (Sanders, 2008). While prior research emphasizes macroeconomic drivers like poverty, it overlooks individual cognitive mechanisms, specifically the role of materialistic money attitudes—the tendency to view money as a symbol of success and a means to enhance social status—in framing sex as an instrumental resource (Sanders, 2008). Therefore, this study investigated the potential pathways through which permissive sexual scripts—cultural norms that approve of casual sex outside of committed relationships—influence transactional sex through the mediation of materialistic money attitudes. Elucidating this scripts–attitudes–behavior pathway will advance theoretical understanding and inform targeted interventions in China’s conflict between consumerism and tradition.

 

This study employed the term “transactional sex” rather than “sex work,” as the latter denotes an institutionalized occupation (Weitzer, 2009), whereas our focus was on non-professional, situational exchanges among students (Sanders, 2008). “Transactional sex” more accurately captures this informal, event-specific interaction (Wamoyi et al., 2016). Terminology shapes perception (Hansen & Johansson, 2023), making “transactional sex” a more neutral and precise choice for examining the behavior and its cognitive antecedents.

 

Prior research on transactional sex and sex work has highlighted structural factors such as poverty (Sanders, 2008) and cognitive factors such as permissive sexual scripts (Gagnon & Simon, 1973). However, these two lines of inquiry have largely existed in parallel, leaving the interplay between structural and cognitive mechanisms underexplored. While general materialism has been linked to sexual risk taking (Zurndorfer, 2016), critical gaps remain. Further, materialism has rarely been examined as a specific money attitude driving transactional sex, and the theoretical link between permissive scripts and this attitude is unexplored. Consequently, the literature lacks a direct test of materialistic money attitudes as the core motivational schema channeling scripts into this specific behavior.

 

To explain why permissive sexual scripts might lead to transactional sex through materialistic money attitudes, we proposed a sequential scripts–attitudes–behavior pathway. First, the cognitive blueprint of permissive sexual scripts legitimize an instrumental view of sexuality, framing it as a potential resource for personal gain (Gagnon & Simon, 1973). This creates a generalized predisposition toward viewing intimate relationships in transactional terms. Second, within the potent consumerist context of contemporary China, individuals who already hold such scripts are more psychologically receptive to internalizing broader materialistic narratives (Bandura, 2001), thereby developing stronger materialistic attitudes regarding money (Chen, 2016). Finally, these internalized materialistic money attitudes become the proximal motivational driver. They supply the specific currency that makes the instrumental exchange of sex into a subjectively rational and desirable option. Thus, we proposed the following hypotheses:

Hypothesis 1: The case group will score higher than the control group on permissive scripts, materialistic money attitudes, and impulsivity.

Hypothesis 2: Permissive scripts, materialistic money attitudes, and impulsivity will positively correlate with transactional sex.

Hypothesis 3: Materialistic money attitudes will mediate the scripts–behavior relationship, after we control for impulsivity as a covariate.

We conducted a field-based survey across 10 universities in Shandong Province, China, selected to ensure diversity in institutional tiers (e.g., “Project 985,” “Project 211,” and provincial universities) and academic disciplines (e.g., comprehensive, science and engineering, teacher training), enhancing the potential representativeness of women college students in similar regional and educational contexts within China. We distributed 5,000 questionnaires (500 per university). After excluding invalid responses (e.g., incomplete, patterned answering), we obtained 4,810 valid questionnaires from women college students.

 

In terms of screening and case validation, all participants completed the same paper-based questionnaire. The questionnaire first included a dichotomous screening question: “Have you ever engaged in sexual acts in exchange for money or expensive gifts?” We formed a preliminary case group comprising the 228 (4.74%) participants who responded “yes” to this screening item. To verify the authenticity of these reports while preserving anonymity, the questionnaire design immediately required this group of participants to provide specific details about this behavior—including frequency, context, and the nature of compensation—within the same questionnaire. The research team then manually reviewed all written descriptions provided by these participants against predefined validity criteria. Valid responses had to clearly describe a specific context (e.g., through online platforms, through introduction by acquaintances) and specify a form of compensation with clear monetary value (e.g., cash, luxury goods, payment of bills or rent). After review, eight responses were excluded for being too vague (e.g., “received some gifts”), contradictory, or stereotypical. Ultimately, 220 responses were validated as confirmed cases. This process, which relied on detailed descriptions embedded within a single questionnaire and manual review, strengthened the validity of case group identification. However, it is important to note that the control group (those who answered “no” to the screening question) may have included individuals who engaged in similar behaviors but chose not to report them, which highlights a limitation inherent to self-report studies of sensitive behaviors.

 

We then individually matched each of the 220 validated cases (not including the eight excluded cases) with a control participant selected from the remaining 4,582 participants who answered “no” to the screening item. Matching was based on age (within ± 1 year), academic major, and year of study. We successfully matched 216 cases, resulting in a final case-control analytic sample of 432 participants (216 matched pairs). All subsequent hypothesis testing and mediation analyses were conducted exclusively on this matched sample of 432.

 

This research adhered to the ethical principles of the Declaration of Helsinki. The protocol was approved by the Institutional Review Board (IRB-2023-066) of Ludong University, China. Informed consent ensured participants were aware of the study’s purpose, procedures, risks, and their right to withdraw. To ensure confidentiality, the study was anonymous, collecting no personally identifiable information (e.g., names, student identifications). All encrypted data were stored on a secure, password-protected server with access restricted to the principal investigator. Upon completion, all participants were provided with contact information for psychological support services.

For all multi-item scales, the mean score for each scale was computed and used in subsequent analyses. This approach facilitated a uniform interpretation across measures with differing numbers of items.

 

As described above, a dichotomous screening question (i.e., “Have you ever engaged in sexual acts in exchange for money or expensive gifts?”) was used for primary group classification and subsequent statistical modeling. Participants who answered “yes” also completed supplementary questions regarding the frequency, context, and nature of compensation in a separate section. These detailed behavioral data were used solely for the purpose of validating case group membership and ensuring data integrity during the two-stage verification process. They were not utilized as variables in the primary mediation analysis, in which we focused on testing the psychological model.

 

To capture the multifaceted nature of permissive sexual scripts, a composite score was created by integrating two sets of scale items derived from established measures of permissive sexual attitudes. These were Hendrick and Hendrick’s (1987) scale (e.g., “Casual sex between consenting adults is acceptable without emotional commitment”) and Simpson and Gangestad’s (1991) scale (e.g., “Sex can be a fair exchange for gifts, money, or other material benefits”). We selected and adapted these items to tap into distinct yet complementary facets of a broader permissive sexual script construct relevant to the contemporary Chinese context. The raw total scores from both scales were first converted to z-scores (standardized within the current sample) to eliminate scale differences, and were then averaged to form a single composite score. We conducted a confirmatory factor analysis on the combined items and this supported unidimensionality, with all fit indices meeting conventional standards for good fit, χ²/df < 3.00, comparative fit index > .90, root-mean-square error of approximation < .08, standardized root-mean-square residual < .08 (Hu & Bentler, 1999). This composite indicator demonstrated adequate psychometric validity. It ranged from 1 to 5, with higher scores indicating stronger endorsement of permissive and instrumental sexual scripts.

 

We measured materialistic money attitudes using the Material Values Scale developed by Richins and Dawson (1992). This scale consists of 18 items measuring three dimensions: acquisition centrality (e.g., “I admire people who own expensive homes, cars, and clothes”), happiness (e.g., “I would be happier if I could afford to buy more things”), and success (e.g., “I respect people who are successful financially”). Participants rated each item on a 5-point Likert scale ranging from 1 = strongly disagree to 5 = strongly agree, with higher scores indicating stronger materialistic attitudes toward money. The mean score of the overall scale was calculated by averaging all items, which showed high reliability in the current sample (α = .88, M = 4.05, SD = 0.88). The original scale validation reported strong psychometric properties, with a confirmatory factor analysis supporting its three-factor structure (Richins & Dawson, 1992).

 

We measured impulsivity using the widely adopted Chinese revised version of the Barratt Impulsiveness Scale (Wang et al., 2018). This scale has 30 items rated on a 4-point Likert scale ranging from 1 = rarely/never to 4 = almost always/always, with higher scores indicating greater trait impulsivity. The total mean score across all items was used in the analysis. The scale demonstrated good internal consistency of .88 in this study and Wang et al. (2018) previously reported similar reliability (α = .87).

We analyzed data using SPSS 26.0. We conducted preliminary analyses (i.e., independent-sample t tests, Pearson correlations) to examine group differences and bivariate relationships (Hypotheses 1 and 2). Further, to test the hypothesized mediation (Hypothesis 3), we employed a two-step analytical strategy (preliminary mediation analysis and primary mixed-effects mediation analysis) to ensure methodological rigor, given our matched case-control design. First, for the preliminary mediation analysis, we conducted a standard logistic regression-based mediation analysis using Model 4 of the PROCESS macro for SPSS 4.0 (Hayes, 2018) with 5,000 bootstrapped resamples. This preliminary analysis confirmed the basic mediation model where materialistic money attitudes mediated the relationship between permissive sexual scripts and transactional sex. Second, a primary (formal) mixed-effects mediation analysis crucial for accounting for the nonindependence of observations within the 216 matched pairs—a core feature of our design—and the primary test of Hypothesis 3 was conducted using generalized linear mixed-effects models. Specifically, we performed a multilevel mediation analysis using 5,000 Monte Carlo simulations to estimate the 95% confidence interval (CI) of the indirect effect. The models included a random intercept for each matched pair (Pair ID) to control for shared variance within pairs. A linear mixed model was specified for the continuous mediator (i.e., materialistic money attitudes), and a mixed-effects logistic regression model (melogit) was specified for the binary outcome (i.e., transactional sex). In all mediation models, we controlled for age, academic year, monthly disposable income, parental educational attainment, family location (i.e., urban/rural), and trait impulsivity as fixed effects to isolate the specific indirect effect. Mediation was considered statistically significant if the Monte Carlo CI for the indirect effect did not include zero.

First, we conducted preliminary analyses to examine data quality. To assess the potential for common method bias, we used Harman’s single-factor test. The results indicated that the first factor explained 32.54% of the total variance, which was below the commonly recommended threshold of 40%, suggesting that severe common method variance was not a pervasive issue in the dataset.

 

Independent-sample t tests were conducted to test Hypothesis 1. As presented in Table 1, the transactional sex group reported significantly higher levels of permissive sexual scripts, materialistic money attitudes, and impulsivity. The differences were all statistically significant with medium-to-large effect sizes, with all p values < .001, and Cohen’s d ranging from 0.67 to 0.79.

We conducted Pearson correlation analyses for the total sample (N = 432) to address Hypothesis 2. The correlation matrix is displayed in Table 2. As predicted, permissive sexual scripts showed a significant positive correlation with transactional sex behavior, r = .38, p < .001. Materialistic money attitudes, r = .42, p < .001, and impulsivity, r = .32, p < .001, were also significantly and positively correlated with transactional sex behavior. Furthermore, all three predictor variables were significantly intercorrelated.

We formally tested Hypothesis 3 using the mixed-effects mediation model described above, which accounted for the paired design. The results supported the hypothesized mediation. The total effect of permissive sexual scripts on transactional sex was significant, b = 0.85, p < .001 (for detailed path coefficients and effect decomposition, see Tables 3 and 4). Path analysis from the mixed-effects model indicated that permissive sexual scripts significantly predicted materialistic money attitudes, which, in turn, were associated with a higher likelihood of transactional sex (Figure 1). A significant indirect effect was found, b = 0.30, Boot 95% CI [0.15, 0.47], supported by a bootstrapped CI that did not include zero. Furthermore, the direct effect was no longer statistically significant at the conventional alpha level of .05, p = .06. These findings suggested that materialistic money attitudes largely accounted for the relationship between permissive sexual scripts and transactional sex in this model.

 

In all analyses for testing Hypothesis 3, we controlled for a comprehensive set of covariates, including age, academic year (as a proxy for educational progression), monthly disposable income, parental educational attainment (as an indicator of socioeconomic background), family location (urban/rural), and the trait of impulsivity. This approach was taken to reduce potential confounding, and to isolate the specific indirect effect through materialistic money attitudes.

Our principal contribution is the empirical delineation of a scripts–attitudes–behavior mediation model, where materialistic money attitudes serve as the pivotal mediator. This model offers a process-oriented explanation, specifying the cognitive–motivational sequence beyond broad attitudinal or macroeconomic factors alone (Zurndorfer, 2016).

 

Permissive sexual scripts legitimize an instrumental view of sex (Gagnon & Simon, 1973). Our mediation analysis showed that this framework operates indirectly, first fostering stronger materialistic money attitudes—a schema equating financial gain with success and self-worth (Chen, 2016). These internalized attitudes then act as the proximal driver, translating a permissive orientation into a concrete propensity for transactional exchange.

 

This pathway remained robust after we controlled for impulsivity and socioeconomic factors, indicating the behavior was not sufficiently explained by poor self-control or macrostructural determinants (e.g., poverty) alone. While impulsivity may lower the threshold for risk taking (Gottfredson & Hirschi, 1990), it is the mediated pathway, through which permissive scripts promoted materialistic attitudes, that provided the behavior with its specific motivational direction and force, thus nuancing general theories of risk behavior.

 

This process must be contextualized within contemporary China, where individuals may internalize conflicting narratives. In this context, permissive scripts generally circulate via digital and peer networks, while a potent consumerist ethos equates material possession with success. For some, transactional sex has been found to become a pragmatic, if risky, strategy to navigate these pressures and social mobility demands (Sanders, 2008).

Notwithstanding limitations of the study, we robustly identified materialistic money attitudes as a key mediator. Interventions should therefore move beyond generic risk awareness to disrupt the link between instrumental scripts and materialistic values. Theoretically informed programs integrating critical media literacy with values clarification could target this core mechanism to mitigate associated risks. For example, a values clarification exercise could ask students to rank personal goals (e.g., financial success, safety, relationships) and then reflect on how transactional sex might conflict with higher-ranked values. This could be facilitated through group discussion.

Limitations include the cross-sectional design of the study, regional sampling, and dichotomous outcome measure. Future research should employ nuanced measures of frequency, compensation, and reciprocity. Further, while our mediation model suggested a theoretically coherent sequence and controls for key confounds, the cross-sectional design precluded causal inference. Longitudinal or experimental designs are needed to confirm temporal precedence and causality.

In conclusion, this study illuminated the complex landscape wherein Chinese women students negotiate values. Transactional sex, from this perspective, is a situated behavior arising where sexuality and materialism intersect. Effective intervention requires engaging with this framework at its roots.

Bandura, A. (2001). Social cognitive theory: An agentic perspective. Annual Review of Psychology, 52(1), 1–26. https://doi.org/10.1146/annurev.psych.52.1.1

 

Chen, C. (2016). No moral qualms about transactional sex? Materialism and objectification. Social Behavior and Personality: An international journal, 44(11), 1803–1814. https://doi.org/10.2224/sbp.2016.44.11.1803

 

Gagnon, J. H., & Simon, W. (1973). Sexual conduct: The social sources of human sexuality. Aldine Publishing Company.

 

Gottfredson, M. R., & Hirschi, T. (1990). A general theory of crime. Stanford University Press. https://doi.org/10.1515/9781503621794

 

Hansen, M. A., & Johansson, I. (2023). Asking about “prostitution”, “sex work” and “transactional sex”: Question wording and attitudes toward trading sexual services. The Journal of Sex Research, 60(1), 153–164. https://doi.org/10.1080/00224499.2022.2130859

 

Hayes, A. F. (2018). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach (2nd ed.). Guilford Press.

 

Hendrick, S., & Hendrick, C. (1987). Multidimensionality of sexual attitudes. The Journal of Sex Research, 23(4), 502–526. https://doi.org/10.1080/00224498709551387

 

Hu, L.-T., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6(1), 1–55. https://doi.org/10.1080/10705519909540118

 

Richins, M. L., & Dawson, S. (1992). A consumer values orientation for materialism and its measurement: Scale development and validation. Journal of Consumer Research, 19(3), 303–316. https://doi.org/10.1086/209304

 

Sanders, T. (2008). Paying for pleasure: Men who buy sex. Willan Publishing.

 

Simpson, J. A., & Gangestad, S. W. (1991). Individual differences in sociosexuality: Evidence for convergent and discriminant validity. Journal of Personality and Social Psychology, 60(6), 870–883. https://doi.org/10.1037/0022-3514.60.6.870

 

Wamoyi, J., Stobeanau, K., Bobrova, N., Abramsky, T., & Watts, C. (2016). Transactional sex and risk for HIV infection in sub-Saharan Africa: A systematic review and meta-analysis. Journal of the International AIDS Society, 19(1), Article 20992. https://doi.org/10.7448/IAS.19.1.20992

 

Wang, M., Deng, Q., Shou, Y., Lai, H., Deng, J., Gao, Y., & Li, Z. (2018). Assessing impulsivity in Chinese: Elaborating validity of BIS among male prisoners. Criminal Justice and Behavior, 46(3), 492–506. https://doi.org/10.1177/0093854818806033

 

Weitzer, R. (2009). Sociology of sex work. Annual Review of Sociology, 35(1), 213–234. https://doi.org/10.1146/annurev-soc-070308-120025

 

Zurndorfer, H. (2016). Men, women, money, and morality: The development of China’s sexual economy. Feminist Economics, 22(2), 1–23. https://doi.org/10.1080/13545701.2015.1026834

Dafei Feng, Weihai Vocational College, Science and Technology New City, Weihai City, Shandong Province 264210, People’s Republic of China. Email: [email protected]