A stronger interpretation of the Molyko data begins with the Cameroonian literature. University students in Cameroon have long been studied as a population in which emerging-adult relationships, sexual decision-making, and the transition to relative independence intersect. In one of the earliest university-based studies from Buea, Kongnyuy and colleagues surveyed 664 students and found that awareness of emergency contraception was considerably wider than accurate knowledge; the authors concluded that misinformation remained a practical barrier to informed reproductive-health decisions [10]. That study is relevant beyond contraception because it illustrates a recurring feature of student life: intimate decisions are often made within peer networks where information, advice, and norms circulate rapidly, but not always accurately.

Evidence from another Cameroonian university points to the same need for contextual rather than moral interpretation. Noubiap and colleagues studied 411 students attending the University of Maroua medical and social welfare centre and documented high levels of sexual activity, multiple partnerships, pornography exposure, and inconsistent condom use among portions of the sample [11]. The authors framed these patterns principally as matters for HIV and STI prevention. Their findings are useful here because several behaviours measured in the present dating scale—multiple partnerships, pornography exposure, and sexual decision-making—were already being examined elsewhere in Cameroon as public-health and behavioural issues rather than simply as breaches of morality.

More recent studies show that intimate and reproductive concerns remain salient in Cameroonian higher education. At the University of Buea, Halle-Ekane and colleagues found that 42% of 405 sexually active female students were at risk of at least one domain of female sexual dysfunction; lower year of study, poor health, and a history of sexual assault were independently associated with greater risk, while being unmarried but in a relationship was associated with lower risk in their analysis [12]. At the University of Yaoundé I, Akoku and colleagues analysed data from 424 sexually active female students and reported that only about half correctly identified their fertile period, despite substantial contraceptive use [13]. These studies do not test the same constructs as the present research, but together they reinforce a central point: relationship and sexual experiences among Cameroonian university students involve both resources and vulnerabilities, and deserve evidence-based counselling rather than blanket condemnation.

Cameroonian scholarship on emerging adulthood also helps position dating as part of a developmental pathway. Lo-oh’s work with university students in Buea on marriage readiness emphasises that young adults actively construct ideas about intimacy, partnership, commitment, and future family life within particular social and cultural settings [14][15]. This perspective is compatible with Arnett’s account of emerging adulthood and with the present study’s emphasis on university neighbourhoods as spaces where autonomy expands while social controls and expectations are renegotiated [2]. In such settings, the relevant question is not whether relationships occur, but how young adults manage intimacy alongside academic obligations, finances, peer expectations, family expectations, and personal safety.

Comparable evidence from Nigeria and Ghana widens the regional frame. Studies of Nigerian undergraduates have documented dating violence, perceived interpersonal risks, cohabitation, risky sexual behaviour, and the academic and psychosocial consequences that can accompany unhealthy relationships [16]-[18]. Recent Nigerian work on cohabitation identifies emotional companionship, economic convenience, proximity to campus, peer influence, and housing or financial pressures as important motivations, alongside stigma and reproductive-health concerns [19][20]. In Ghana, qualitative work with university students similarly shows that consent, gender expectations, danger, and negotiation are embedded within wider social norms and power relations [21]. These studies caution against treating relationship behaviour as an isolated individual choice.

The wider sub-Saharan evidence is also uneven. A recent scoping review of sexual violence involving university students identified research from only a minority of countries in the region, with Nigeria, South Africa, and Ethiopia dominating the evidence base; Cameroon appeared far less frequently [22]. That geographical imbalance strengthens the case for publishing carefully qualified Cameroonian evidence, including older baseline data, provided that its temporal limits are explicit. A baseline from Molyko cannot establish current prevalence, but it can document what was already present in the ecology of a major university neighbourhood before the most recent expansion of smartphone-based relationship practices.

Much of the literature on student sexuality and relationships uses the university campus as the unit of analysis. The present study deliberately looked beyond the campus boundary. In Buea, as in many African university towns, a large part of student life occurs in privately rented hostels, rooms, bars, eateries, churches, cybercafés, transport points, and other spaces that sit outside direct university administration. The term university neighbourhood is therefore not merely geographical. It describes a social environment in which students exercise greater autonomy, interact intensively with peers, negotiate living costs, and build relationships while remaining connected to academic institutions.

This distinction matters for interpretation. A student may experience relationship conflict in a hostel, obtain advice from friends, borrow money from a partner, miss a morning lecture after a late night, seek reproductive-health information from a pharmacy, and later interact with a university counsellor. No single institution controls that chain of events. The neighbourhood perspective therefore complements individual psychological explanations by drawing attention to the settings through which behaviour is organised. It also helps explain why interventions confined to classroom rules may have limited reach. For international readers, Molyko is best understood as an example of a wider phenomenon: the dense student districts that grow around expanding universities and become important sites of emerging-adult socialisation.

The purpose of the study was to examine the extent to which student dating practices predicted discipline problems among university students living in a Cameroonian university neighbourhood. Reframed for publication in 2026, the study also serves a second purpose: to provide baseline evidence for contemporary conversations about student relationships, digital intimacy, consent, counselling, safeguarding, and university-community responsibility.

One research question guided the study: To what extent did dating practices predict students’ discipline problems in the Molyko university neighbourhood in 2016? The null hypothesis stated that dating practices would not significantly predict students’ discipline problems. The alternative hypothesis stated that dating practices would significantly predict students’ discipline problems.

The study is best read through three complementary theoretical lenses. Erikson’s [3] psychosocial theory positions late adolescence and early adulthood as periods of identity formation, during which intimacy, belonging, and self-definition become central. Bandura’s [23] social learning theory explains how students may observe, imitate, and normalize behaviours modelled by peers, older students, media figures, and romantic partners. Harris’s [24] group socialization theory adds that peer groups can become powerful transmitters of norms when young people live among age-mates away from close family supervision. Together, these perspectives help explain why dating practices in university neighbourhoods may become linked to discipline problems: behaviour is not only individually chosen; it is learned, rewarded, justified, and repeated within a peer environment.

These theories are complemented by a socio-ecological reading of student relationships. Individual developmental needs are expressed within peer groups, intimate partnerships, hostels, universities, families, neighbourhood businesses, and wider cultural expectations. This layered approach is particularly useful in African campus settings because risks do not arise solely from personal characteristics. Housing arrangements, financial pressure, limited privacy, gendered power, alcohol-serving venues, peer norms, and the accessibility of counselling or sexual and reproductive health services can all shape how relationships are experienced. Recent sub-Saharan reviews of campus sexual violence similarly organise risk and protective factors across individual, interpersonal, institutional, and social levels [22].

This article reports one component of a larger study on youth culture and discipline problems. The design combined a cross-sectional survey with a small number of post-survey interviews, in the general spirit of combining different forms of evidence to strengthen interpretation [25]. Because the interview record available for this article consists of researcher-generated summaries rather than an auditable transcript-and-coding trail, the interview material is presented throughout as contextual participant accounts that help interpret the survey findings, not as a convergent mixed-methods strand (see Sections 2.6 and 3.2). A survey was appropriate because the study required standardized responses from a relatively large number of students at one point in time. Interviews were added so that students could describe, in their own terms, how dating experiences entered their academic and neighbourhood lives.

The study was conducted in 2016 in Molyko, Buea, a neighbourhood adjoining the University of Buea. The accessible population was estimated at approximately 19,599 students, based on administrative student data from the Ministry of Higher Education (MINESUP). Using the Krejcie and Morgan [26] sample-size guide for a population close to 20,000, the target sample was 377, and 377 questionnaires were distributed. Of these, 323 were returned in usable form (85.7%); the remaining 54 questionnaires were either not returned by respondents or were returned damaged or soiled to the point of being unreadable, particularly those completed in bars and nightclubs. No count of students who were approached but declined to take a questionnaire was recorded in 2016, and route-specific distribution counts (hostels versus leisure venues) were not logged; these limits of the archived recruitment record are acknowledged in Section 6.

The final survey sample comprised 195 male students (60.4%) and 128 female students (39.6%). Age was recorded in bands: 67 participants (20.7%) were aged 16 - 20 years, 173 (53.6%) were aged 21 - 25 years, and 83 (25.7%) were aged 26 - 30 years. Because age was collected only in these bands, the exact number of participants younger than 18, if any, cannot be determined from the archived record; the implications of this are addressed in the Ethics statement and in Section 6. Respondents spanned seven faculties of the University of Buea and were in their second to fourth years of study. Most participants lived in crowded hostels (62.5%), and many reported living in hostels without caretakers (75.5%). A majority rated their financial status as low (62.5%). Religious identification was also recorded: 82.7% identified as Christian, 3.1% as Muslim, and 14.2% as belonging to other faiths.

Twenty students were interviewed after the survey. Interviewees were selected on a convenience basis from among survey respondents who indicated willingness to discuss sensitive relationship experiences in greater depth and who were available during the interview window (10 March to 10 April 2016).

Three non-probability sampling approaches were combined. Purposive sampling was used to recruit students living in hostels and social spaces close to bars, nightclubs, casinos, and other leisure sites where the behaviours under study were likely to be visible. Snowball sampling helped reach students who were less willing to disclose sensitive dating-related experiences to an unfamiliar researcher. Convenience sampling was used for the interviews because participants had to be available and willing to discuss sensitive experiences in depth. These choices increased access to relevant cases but limited the generalizability of the findings beyond the Molyko setting.

The quantitative instrument was a 70-item structured questionnaire developed for the larger study. It included items on demographics, alcohol use, smoking, dressing habits, dating experiences, gambling, and discipline problems. This article uses the demographic section, the dating-practices scale, and the discipline-problems scale.

The dating-practices scale comprised 10 items covering cohabitation without parental knowledge or approval, frequent partner communication through Facebook or WhatsApp, aggression related to partner stress, multiple sexual relationships, online dating with strangers, pornography exposure, fighting with partners, abortion-related experiences, unwanted pregnancy, and financial deception involving parents.

The discipline-problems scale comprised 10 items, each answered about the respondent’s own behaviour: 1) frequent absence from school; 2) frequent lateness to school; 3) despising, harassing, attacking, or disrespecting staff; 4) quarrelling and causing problems with landlords or fellow tenants; 5) bribing, cheating, or lobbying for success in examinations; 6) quarrelling and fighting with dangerous objects; 7) smoking in public places; 8) participating in rebellious activities; 9) involvement in internet malpractices; and 10) disturbing neighbours with loud music.

All items were rated on a 4-point Likert-type scale ranging from 1 (strongly disagree) to 4 (strongly agree). Each 10-item composite was scored as the sum of its items, giving a possible range of 10 to 40, with higher scores indicating more of the measured construct. Expert judges rated the content validity of the instrument. The dating-practices scale had a validity index of 0.88, the discipline-problems scale had a validity index of 0.90, and the full instrument had a validity index of 0.92. A pilot study with 20 students in Bambili, a university neighbourhood associated with the University of Bamenda, produced Cronbach’s alpha values of 0.71 for dating practices, 0.70 for discipline problems, and 0.72 for the full instrument. Internal-consistency coefficients within the analytic sample itself were not archived and could not be recomputed for this article because the item-level dataset was no longer available; this is acknowledged as a limitation in Section 6.

Only fully completed questionnaires were treated as usable; questionnaires with missing, unreadable, or damaged responses were excluded at data entry. All analyses were therefore complete-case analyses conducted on the same N of 323, and no imputation of missing item responses was performed.

For the present article, the dating-practices score should be understood as an index of dating-related experiences rather than as a psychometrically established unitary trait. The items intentionally covered heterogeneous aspects of students’ relationship lives, ranging from communication and cohabitation to conflict, sexual risk, and financial deception. This breadth reflected the exploratory purpose of the original master’s study, but it also limits construct precision. Accordingly, the composite is retained to preserve the original analysis, while the individual item frequencies are reported so that readers can see what the score contains. The article therefore avoids interpreting a high composite score as a diagnosis, a stable personality characteristic, or a measure of “healthy” versus “unhealthy” dating.

A semi-structured interview guide of seven open-ended items was used for the interviews. Questions asked participants to describe how dating influenced class attendance, study time, conflict with peers or university authorities, financial pressure, substance use, pregnancy-related worries, and relationship conflict.

A letter of introduction from the Faculty of Education, University of Buea, facilitated access to the research setting, and the study was conducted under departmental and faculty academic approval as a supervised Master of Education thesis; at the time, the Faculty did not issue separate research-ethics committee reference numbers for supervised postgraduate studies, and none exists for this study. The lead author administered the questionnaires personally in hostels and leisure venues, explained that participation was voluntary, and answered respondents’ questions about confidentiality before completion. Respondents completed the questionnaires anonymously; no names or identifying information were collected on the survey instrument.

Interviews were conducted between 10 March and 10 April 2016 with students who agreed to provide more detailed accounts. Interviewees signed a written consent form, which described the purpose of the study, the voluntary nature of participation, the right to pause or withdraw at any time without consequence, the approximately 60-minute duration, and the request to audio-record (with the option of note-taking only for those who declined recording). Names and identifying details were removed from the interview records, and audio materials were kept by the researcher and shared only with the thesis supervisor and examination panel. Participants were given the supervisor’s contact details as an independent channel for reporting any concern about harm, breach of trust, or disclosure of identity. No formal referral protocol for disclosures of coercion or violence was in place in 2016; interviewees who raised distressing experiences were reminded of the voluntary nature of the interview and of available university student-support structures. The absence of a formalized referral and safeguarding pathway is acknowledged as a limitation in Section 6.

Survey data were analysed with SPSS Version 20.0. Descriptive statistics summarized the dating-practices items. Pearson correlation and bivariate linear regression tested whether dating practices predicted discipline problems. Statistical significance was evaluated at p < 0.05, with threshold p values reported in APA style. All inferential analyses were complete-case analyses with N = 323.

The inferential statistics reported in Section 3.3 reproduce the archived SPSS Version 20.0 output tables preserved in the thesis record (correlation, model summary, ANOVA, and coefficients tables). Composite-level descriptive statistics were not separately archived; the composite standard deviations and the discipline-problems composite mean reported in this article were therefore derived algebraically from the archived regression output, as indicated in the notes to Table 1 and in Section 3.3. In the original thesis tables, the figure reported as the dating composite standard deviation (9.95) was in fact the sum of the ten item standard deviations; this labelling error has been corrected here.

Interview responses were summarized by the researcher into recurring content categories at the time of the original study; independent double-coding was not undertaken, and verbatim transcript excerpts were not available for verification during article preparation. Consistent with this audit trail, the interview material is presented in Section 3.2 as contextual participant accounts—careful summaries of what interviewees reported—rather than as a formally coded qualitative analysis, and no thematic claims are made beyond what those summaries support.

Table 1 presents the frequencies, percentages, means, and standard deviations for the 10 dating-practices items.

Table 1.Frequencies, percentages, means, and standard deviations for dating-practices items (N = 323).

Note. Items were rated on a 4-point Likert-type scale (1 = strongly disagree to 4 = strongly agree). Agree combines agree and strongly agree responses. The composite is the 10-item sum (possible range 10 - 40); the average item mean was 3.21. The composite SD was derived algebraically from the archived regression output (Section 3.3), because composite-level descriptive output was not separately archived; the value of 9.95 reported in the original thesis tables was the sum of the ten item SDs and is not a valid composite SD.

Using a midpoint criterion of 2.50 on the 4-point scale, the average item mean of 3.21 indicates agreement-oriented responses across the dating-practices items. The most frequently endorsed item was partner communication through Facebook or WhatsApp (89.2%), followed by aggression related to stress from a partner (86.4%), financial deception involving parents (82.0%), pornography use with a partner (81.7%), and multiple sexual relationships (79.9%). These percentages should be read as 2016 baseline patterns in a purposively sampled student neighbourhood, not as current prevalence estimates for Cameroonian university students generally.

The 20 post-survey interviews provided contextual accounts of how dating practices entered students’ academic and neighbourhood lives. As explained in Section 2.6, these accounts are researcher-generated summaries of what interviewees reported; they are presented as interpretive context for the survey association, not as a formally coded qualitative analysis, and no illustrative quotation has been added because none could be verified from the original transcript record. Two broad patterns were most prominent in the summaries.

The first pattern concerned academic time displacement. Students described missing classes, arriving late, keeping late nights, spending weekends with partners, and using study time for relationship communication. In their accounts, the immediate problem was not the existence of a romantic relationship but the displacement of academic routines by relationship demands. This account gives practical meaning to the survey association: dating became disciplinary when it reorganized time away from lectures, assignments, sleep, and study.

The second pattern concerned emotional strain, financial pressure, and risky coping. Students described anxiety related to dishonesty, pregnancy, money, and partner commitment. Some accounts linked relationship stress to aggression, substance use, multiple partnerships, and sexual coercion. These accounts help explain why dating practices in the survey were positively associated with discipline problems: relationship stress entered academic life through time loss, financial deception, conflict, and harmful coping.

Dating practices were positively correlated with discipline problems, r (321) = 0.68, p < 0.001, which is a large effect according to Cohen’s [27] conventional guidelines (Table 2). For the discipline-problems composite (possible range 10 - 40), the mean derived from the archived regression output was 31.34 and the derived standard deviation was 7.38.

Table 2.Pearson correlation between dating practices and discipline problems (N = 323).

Note. *p < 0.001.

A bivariate linear regression was then conducted with discipline problems as the outcome variable and dating practices as the predictor. As shown in Table 3, the model explained 46.8% of the variance in discipline-problem scores.

Table 3.Model summary for the regression of discipline problems on dating practices.

Note. Predictor: Dating practices. N = 323 (complete cases).

The regression model was statistically significant, F (1, 321) = 282.19, p < 0.001 (Table 4).

Table 4. Analysis of variance for the regression model.

Note. Dependent variable: Discipline problems. Predictor: Dating practices.

Table 5 presents the regression coefficients. The unstandardized coefficient indicated that each one-point increase in the dating-practices score was associated with a 0.64-point increase in the discipline-problems score. The regression equation was: discipline problems = 10.72 + 0.64 (dating practices). The null hypothesis was rejected.

Table 5.Regression coefficients for dating practices predicting discipline problems.

Note. CI = confidence interval. All values reproduce the archived SPSS Version 20.0 output.

The age of the data is the manuscript’s clearest limitation, but historical data can still be useful when the claim is appropriately bounded. The present study captures a particular moment in Cameroonian university life: Facebook and WhatsApp were already established in students’ relationship communication, yet the current platform ecology, smartphone penetration, image-based communication, and app-mediated dating had not developed to their present scale. The value of the dataset therefore lies in documenting a pre-current-platform baseline rather than in providing a 2026 estimate.

For that reason, the 2016 data should be read as a pre-2026 baseline for questions that have become more urgent: How have WhatsApp and other platforms changed student dating? How do digital interactions affect consent, jealousy, coercive control, and academic attention? Are students able to report sexual coercion safely? Are counsellors available and trusted? Do landlords, hostel caretakers, families, and universities coordinate around student safety? Recent work on cyber-dating abuse shows that technology can facilitate control, monitoring, humiliation, and psychological harm within intimate relationships [6][32]. This makes the Molyko findings more relevant, not less, provided they are not overstated as current prevalence data.

The findings also call for a shift in language. A purely moral framing risks blaming students while ignoring the conditions that shape behaviour. A contemporary framing should use the language of campus well-being, healthy relationships, consent, safeguarding, digital safety, mental health, and university-community responsibility. The article therefore does not argue that dating causes indiscipline. It argues that, in the 2016 Molyko setting, particular dating-related practices were strongly associated with discipline problems and should be treated as a legitimate student-affairs, counselling, and public-health concern.

This shift is also consistent with contemporary public-health guidance. WHO treats adolescent and young people’s sexual and reproductive health as a matter of information, life skills, supportive environments, accessible services, protection from coercion, and the capacity to make informed decisions [33], and frames adolescent health more broadly as a developmental and rights-based concern rather than a disciplinary one [34]. WHO-supported estimates published by Sardinha and colleagues show that intimate partner violence begins early for many young women and can affect education, mental health, pregnancy risk, and future relationships [7]. These concerns are especially salient in central sub-Saharan Africa, where the estimated burden of partner violence among adolescent girls is comparatively high.

Cameroonian programme evidence points in the same direction. UNICEF’s 2025 situation analysis reports that adolescent girls in Cameroon continue to face sexual violence, early reproductive transitions, and gaps in HIV knowledge, while its adolescent-girls programming explicitly combines education, gender-based violence prevention, sexual and reproductive health, and youth participation [35]. UNFPA has similarly supported age-appropriate sexual and reproductive health information and services in Cameroon’s North West and South West Regions and, at the 2024 University Games, provided university-age young people with sexual and reproductive health and gender-based violence prevention information and services [36][37]. These are relevant precedents for universities because they frame prevention around services and capabilities, not only prohibition.