Romantic relationships are deeply significant in the transition from adolescence to adulthood, particularly during college years. Understanding young people's attitudes towards love is crucial for guiding them towards positive and healthy relationships and future marriages. Existing research predominantly focuses on Western perspectives, neglecting the cultural nuances affecting attitudes in Eastern societies, such as China. This study aims to address this gap by investigating the trajectories of Chinese college students' attitudes toward love and identifying the factors influencing these trajectories. The decreasing marriage rates and social problems associated with late or absent marriages in China provide further impetus for this research, as understanding attitudes towards love is crucial to understanding marriage intentions. This research seeks to contribute to existing theories of love by analyzing developmental patterns and influential factors within the specific context of Chinese culture, providing valuable insights for societal interventions and educational strategies.

The study reviewed existing theories of love, including Rubin's distinction between "like" and "love," Lee's six love styles, Shaver and Hazan's attachment theory, Hatfield and Walster's passionate and companionate love, and Sternberg's triangular theory of love. It also examined the cultural context of love in China, noting the historical association of love with marriage and the significant shifts since the implementation of monogamy in 1950. The literature review explored previous research on developmental trajectories of romantic relationships in adolescents, highlighting the later onset of romantic activity in Asian adolescents compared to Western counterparts, and the influence of sociocultural background. Finally, it explored the influencing factors on attitudes toward love, including gender differences, age, hometown location, extroversion, major, ethnicity, siblings, BMI, self-esteem, and parental education levels and socioeconomic status. The review revealed inconsistent findings regarding the relationship between these factors and attitudes towards love, emphasizing the need for further investigation, particularly within the Chinese cultural context.

This longitudinal study utilized a five-year dataset from the Beijing College Student Panel Survey, encompassing data from 2473 students (1166 female, 1307 male) from their freshman to senior years. The core variable, attitudes toward love, was measured using a 23-item Romantic Love Scale, adapted from Knox and Sporakowski (1968) and further refined through principal component analysis to extract romantic and realistic attitude indices. Cronbach's alpha values for the romantic and realistic indices ranged from 0.67 to 0.76 across the four years. Other variables included gender, ethnicity, hometown location, siblings, major, BMI, extroversion, age, parental education levels, family economic and social status, and self-esteem. Descriptive statistics were used to analyze changes in romantic and realistic attitudes over time. Pearson correlation analyses investigated relationships between attitudes toward love and other variables. Growth mixture modeling (GMM) using Mplus 8.3 was employed to identify latent trajectory classes for both romantic and realistic attitudes. Model fit was evaluated using AIC, BIC, SABIC, LMR-LRT, B-LRT, and entropy. Finally, multinomial logistic regression analysis in Stata 15.0 examined the influence of various factors on trajectory class membership.

Descriptive statistics revealed that both romantic and realistic attitudes generally increased from freshman to junior year, then slightly decreased in the senior year. Realistic attitudes consistently scored higher than romantic attitudes. Correlation analysis showed a significant negative correlation between romantic and realistic attitudes. Romantic attitudes were positively correlated with gender (male), major (humanities/social sciences), age, and BMI; negatively correlated with hometown location (rural), parental education levels, and self-esteem. Realistic attitudes were positively correlated with age, siblings, extroversion, hometown location (urban), parental education levels, and self-esteem; negatively correlated with gender (female), major (humanities/social sciences), ethnicity (minority), and family economic and social status. Growth mixture modeling identified three trajectory classes for romantic attitudes: "high-increasing" (1.05%), "low-decreasing" (13.79%), and "low-increasing" (85.16%). Three classes were also identified for realistic attitudes: "high-increasing" (1.58%), "low-increasing" (96.97%), and "high-decreasing" (1.46%). Multinomial logistic regression analysis revealed that gender significantly influenced romantic attitude trajectories, with males more likely to be in the "low-decreasing" group. Major significantly influenced realistic attitude trajectories, with students in science/engineering/agriculture/medicine more likely to be in the "high-decreasing" group. Extroversion and hometown location also significantly influenced realistic attitude trajectories.

The findings highlight the complex interplay between romantic and realistic attitudes toward love among Chinese college students. The generally increasing then slightly decreasing trend in both attitudes across the four years of college likely reflects the unique cultural context in China, where romantic relationships are often discouraged before college entry. The higher scores for realistic attitudes compared to romantic attitudes might be attributed to the collectivist nature of Chinese society, which emphasizes social norms and family expectations in partner selection. The influence of gender, major, extroversion, and hometown location on trajectory class membership underscores the importance of considering diverse individual characteristics and social contexts when understanding attitudes toward love. These findings challenge some previous research in the field and add a crucial cultural dimension to the study of romantic attitudes. The study's findings have practical implications for guiding college students towards healthy relationships.

This study provides valuable insights into the developmental trajectories and influencing factors of romantic and realistic attitudes toward love among Chinese college students. It demonstrates the importance of cultural context in shaping these attitudes, showing that realistic attitudes often outweigh romantic ones. Gender, major, extroversion, and hometown location emerged as key predictors of trajectory classes. Future research should investigate a wider range of potential influencing factors and employ more diverse methodologies to further refine our understanding. The findings suggest that colleges and society should incorporate education on healthy relationships into curricula and actively guide students toward developing mature and balanced attitudes toward love.

The study's limitations include the use of a selective sample, potentially limiting generalizability to other populations and countries. The reliance on self-report measures could introduce subjective biases. The relatively low Cronbach's alpha values for some scales also warrant caution in interpreting results. Finally, the inclusion of only a limited number of intervenable factors restricts the explanatory power of the study. Future research should address these limitations to enhance the robustness and generalizability of the findings.