This study examined the correlates and predictors of prosocial behavior among Chinese adolescents in Hong Kong. A sample of 518 high school students responded to a questionnaire containing measures of antisocial and prosocial behavior, prosocial norms, pragmatic values, moral reasoning, and empathy. Preliminary analyses showed that there were gender differences in some of the measures. While correlation analyses showed that parental education, prosocial norms, pragmatic values, moral reasoning, and empathy were related to prosocial behavior, regression analyses showed that prosocial norms, pragmatic values, and empathy dimensions (personal distress and empathy) were key predictors of it. The findings are largely consistent with theoretical predictions and previous research findings, other than the negative relationship between personal distress and prosocial behavior. The study also underscores the importance of values and norms in predicting prosocial behavior, which has been largely neglected in previous studies.
Prosocial behaviors are actions that aim to fulfill another person’s need for support or to promote and sustain a positive benefit for them [
The cultivation of prosocial behavior has long been an important objective of compulsory education and youth development programs. Prosocial development is closely linked to various positive developmental outcomes for young people including academic success, positive self-worth, positive relationships with others, and higher social competence [
In contrast to the abundance of western research on adolescent prosocial behavior, very few studies have been done with Chinese adolescents. In addition, while there are many studies on antisocial and deviant behaviors among adolescents, comparatively fewer look at prosocial behavior, including the predictors of it in adolescents. For example, Ma and colleagues [
It is likely that prosocial development has largely taken shape by mid-to-late adolescence. By then, young people are likely to have developed a set of values to guide their behavior. Unfortunately, few studies have examined how personal values may influence prosocial behavior among young people. In this study, it is hypothesized that the adoption of prosocial norms and rejection of pragmatic values will play a part in determining prosocial behavior of young people. Prosocial norms are standards and beliefs, or the set of shared social expectations of healthy, ethical, appropriate, and culturally desirable actions, that promote prosocial behavior and minimize health risks [
There are also theoretical accounts proposing that individual differences in prosocial development are closely linked to empathy-related constructs like sympathy, personal distress, perspective and role taking, social awareness, and moral reasoning [
Finally, one might ask how sociodemographic factors (such as age, gender, and level of parental education) relate to adolescent prosocial behavior and the attributes which may be associated with it (such as prosocial norms, pragmatic values, moral reasoning, and empathy constructs). In terms of age, since adolescents develop better moral reasoning as they get older, it may be expected that age is linearly related to prosociality. On the other hand, some theorists may also argue that an increase in age may not necessarily lead to an advancement of moral values, and there may be no relationship between these factors. Since there is not much research in this area, the present study attempts to examine the relationship between age and prosocial behavior. Furthermore, as females tend to show more empathy and to be more relationship oriented, it may also be expected that they will demonstrate higher levels of prosocial norms, moral reasoning and prosocial behavior, and lower levels of pragmatic values. Finally, as parental education implies greater social capital (such as better parenting and more involvement), it can also be predicted that parental education levels will be positively related to adolescent prosocial behavior. Against this background, the specific objectives of this study were (a) to examine the correlations between basic demographic factors (age and gender) and adolescent prosocial behavior and its related attributes (including prosocial norms, pragmatic values, moral reasoning, and empathy-related constructs); (b) to study the correlates of prosocial behavior, including prosocial norms, pragmatic values, prosocial reasoning, and empathy-related constructs; (c) to identify the key predictors of prosocial behavior among late adolescents. A cross-sectional survey was conducted using a sample of high school students in Hong Kong.
The participants in this study were recruited from a group of high school students (secondary four to six) who attended a one-day “Teen Talk” event organized by the Hong Kong Law Society. The event was titled “Love Yourself, Love Others,” and its purpose was to engage young people to discuss their core values as well as those of society in a one full day seminar. As well as this, the event attempted to increase participants’ understanding of legal and social issues.
A total of 533 participants completed and returned survey questionnaires through their schools, giving a response rate of around 35.5%. Fifteen questionnaires were not included in the analysis either because more than 10% of items were incomplete or because they were extreme outliers. After discarding these questionnaires, 518 remained in the dataset. The participants were full-time students aged 14–22 (M = 16.2, SD = 1.1). There were more females (69.6%) than males (30.4%). They were recruited from 36 secondary schools and studying in secondary 4 (54.0%), 5 (40.2%), or 6 (5.8%). When asked about their educational achievements compared to classmates, more than one-third (37.1%) regarded themselves as better than average, and around half (48%) said they were average. A large proportion of participants regarded their conduct as better than average (48%) or very much better than average (14.3%).
Most of the participants were the only child in their family (72.4%), and 23.7% had one sibling. The median education level of both respondents’ parents was secondary three. Financially, only 7.8% (
Ethical approval for this study was obtained beforehand from the Departmental Research Committee of The Hong Kong Polytechnic University. After access had been obtained, the schools which had participated in the Teen Talk event helped to distribute the questionnaires, invitation letters, and consent forms. If potential participants were under 18, an invitation letter was also sent to their parents. Parents were asked to sign a consent form and return it to the school. All participants were also requested to sign a consent form for voluntary participation. The schools sent the completed questionnaires and consent forms back to the researchers.
The adolescent behavior questionnaire (ABQ) is a generic Chinese-language instrument designed to measure the pro- and antisocial behavior of adolescents [
Three items were taken from the Chinese youth positive development scale (CYPDS) to measure how willing participants were to provide help to the needy, participate in volunteer work and follow school rules. The reliability and validity of the CYPDS have been demonstrated in previous validation studies [
The items used in a youth opinion poll titled “Young People’s Outlook on Life” [
The Chinese version of the prosocial reasoning objective Measure (PROM) was used to assess participants’ prosocial reasoning. The PROM is a measure assessing prosocial moral reasoning in young people and adults [
The 21-item Chinese interpersonal reactivity index (C-IRI) is a self-report questionnaire consisting of three subscales: fantasy (FS), empathy (ES) and personal distress (PD). The participants were asked to indicate the degree to which each item described them using a 5-point Likert-type scale, which varied from 0 (does not describe me well) to 4 (describes me very well). A higher score in a subscale represents a higher functioning in each aspect of empathy. The C-IRI has acceptable psychometric properties in Chinese adolescent samples [
A multivariate analysis of variance (MANOVA) was conducted to examine gender and age differences in prosocial behavior and its related attributes. The sample was stratified into 4 age groups: 14-15 (
Further analyses using univariate ANOVAs showed that males reported more antisocial behavior than females, but there were no significant gender differences in prosocial behavior. However, using the overall ABQ scores as the outcome indicator, females were significantly more prosocial than males (
Sex differences in prosocial behavior and its correlates.
Variablea | Sex | M | SE | 95% CI | |||
Lower | Upper | ||||||
Antisocial behavior | M | 3.46 | .22 | 3.03 | 3.88 | 20.64*** | .04 |
F | 2.25 | .15 | 1.95 | 2.55 | |||
Prosocial behavior | M | 3.31 | .14 | 3.02 | 3.59 | .49 | .00 |
F | 3.43 | .10 | 3.23 | 3.63 | |||
ABQ score | M | −.15 | .22 | −.59 | .29 | 23.77*** | .03 |
F | 1.18 | .16 | .87 | 1.49 | |||
Prosocial norms | M | 4.65 | .05 | 4.54 | 4.75 | 12.79*** | .03 |
F | 4.88 | .04 | 4.81 | 4.95 | |||
Pragmatic values | M | 2.72 | .04 | 2.63 | 2.81 | 6.67* | .01 |
F | 2.86 | .03 | 2.80 | 2.92 | |||
PROM overall weighted | M | 6.46 | .08 | 6.31 | 6.61 | 20.47*** | .04 |
F | 6.88 | .05 | 6.77 | 6.99 | |||
Personal distress subscale | M | 1.98 | .05 | 1.88 | 2.07 | 20.03*** | .06 |
F | 2.28 | .03 | 2.22 | 2.35 | |||
Fantasy subscale | M | 2.07 | .06 | 1.95 | 2.19 | 11.01** | .02 |
F | 2.32 | .04 | 2.24 | 2.41 | |||
Empathy subscale | M | 2.45 | .04 | 2.36 | 2.52 | 10.92** | .02 |
F | 2.60 | .03 | 2.54 | 2.65 |
Note: aestimated marginal means are shown. *
Correlation analyses were conducted to identify potential predictors of prosocial behavior (Table
Correlations between potential predictors and adolescent antisocial and prosocial behavior.
Potential predictors | Antisocial behavior | Prosocial behavior | Adolescent behavior |
---|---|---|---|
Age | .06 | .05 | −.02 |
Parent education | −.08 | .18*** | .19*** |
Prosocial norms | −.28*** | .30*** | .46*** |
Pragmatic values | .32*** | −.09* | −.36*** |
PROM overall weighted | −.16*** | .09* | .21*** |
Personal distress subscale | .11* | .02 | −.09* |
Fantasy subscale | .10* | .14*** | −.00 |
Empathy subscale | −.18*** | .32*** | .39*** |
The findings on the predictors of prosocial behavior in adolescents are shown in Table
Prediction of prosocial behavior (ABQ) among adolescents (
Predictors | SE | Collinearity statistics | |||||
Tolerance | VIF | ||||||
Pragmatic values | −.90 | .20 | −.19 | −4.52 | <.001 | .81 | 1.23 |
Prosocial norms | 1.28 | .17 | .32 | 7.53 | <.001 | .75 | 1.33 |
Prosocial reasoning (PROM overall weighted) | .13 | .11 | .05 | 1.17 | .24 | .84 | 1.19 |
Empathy subscale | .98 | .25 | .18 | 3.97 | <.001 | .68 | 1.46 |
Personal distress subscale | −.63 | .17 | −.14 | −3.59 | <.001 | .92 | 1.08 |
Note:
There were some differences in the results of the regression analyses conducted for the male (
Prediction of prosocial behavior (ABQ score) among males and females.
Variables | SE | |||||
---|---|---|---|---|---|---|
Males | Personal distress | −1.45 | .35 | −.29 | −4.12 | <.001 |
Empathy | 1.05 | .46 | .17 | 2.27 | .02 | |
Prosocial norms | 1.15 | .27 | .31 | 4.30 | <.001 | |
PROM overall weighted | .19 | .19 | .07 | 1.00 | .32 | |
Pragmatic values | −.60 | .34 | −.13 | −1.77 | .08 | |
Females | Personal distress | −.48 | .20 | −.11 | −2.37 | .02 |
Empathy | .82 | .29 | .16 | 2.81 | .01 | |
Prosocial norms | 1.48 | .22 | .35 | 6.77 | <.001 | |
PROM overall weighted | .04 | .14 | .01 | .26 | .80 | |
Pragmatic values | −1.02 | .24 | −.21 | −4.25 | <.001 |
Note: For male subsample (
In terms of sociodemographic correlates, while the findings do not reveal any age effect, they demonstrate significant gender differences in ABQ scores, prosocial norms, pragmatic values, prosocial reasoning, and empathy (personal distress, fantasy, and empathetic concern). These findings are generally consistent with those reported in the literature as well as with the hypotheses of this study.
Consistent with the original expectations, several factors are related to prosocial behavior in both the correlation and regression analyses. Higher levels of empathy and prosocial norms as well as a lower level of pragmatic values are associated with a higher level of prosocial behavior. These three predictors remain significant in separate regression analyses with the male and female subsamples. These results are highly consistent with theoretical predictions and previous empirical results [
The present findings indicate a significant association between personal distress and prosocial behavior, but the relationship is negative rather than positive in nature. Most previous studies show that higher levels of personal distress and sympathy are associated with more prosociality towards persons in need [
While prosocial reasoning as assessed by the PROM score is correlated significantly with prosocial behavior, it fails to predict prosocial behavior in the regression analyses. This may be due to the significant collinearity among prosocial reasoning, empathy, and prosocial norms. Analysis using condition indices shows that these three variables share significant variance proportions in several dimensions. Nevertheless, it is noteworthy that prosocial reasoning only explained a small proportion of variation in prosocial behavior (4.5%) when it was entered as the only predictor in the regression.
The four significant predictors, including empathetic concern, prosocial norms, pragmatics values, and personal distress, were able to predict around 30% of variation in prosocial behavior. The variance explained was a bit higher for the female (33%) than male (27%) sample. The percentage of variance explained was statistically significant, but much of the variation was not accounted for. In future studies, it will be a good idea to add predictors involving social influences such as peer and teacher influence, parent and school socialization, school discipline and encouragement for prosocial involvement, or interpersonal competence [
These findings should be interpreted with reference to several factors. Firstly, while the sample was large, it was a convenience sample recruited from high schools with the age range largely limited to late adolescence (age 15–18). Secondly, the students involved had been nominated by their schools to take part in the “Teen Talk.” Since it is likely that schools will prefer to have students with better academic or conduct records to take part in these types of community events, the sampled group may have displayed better previous conduct. Thirdly, we found that the range of prosocial behaviors in the ABQ could be expanded further. While it covers prosocial acts in home, school and social situations, additional items could be added to sample a wider range of prosocial behaviors like providing support or assistance to people one knows or does not know: sharing, listening, and comforting; appreciating others; working in a team; involvement in prosocial groups (such as service teams or religious groups) [
In summary, this study has shown that empathetic concern, personal distress, prosocial norms, and pragmatic values are key predictors of prosocial behavior among older adolescents in Hong Kong. These results are largely consistent with theoretical expectations and the findings of previous work, other than the negative relationship between personal distress and prosocial behavior. Further study is needed to examine the unique and combined effect of empathy-related constructs (empathy, sympathy, perspective taking, and personal distress) on prosocial development. This study also provides support for the importance of values and norms in predicting prosocial behavior, which has seldom been explored in previous studies. While prosocial reasoning shows a significant correlation with prosocial behavior, it is not a significant predictor of it. This is probably a result of the collinearity of prosocial reasoning with empathy and prosocial norms. In further studies looking at predictors of prosocial behavior, it will be necessary to design or employ a standardized measure that provides a wider coverage of such activity. It will also be necessary to select and measure additional predictors which reflect peer, family, and school influences on prosocial development.