The Relationship between Media Usage and Positive Youth Development with the Mediate Effects of Internet Use Gratification

Internet offers countless opportunities that affect students' lives and their career potentials. With the help of Internet, youth development processes are positively encouraged. In order to understand how Internet can influence Positive Youth Development (PYD), this study was carried out with an attempt to identify the relationship between problems of using Internet, income


INTRODUCTION AND PROBLEM STATEMENT
Malaysia is an Asian country with visions of being a developed country in the region. In 2015 the Malaysian Youth Policy attempted to empower the youth with leadership qualities enabling them to lead successfully (Arshad & Hong, 2019). Thanks to the guidelines of the Positive Youth Development approach, a large number of young workers have designed specific programs for adolescents and young adults (Wiium & Dimitrova, 2019). In Positive Youth Development (PYD), youth attendance in leadership activities have a direct relationship with the youth developmental processes (Li & Wang, 2012). Leadership development plans has a direct influence on the efficiency of youth mentoring programs. Implementing mentoring skills in youth leadership development is quite vivid in the production of a better perception through the cultivation of knowledge acquisition, attitude, and practice. Nevertheless, mentoring cannot be easily assessed as skills, because it is not merely transferring knowledge; mentoring highlights incorporation of leadership qualities into elements of PYD in the youth development .
Earlier research on the issue of youth development were based on some inaccurate presumptions (Lerner & Steinberg, 2009), such as Hall's views who observed youth in accordance with what they were lacking in comparison to adults (Hall, 1904). As an example, the concept of "storm and stress" introduced by Hall in 1904, is known as starting point of adolescent psychology which was the dominant attitude in the realm of teaching and learning for many years. Judging the youth in terms of what they do not have, affected people's expectations from teenagers ). However, later in 1990s, this dominant framework changed and viewed youth as a product of connections between people and their worldview (Lerner, 2005). Lerner (2006) mentioned that plasticity was considered as a key point in this new outlook which means that people have certain potentials for change in their life time; such potentials are truly significant because adolescents' course of life is not constant and it can be easily impressed by the environment. The Internet provides people with numerous opportunities that can influence a person's life. Therefore, the present study attempts to provide a better understanding of Internet usage and the following factors of youth positive development: confidence, competence, character, connection, and caring (5Cs). Considering earlier research findings, this paper aims to point out the following questions: 1. What is the mediating effect of Internet gratifications use in the relationship between 'problems of using Internet' and 'income' with the selected dimensions of positive youth development? 2. What is the mediating effect of Internet gratifications use in the relationship between field of study and the selected dimensions of positive youth development?
Viruses and malwares are also among the problems of Internet usage. Muniandy et al. (2017) carried out a study to look into the issue of Internet use problems and concluded that users are not aware of or ignore online threats; not having knowledge on the application of security features and anti-virus status will definitely increase malware risks. In addition, Al-khatib & Hammood (2017) reported that users do not have much information on software security, and they merely care about recent features of new applications. Similarly, Tariq (2020) stated that according to a cyber-security department (MyCERT) in Malaysia, more than 400 cases of defacement took place in Malaysia including 19 governmental sections. Azlan et al., (2020) noticed that the most significant problem in using Internet is that many users use the Internet through unstable connections because fixed-line Internet is not affordable to everyone. In a developing Asian country like Malaysia, Internet connection technologies are available to almost all users; the speed and stability of the connections are not the same though. Rural districts lack serious connection facilities compared to urban areas, which could be considered as a serious issue (Demographics Malaysia, 2019). The issue of poverty and the fact that not all families can afford digital devices, adds to the problem as well (Azlan et al., 2020).
According to Ayub, Hamid, and Nawawi (2014), students' field of study is highly influential in using Internet for educational purposes. In 2016, Hanchard carried out a study in Malaysia among different ethnic groups to identify the amount and quality of Internet usage. Hanchard (2016) concluded that there is a wide gap between the educated users and low-income users. Participants with higher income and educational levels have more chances of using social media through their cell phones while this is not the case with lower income and less educated participants. In other words, users with richer economic resources are more susceptible to value obtaining information from social media, while users with weaker socio-economic backgrounds are more susceptible to value religious sources for acquiring information. According to Kavanaugh et al., (2005), skillful and educated users can develop social distances via searching information and as a result they encounter more chances of success in their life.

Media Usage Among Youth
Statcounter (2020) revealed that the number of Internet users in Malaysia is constantly growing up to 2020. Dalol et al., (2021) also stated that 34% of the subjects in his study were online six hours a day. According to Roy (2019), modern media affects people differently based on their character and attitude which in turn are impressed by culture and society.
In addition, communication development technology provides comfort for people, as the development of borderless Internet changed their lifestyle strikingly. As Norhabiba and Putri (2018) put it, young generations nowadays are Internet-literate and digital-oriented empowered with the essential computer skills. Bowyer et al. (2017) stated that the youth use the media for political goals as well; for example YouTube is mostly used for personal media sharing and subscribing to political videos. In another place, Hassan and Azmi (2018) reported that Internet is mainly used for searching information and making decisions. Simon and Sarah (2019) also stated that 80% of people shop online in Malaysia which highlights the significant role of online social media particularly during the Covid-19 lock down. During the Pandemic, people had to restrict their physical relations and stay home longer hours and thus online activities became more popular (Mat Dawi et al., 2021).

Positive Youth Development and Internet Usage
According to the outlooks of Positive Youth Development (PYD), trying will lead to participation and contribution. Wiium et al., (2019) carried out a cross-sectional study among 858 Ghanaian senior high school students and reported that boys scored higher on the element of competence compared to girls and that girls scored higher on the element of caring compared to boys. Boys and girls scored almost the same on socio-economic part Gender differences could be a result of the culture of the nation and social procedures sustaining gender inequalities in a society.  explored the way youth are nourished to impress positive constructs introduced by PYD. The youth have already attended leadership courses including mentoring. The results revealed that competence development occurs during mentoring as well as in the position of the mentor's organizational community. Therefore learning and connection are not restricted to what happened within the one-to-one mentor-protege dyad . It was also reported that traits characterized by the '5Cs' in Positive Youth Development (PYD) progressed through mentoring.
According to Leman et al., (2017), studies on Positive Youth Development (PYD) aim to perceive and encourage positive features of development among the youth. Role of cultural and social background is also very important in PYD and it should be considered in any similar research. Rosenfeld, Richman and Bowen (2000), investigated social support networks and schools, stating that social support promotes positive youth development as well as positive school development. Social support seeks for help and interaction from potential supporters such as parents, friends, and teachers who affect students' school performance. Better outcomes are obtained once teachers' support is accompanied with that of parents and friends. Gomez and Ang (2007), stated that Malaysia is encouraging positive youth development programs which can produce positive effects on young people's behavior, academic success and other areas of development.
Abdul Kadir et al., (2012) investigated the relationship between the level of financial welfare and positive feelings of individuals and found out that external assets can predict positive feelings and can increase positive development among the youth. In another study, Meng, Hamzah, Suandi, and Abullah (2012), found that positive youth development and youth learning enable the youth to find their way of life. The right path is shown to them instead of dictating them what to do. According to Datuk Saifuddin Abdullah, Malaysia's deputy minister of higher education (2012), the younger generations need to be empowered with trust, motivation, energy, and the ability to think well and decide. Meng et al. (2012), contended that young generations are valuable human resources of a nation and therefore proper programs, such as youth leadership, need to be generated to enable them for a better life. Kaur et al., (2019) studied the development of prosocial behaviors in adolescents and concluded that cognitive autonomy is envisioned by only being engaged in the community, while having a purposeful life in envisioned by school, family, friends, and society engagement which is an indicator of prosocial behaviors among the youth. According to Wan et al. (2019), with the rapid increase of smart phones, more than 66% of the people in the world have access to social media. Mobile social networking has become a key communicative tool in everyday life of people.
Samsuddin and Mohamed Shaffril (2019), investigated dimensions of positive youth development (PYD) which are related to using Information Communication Technology (ICT) in rural libraries among young Malaysians. The outcome revealed that the level of education and employment status are important and also the variable of age is positively related to PYD and using ICT at rural libraries. It also became apparent that implementing PYD though using ICT in such places deemed necessary. Sarwar, Zulfiqar, Aziz, and Ejaz Chandia (2019), reported that usefulness, availability and joy are closely connected with using social media which functions as a powerful tool to accelerate learning by motivating students through communication and cooperation. Sarwar et al. (2019) also referred to a negative relationship between joy and collaborative learning; cyber-bullying was also proved to be a suppressing element in the positive relationship between collaborative learning and learner performance. Erdem, DuBois, Larose, De Wit, and Lipman (2016), discovered that mentoring programs are useful prevention tools for emotional and behavioral problems of at-risk youths. They considered the indicators of PYD as mediators of the relationship between mentoring support and the youth emotional and behavioral problems. Indicators of PYD are: competence, confidence, connection, care and compassion, and character. The results confirmed that PYD mediates the relationship between the two variables of the study; however, the relationship between mentor support and PYD was confined to youth in active mentoring relationships. Consequently, the present research investigates the relationship between positive youth development and the following four groups of independent variables: Internet use gratification, problems of using Internet, income level, also filed of study considered as a moderator.
In this study, two variables of problems in using Internet and income were considered. Each variable directly affected Internet use gratification and positive youth development. Gratification of using Internet was considered to have mediation effect in using Internet and PYD. The question is whether students' field of study is important in the quality of Internet use gratification and PYD. The research framework is presented below.

Uses and Gratification Theory and Media Usage
The theory of Uses and Gratification (U&G) describes the reasons and ways of using the media by the users who consciously choose their preferred media based on their individual needs (Stafford et al., 2004). According to (Flaherty et al., 1998, p. 252), the U&G theory has five main suppositions: a) Communication is goal-oriented b) Users actively choose their desired communicative media  Samani and Guri (2019), perceived gratification is obtained once the users' sense of need is satisfied and the gratification is verified once experienced. Internet users normally surf the web pages with a purpose in their minds. Unlike the traditional mass media, websites offer a specific function for people who need them.
According to Teng and Joo (2017), people use the media to achieve their unique and personal demands; for instance older generations mostly use the social media to keep in touch with family and friends and attempt to keep up with the pace of modern life which is known as the cognitive needs in the theory of Uses and Gratification. Musa et al. (2016) studied the application of U&G theory in using social media in Nigeria and concluded that most people use social media for the purpose of communication, passing time, searching, socializing, education, self-expression, and having fun.

Location and Sampling
Local male and female students of University Putra Malaysia (UPM) participated in this study. They were between 18 to 40 years old from three major races of Malay, Chinese, and Indian who were doing their Bachelor, Master and PhD in this university. According to the Malaysian definition, the age range of 15 to 40 years old is defined as youth (Yunus, 2007). Total number of 13726 undergraduate students and 7444 postgraduate students were considered as the population. Postgraduate and undergraduate students were identified once they were given the questionnaires; the participants also needed to specify their educational level in the questionnaire as well. Simple random sampling was used in the present study according to which, homogenous subgroups are created first and then a proportionate sample from each subgroup is selected (Keyton, 2006). The questionnaire was distributed in16 faculties according to the population of each faculty. To determine the sample size for this research a formula by Israel (1992) was used. The formula is n= , n = Sample size, N = Population (postgraduate and undergraduate), e = Precision level (when confidence level is 95% or 0.05). Using the formula, the sample size in the present study was calculated as: 1) n = 4) n = 2) n = 5) n = 392.58

3) n = 6) n = 392 Respondents
The result of Israel Formula was 392; however, more data was collected from 440 students (n=440) of 16 faculties at UPM. The sample size for each faculty was determined by dividing the student population of each faculty by the total population and multiplied by 440. For instance, for the Faculty of Agriculture, is calculated for undergraduate, and for postgraduate . Therefore, the total number of questionnaires to be filled in the faculty of agriculture is 36, for undergraduate 26, and for postgraduate 10. This study is part of a larger study measuring the impact of Internet usage on youth negative or positive development among University students in Malaysia.

Participants
The design of the study required data to be collected from local students from three major races of Malay, Chinese, and Indian who were doing their Bachelor, Master and PhD at UPM. Questionnaires were distributed among students who were Malaysian citizens, or permanent residents or people whose family lived and worked in Malaysia and their children or even any one of the family members studied at UPM. The random sample of undergraduate and postgraduate students was asked to participate in taking the short form of the positive youth development survey addressing topics such as problems of using Internet, and gratification of using Internet. Other questions addressed some personal information of the respondents such as income and field of study. Descriptive statistics were used to summarize the sample. The inclusion criteria consisted of age between 18 to 40 years old based on the definition of the youth in Malaysia. Therefore, questionnaires were distributed among students whose ages were below 40 and use Internet. Students older than 40 years old were deleted from the analysis.

Measurement
Undergraduate and postgraduate students were asked to participate in taking the short form of the positive youth development survey addressing topics such as problems of using Internet and gratification in using Internet. Other questions addressed demographic of the respondents such as income level and field of study. The study included a set of items for measuring IVs and DVs with 112 items. 19 items addressed problems of using Internet with five response alternatives from 1= "Not at all" to 5 = "Very Frequent". The range of the score is from 12 to 60. This construct was operationalized as interval measurements.
The gratification of using Internet was measured by 23 items, on a five-point Likert scale, respondents were asked for their gratification in using the Internet ranging from 1= "Strongly Disagree" to 5= "Strongly Agree" for each statement. Range of score for this dimension is from 23 to 115. This scale was operationalized as an interval measurement. It was divided into five dimensions namely cognitive with 6 items, personal integration with 5 items, escape, social integration, and affective with 4 items.
Positive youth development was measured by 68 items form institute for applied research in Positive Youth Development (PYD) developed by Lerner (2010). It was measured by 5-point Likert scale. This scale was operationalized as an interval measurement. It was divided to five dimensions namely Confidence, Competence, Character, Caring and Connection. Confidence was measured by 9 questions which were divided into two parts; self-worth with 3 questions ranging from 1= "never true" to 5= "always true") and positive identity with 6 questions (ranging from 1= "Strongly disagree" to 5= "Strongly agree"). Competence was measured by 10 questions which were divided into four categories: 3 questions for academic competence, 1 question for grade, 3 questions for physical competence, and 3 questions for social competence. Character was measured by 18 questions, which were categorized by personal value with 5 questions, social conscience with 6 questions, value of diversity with 4 questions and conduct behavior with 3 questions. Caring was measured by 9 questions related to sympathy disadvantaged, loneliness, unfortunate, pain and rejection. Connection was measured by 22 questions which were divided to family with 6 questions, University with 7 questions, community (neighbor) with 5 questions and peers with 4 questions.

Data Analysis
This study used Partial Least Square (PLS) 1 as the most important motivations are exploration and prediction, which is recommended in an early stage of theoretical development in order to test and validate exploratory models. It is also suitable for prediction-oriented research. Thereby, this methodology assists the explanation of endogenous constructs. In addition, descriptive analysis made use of mean scores, standard deviation (SD), frequency, and percentage to determine demographic of the respondents such as income level, field of study, problems of using Internet, and gratification of using Internet as mediation, and the dependent variable is Positive youth development (PYD). PYD was divided into five main categories of competence (social competence, physical competence, academic competence and grade), confidence (positive identity and self-worth), connection (family, community, school, peer), character (social conscience, conduct behavior, value of diversity and personal value), and caring. PLS analysis was employed for the regression analysis for mediating effects of field of study with positive youth development.

Internet Use Gratification
Based on the results, the highest overall mean belonged to cognitive dimension followed by escape and social integration dimension. In the cognitive dimension, the highest mean belonged to "To get information about something" (M =4.38, SD =0.70) and the lowest mean was related to "To keep up to date on popular sites" (M =3.90, SD =0.89). In the escape dimension, the highest mean belonged to "To relax and unwind" (M =4.21, SD =0.74) followed by "Because it makes me feel less tense" (M =4.00, SD =0.83). The highest mean in the social integration dimension is related to the "To keep in touch with people" (M =4.15, SD =0.73) followed by "Because I can talk with different people" (M =3.79, SD =0.92). The item "Because it's entertaining" had the highest mean in affective dimension (M =4.12, SD =0.76) and the lowest mean belonged to "To role play or experiment with my identity" (M =3.41, SD =0.99). Last and least overall mean belonged to personal integration dimension "I can do things in my own space "(M =4.09, SD =0.80) followed by "Because I can do whatever I want" (M =3.74, SD =0.90).

Problems of Using Internet
Regarding the problems of using Internet, the highest mean was related to slow Internet connection (M=3.58, SD=0.96) followed by overload of information on the Internet (M=3.50, SD=0.91). The lowest mean was related to the Internet being too complicated to use (M=2.25, SD=0.89). The findings of this study are consistent with previous studies that reported the main problem of using the Internet is related to the slowness in downloading (Mui et al., 2002;Riahinia & Azimi, 2008;Suryani, 2007).

Positive Youth Development
The element of confidence in positive youth social development was divided into two categories: positive identity and self-worth. Positive identity had the highest mean of 21.99 in the confidence dimension. In the positive identity dimension "when I am an adult, I'm sure I will have a good life" had the highest mean (M=4.41, SD=0.84) and "at times, I think that I am no good at all" (M=2.81, SD= 1.20) had the lowest mean score. In self-worth dimension "some students are pretty pleased with themselves" (M=3.61, SD=0.83) had the highest mean and "some students do like the way they are leading their life" (M=3.54, SD=0.85) had the lowest mean. In the confidence dimension, positive identity had the highest mean (M=3.66) and self-worth had the lowest (M=3.58). In the confidence dimension "when I am an adult, I am sure I will have, a good life" had the highest mean which showed that they are optimistic toward the future. In the self-worth dimension, "students are pretty pleased with themselves" had the highest mean due to this fact that they do not have problems such as depression, anxiety, hopelessness. In addition, they are far from some abnormal behaviors such as suicide and inferiority complex that are the sources of many problems in the society. This study is supported by the previous findings of Alberts et al., (2006),  and Bloomquist (2010).
Competence dimension was divided into four dimensions: academic competence, social competence, physical competence, and grade. In the academic competence dimension "Some students can remember things easily" had the highest mean (M=3.68, SD=0.89) and "Some students can almost always figure out the answers" (M=3.54, SD= 0.81) had the lowest mean. In physical competence dimension, "Some students don't feel they can play as well" had the highest mean (M=2.49, SD=0.87) and "Some students don't feel that they are very good when it comes to sports" had the lowest mean (M=2.33, SD=.088). In the social competence dimension, "some students don't have many friends" (M=2.57, SD=.094) had the highest mean and "some students usually do things by themselves" (M=2.35, SD=.091) had the lowest mean.
In the competence dimension, academic competence had the highest mean (M=3.63) and physical competence had the lowest mean (M=2.40).
In the academic competence dimension, students pay attention to their course, and it is important for them because of this reason, "some students remember easily things" had the highest mean. In the physical competence dimension, "Some students don't feel they can play as well" had the highest mean because they had high self-confidence and their motivations were high to start new activities. In the social competence dimension, "some students don't have many friends" had the highest mean due to this fact that perhaps they were shy or depressed and other students had to encourage them to communicate and participate in group activities. Regarding grade achievement, the highest percentage of the grade was related to the "half B and half A", grade 48.8 followed by "mostly B", 18.5 percent. Most respondents had good grades which is very helpful in increasing their self-confidence, finding better jobs in the future, and continuing their studies in higher levels or abroad.
Character dimension had four parts, namely: personal value, social conscience, value of diversity and conduct behavior. In the personal value dimension "accepting responsibility for my actions when I make a mistake or get in trouble" had the highest mean (M=3.94, SD= 0.77) and "doing what I believe is right, even if my friends make fun of me" (M= 3.79, SD= 0.86) had the lowest mean. In social conscience dimension "speaking up for equality (everyone should have the same rights and opportunities" (M= 3.82, SD= 0.84) had the highest mean and "giving time and money to make life better for other people" (M= 3.47, SD= 0.84) had the lowest mean. The results showed that people tended to have freedom of speech and they wanted to participate in decision-making in the society because of this speaking up for equality had the highest mean. In the value of diversity dimension "respecting the values and beliefs of people who are of a different race or culture than I am (M= 3.96, SD= 0.78) had the highest mean and "knowing a lot about people of other races." (M=3.59, SD= 0.78) had the lowest mean. The findings showed that respecting people's difference had the highest mean because Malaysia is a multicultural country and people, especially the youth, need to respect each other's religion and culture. In the conduct behavior dimension "Some students usually do not do things that get them in trouble" (M=3.65, SD= 0.91) had the highest mean and some students hardly ever do things they know they shouldn't do" (M= 3.49, SD= 0.78) had the lowest mean. In the character dimension, personal value had the highest mean (M=3.88) and conduct behavior had the lowest mean (M=3.59). As shown in the results, "students do not do things that get them in trouble" had the highest mean because students wanted to behave as an educated person; therefore, they tried to respect other people, rules and regulations. The findings of the current study are supported by the previous studies of ( Alberts et al., 2006;Bloomquist, 2010;Lerner & Lerner, 2005).
In the caring dimension "when I see another person who is hurt or upset, I feel sorry for them" had the highest mean (M= 4.03, SD= 0.84) and "I feel sorry for other people who don't have what I have" (M= 3.56, SD= 0.98) had the lowest mean. The results showed that feeling sorry for the person who is hurt or upset had the highest mean because youth are caring about the other people, their feelings and their life is important for them. The findings are supported by previous studies (Bloomquist, 2010;Lerner, 2010).
Connection is divided into four parts, namely family, community, university and peer connection. In family connection dimension "my parents give me help and support when I need it" (M= 4.49, SD= 0.66) had the highest mean and "talk to your parents if you have an important concern about drug, alcohol, sex" (M= 3.02, SD= 1.45) had the lowest mean. In the community dimension "I'm given lots of chances to make my town or city a better place to live (M= 3.71, SD= 0.90) had the highest mean and "In my town or city, I feel like I matter to people" (M=3.32, SD= 0.87) had the lowest mean. 26% of the respondents sometimes talk about drug, alcohol, and sex to their parents and 21.5% of the respondents never talk about these issues to their parents. In the University dimension "I care about my University that I go" (M=3.94, SD= 0.82) had the highest mean and "my lecturers really care about me." (M= 3.45, SD= 0.80) had the lowest mean. In the peer dimension, the highest mean was related to the statement "My friends care about me" (M= 3.89 SD= 1.07) and the lowest mean was related to the statement "I trust my friends" (M=3.66, SD=1.05). In the connection dimension, family had the highest mean (M=4.05) and community had the lowest (M=3.47).
In the family dimension, "family give support to their children when they need it" had the highest mean because family usually cares and supports its members, however, in serious problems such as drug usage, alcohol usage and sexual issues, people do not usually talk in the family, maybe it is because of culture, conservative beliefs, or parents' lack of enough knowledge. In the community dimension, it became clear that younger people feel more responsible to build the society in the future and develop their country. One quarter of the respondents talk to their parents about sex, alcohol and drug usage. Family should be a safe place to talk in order to avoid serious problems. In the University dimension, students care about their University and field of study. In the peer dimension, "my friends care about me" had the highest mean which is a sign of good relationship with their friends. Such deep relationships prevent some serious disorders such as depression, anxiety and so forth. The findings of the current study support previous findings of the investigation on the positive youth social development (Benson et al., 2006;Lerner & Lerner, 2005;.

Measurement Model
Measurement model displays how the latent variables are measured in terms of the observed variables, and it describes the measurement properties of the observed variables.

Convergent Validity
As a general rule of thumb, the (standardized) outer loadings must be 0.708 or higher (Hair et al., 2017). Indicators with very low outer loadings (below 0.40) should always be removed from the scale (Hair et al., 2011). Commonly, indicators with outer loadings between 0.40 and 0.70 should be considered for removal from the scale only when omitting the indicator leads to a substantial increase in the composite reliability and AVE (Henseler et al., 2009). Table 1 shows the outer loadings of all items for all variables in the initial and modified measurement model. According to these results, all are outer loadings except the following items: in Confidence indicator SW3, in caring indicator, CAR8,9, in positive identity indicator, PI4,5,6 in gratification of using Internet indicator GRAT3,4,7,12,13,17, in problems of using Internet indicator PRO2,6,7,8,9,10,11,12,13,14,15,16,17,18,19 were eliminated from initial measurement model due to low loading factor which were less than 0.5, that confirmed their low contribution to related constructs. Regarding to the findings of this study, the Composite Reliability (CR) ranged between 0.53 and 0.93 and AVE ranged between 0.254 and 0.71. In addition, to measure collinearity at the indicator level, the variance inflation factor (VIF) is used. Results indicated that all values are below 5, which indicates that there is no collinearity issue.

Discriminant Validity
Discriminant validity demonstrates the extent to which a construct is empirically distinct from another construct (Hair, Hult, Ringle, & Sarstedt, 2021). The Heterotrait-Monotrait (HTMT) ratio of correlation is a new criterion for measuring discriminant validity in PLS -SEM models (Henseler et al., 2014). For Hair et al. (2010) the HTMT value has to be less than 0.85, meaning that two constructs were distinct. In this study all the variables demonstrated that the measurement model used meets and exceeds the requirements for establishing discriminant validities (Table 2).

Path Model for Second Order Factor
Since perceived value was second order latent variables, in order to evaluate the significant contribution of all first order latent variables were investigated using bootstrap approach 2 (Hair, Hult, Ringle, Sarstedt, et al., 2021). The first stage produces parameter estimations for the components and computes their factor scores. The factor scores for school, family, peer, community (connection), positive identity, self-worth (confidence), physical, academic and social competence and grade (competence), social conscience, conduct behavior, value of diversity and personal value (character) were then used as the scores to compute the higher-order. Hair et al., (2006) suggest that a higher-order construct should be assessed in a similar manner as in the lowerorder construct structure. Therefore, reliability, discriminant and convergent validity were checked again against acceptable threshold values.
Outputs from this stage showed that school, family, peer, community had significant loading on connection (p<0.001). Positive identity, self-worth had significant loading on confidence (p<0.001). Physical, academic and social competence and grade had significant loading on competence (p<0.001). Social conscience, conduct behavior, value of diversity and personal value had significant loading on character (p<0.001).

Path Model Using Bootstrap
The path coefficients obtained from this model are consistent. The respective confidence intervals can be obtained by bootstrapping (Streukens & Leroi-Werelds, 2016). This step also provides estimates for indirect and total effects. Some inputs have impact on positive youth development such as the relationship between gratification of using Internet and caring is positive and significant (β= 0.393, p<0.05). Significant and positive relationship is found between gratification of using Internet and character (β= 0.282, p<0.05). The relationship between gratification of using Internet and competence is negative and significant (β= -0.333, p<0.05). There is positive and significant relationship between gratification of using Internet and confidence (β= 0.443, p<0.05). Finally, yet importantly positive significant relationship belongs to gratification of using Internet and connection (β= 0.266, p<0.05). There is no significant relationship between income and four dimensions of positive youth development: caring, connection, character, and competence except for confidence, which is positive and significant relationship (β= 0.241, p<0.05). There is no significant relationship between income and gratification of using Internet (β= 0.241, p>0.05). There is no significant relationship between problem in using Internet and five dimensions of posi-tive youth development (5Cs) (p>0.05), however this relationship with gratification of using Internet is positive and significant (β= 0.161, p<0.05) (Table 4). The change in the R 2 value while a particular independent construct is eliminated from the model can be used to evaluate whether the omitted construct has a basic influence on the dependent construct. This measures indicator the f 2 or effect size. Recommended guideline for assessing effect size are: f 2 ≥ 0.02, f 2 ≥ 0.15 and f 2 ≥ 0.35, respectively representing small, medium and large effect size of exogenous construct (Cohen, 1988). According to the result of f 2 indicated that effect size of exogenous construct for gratification of using Internet indicate effect size for competence (f 2 =0.121), connection (f 2 =0.075), and character (f 2 =0.085), is small, for confidence (f 2 =0.242), and caring (f 2 =0.181), is medium. The highest effect size belonged to confidence and caring. For income, the highest effect size belonged to confidence (f 2 =0.013), and for problem in using Internet the height effect size belonged to gratification of using Internet (f 2 =0.026) (Table 5).

Coefficient of Determination (R 2 )
In addition, problems of using Internet and income are able to explain 16% for caring, 10%, for character 0.08%, for competence 10%, 20% for confidence, for connection 0.07% of the variance of youth positive development. Furthermore, the effect size of gratification of using Internet is 0.02% of the variance of youth positive development (Table 6).

Predictive Relevance Q 2 of Structural Model
An important aspect of structural model is its capability to determine the predictive relevance of the model. Blindfolding procedure was employed to establish crossvalidated redundancy measurement for each construct. The results revealed that the Q 2 value of gratification of using Internet (0.009), caring (0.072), character (0.029), competence (0.037), confidence (0.081) and connection (0.028) are larger than zero, recommending that the independent construct have predictive for mediator and dependent construct under the condition in this study Hair et al., (2011) (Table 7).  RQ 1: What is the mediating effect of Internet gratifications use in relationship between 'problems of using Internet', and 'income level', with the selected dimensions of positive youth development?

Mediation Test of Internet Use Gratification
There is a mediatory effect between Internet use gratification, the income level, and problems of using Internet, with five dimensions of youth positive development. In general, the mediatory effect of Internet use gratification variables in the path model is significant for problems of using Internet, and the 5Cs. There is no mediatory effect for income and 5Cs with Internet use gratification of mediatory variables (p>0.05).
There is a mediatory effect between problems of using Internet, Internet use gratification, and caring (β=0.063, p<0.05); this relationship is positive. Significant and positive relationship belongs to problems of using Internet, Internet use gratification and connection (β= 0.043, p<0.05). There is mediatory effect between problems of using Internet, Internet use gratification, character and confidence (β=0.045, p<0.05), (β=0.071, p<0.05) respectively. The relationship between problems of using Internet, Internet use gratification, and competence is negative and significant (β=-0.054, p<0.05) ( Table 8).

Moderator Effects of Field of Study
According to partial least square analysis, there are no significantly differences between the path coefficient between science and social science for the entire path coefficient (Table 9).

FINDINGS AND DISCUSSION
The present study mainly concentrated on perceiving the direct effects of using Internet on positive youth development and the indirect effects of using Internet on positive youth development through Internet use gratification.
The first main outcome in terms of gratification of using Internet, is that the highest overall mean belonged to cognitive dimension followed by escape and social integration dimension. It explains that students use Internet to get information and escape from the routine at the same time, and get in touch with others.
Regarding problems of using the Internet, the highest mean score was related to slow Internet connection followed by overload of information on the Internet. These findings are consistent with results of previous studies that reported the main problem in using the Internet is related to the slowness in downloading (Mui et al., 2002;Riahinia & Azimi, 2008;Suryani, 2007). Also it is consisted with the study of Azlan et al., (2020) who detected that the main barrier in using Internet is that not all participants had enough money for a stable Internet connection even with the increasing growth of technology in Malaysia, Internet speed and connection quality is not the same everywhere.
Positive youth development was divided into five categories (5Cs) namely confidence, character, connection, caring, and competence. In confidence dimension, positive identity had the highest mean, indicating that the participants have a positive attitude about their lives and future. In the competence dimension, academic competence had the highest mean because the participants are students and their academic achievement is very important to them as it directly affects their professional future. Regarding character dimension, the highest mean belonged to personal value followed by social conscience; indicating that sense of responsibility is among the most important marks for the future. In the caring dimension, "when I see another person who was hurt or upset, I feel sorry for them" had the highest mean which means that people care about others and their fate is important to them. In the connection dimension, the highest mean score belonged to "family", indicating that parents mostly used Internet for communicating when they are not home. The present results are supported by the previous findings of Alberts et al., (2006), Bloomquist (2010).
The second main finding is that positive youth development (PYD) has significant relationships with all their sub-dimensions. In general, Internet use gratification has significant relationships with the 5Cs which indicate that students use Internet and feel gratified and it directly influences positive development. Aburub & Alnawas (2019) also found similar results stating that cognitive gratification has a great influence on adopting mobile learning while hedonic gratification had almost no effect. Learning in the present study is attributed to students' academic performance. Rokito et al., (2019), studied the role of using Facebook on gratification discrepancy and strength of habit and concluded that the influence of habit and gratification satisfaction can determine the amount of using Facebook in a day.
The third major finding is that the mediatory effects of Internet use gratification are significant for problems of using Internet and the 5Cs. Internet use problems (poor connection, having no signal, and getting disconnected) influenced Internet use gratification and PYD. Moreover, there was no mediatory effect for the income level and the 5Cs. Nowadays, people use the Internet to a great extent as it has become an essential part of people's daily life. People are willing to pay the cost of Internet because of their online activities and income does not have any effect on their usage. However, Hanchard (2016) found that income and education directly influence using social media.
The Fourth finding is that there is no significant difference between the path coefficient between science and social science for the entire path coefficient. Therefore, field of study does not have any effect on Internet use gratification and the 5Cs. This result are not consisted with the study of Ayub, Hamid, and Nawawi (2014), in which students' field of study is highly influential in using Internet for educational purposes.
Also, Samani and Guri (2019) found that finding tourist information in the website will provide appropriate gratification. Internet use can be properly explained through the theory of Uses and Gratification as users' sense of gratification has a direct effect on PYD supporting the presupposition that people consciously use the media and once they are satisfied they will use it again. The audience uses the media for communicative purposes and this process in influenced by various social and psychological factors. Musa, Azmi and Ismail (2016) also reported that most people use social media for the following reasons: communication, passing time, fun, finding information, socializing, self-expression, education, and supervision. Moreover, according to Teng and Joo (2017), people use the media for their own personal reasons. Samani and Guri (2019) also concluded that users were gratified once they found the information they needed on the Internet.

CONCLUSION
The present study was conducted as an attempt to find out any relationship among income and problems of using the Internet with gratification of using Internet as mediation in relationship with positive youth development among Malaysian youth based on their field of study factor. The current investigation had five main findings; 1) the mediatory effects of Internet use gratification are significant for problems of using Internet and the 5Cs 2) the mediatory effects of Internet use gratification are not significant for income level and the 5Cs; 3) there is no significant difference between field of study and Internet usage gratification and 5cs. 3) There is no significant relationship between problem of using Internet, the 5Cs; 4) The relationship between problems of using Internet and Internet use gratification is positive and significant; 5) there is no significant relationship between income and 5 dimensions of PYD except for confidence; 6) the relationship between income and gratification of using Internet is not significant.

RECOMMENDATION
Policy planners in the society need to equip students with proper plans to grow their level of gratification and positive development. In other words, increasing positive development is the prerequisite of having a developed nation. At the same time, people need to be aware of negative developments like pornography and cybercrime. Through enhancing the 5Cs among the younger generation of a society, the amount of abnormal and risky behavior will decrease dramatically. According to Manrique-Millones et al., (2021), with the help of positive youth development as a promising model, young people of Peru and Colombia reduced their alcohol usage, drug consumption, street crimes, and attempting suicide.
It is suggested to integrate quantitative and qualitative practices in order to get a more inclusive view about using Internet and PYD. A mixture of survey and interview with policy makers as well as youth is also suggested to obtain more comprehensive results; in this way youth are also invited to have a role in their future. Holding proper workshops to increase people's awareness on negative influences of using Internet deems necessary. Youth need to be prepared when they face problems. Once the young generation is equipped with sufficient knowledge to use the Internet properly, a better future from every possible aspect will await them.