In 2018, road accidents were found to be the 8^th leading cause of death among the world’s population, whereas 1.35 million deaths were found annually. Most involved children and young adults were aged 5 to 29 years old [1]. Loss from road accidents has driven road accident prevention programs in many countries through the Global Ministerial Conference on Road Safety [2]. Additionally, the World Health Organization (WHO) has declared a Decade of Action for Road Safety from 2011 to 2020, intending to reduce road accident fatalities by 50% in 2020. However, this goal has yet to be accomplished [3]. Since 2020, WHO has called on member states to implement continuous road accident prevention programs in line with Target 3.6 of the Sustainable Development Goals (SDGs), aiming to halve the number of global deaths and injuries from road traffic accidents by 2030 [4].

In Thailand, despite efforts to improve road safety during the Decade of Action for Road Safety [5], the number of deaths due to road accidents decreased by 10% from 30.4 per 100,000 population in 2019 to 27.2 per 100,000 population in 2020. Most fatalities involved males aged 20 to 24 due to motorcycle riding accidents [6]. Considering specific areas of Thailand classified by Health Regions (13 units), Health Region 4 was the top 5 ranking over the past three years, from 2019 to 2021. Amidst all eight provinces in Health Region 4, Phranakhon Si Ayutthaya Province was the top 3-ranked with the highest number of road accident fatalities among people aged 20 to 24 riding motorcycles during the duration mentioned [7]. The road accident victims in Phranakhon Si Ayuttaya Province were primarily higher education students aged 18 to 22 who used motorcycles as their primary vehicle because the largest university was located in the central area of ancient historical sites, and most roads were narrow [8, 9]. From 2019 to 2021, 79 cases of motorcycle accidents and eight deaths were reported. Causes included crashing with a car (35.0%), high-speed riding (24.0%), failure to wear a helmet (22.0%) and riding a motorcycle in a reverse direction (19.0%) [10].

The comprehensive enforcement strategy for preventing road accidents included wearing a helmet, not riding a motorcycle while drunk, having a safe motorcycle, having a motorcycle license and riding within the speed limit [11]. These measures were developed based on the concept that road accidents are caused by people’s behaviors [12]. Related research has revealed that most studies emphasized specific causes of road accidents, particularly in helmet use behavior [13-18]. Few studies have applied the health behavioral theory [15, 17]. Theories can guide in identifying the influential factors of a particular person and provide information to understand behaviors and behavioral change [19]. The Theory of Planned Behavior (TPB), developed by Ajzen [20], is one of the most applied theories in the social and behavioral sciences. TPB was intended to predict and explain a wide range of intentions and health behaviors, composed of four core components: attitude, subjective norms, perceived behavioral control (PBC) and behavioral intention (BI) [21].

Therefore, this study applied TPB as a research conceptual framework to study behaviors and factors predicting safe motorcyclist riding among university students in Phranakhon Si Ayutthaya Province. According to TPB, students’ safe motorcyclist riding behavior depends on their attitudes and subjective norms influenced by family members, friends and their favorite group of superstars or celebrities participating in a road safety campaign, PBC and BI. These factors together can affect the target behaviors: safe motorcyclist riding behavior. The study’s findings can be beneficial for planning the prevention and promotion program to enhance students’ safe motorcyclist riding behaviors.

The research findings were divided into two parts. The first part described the general characteristics of respondents, while the second presented the factors predicting safe motorcyclist riding behavior. Bivariate statistics analysis was used to present the results of factors associated with safe motorcyclist riding behavior, and then the significant variables were assigned to multivariate analysis.

Altogether, 54.6% of participants had a level of safe motorcyclist riding behavior covering five main aspects: helmet use, not being drunk while riding a motorcycle, having a safe motorcycle, having a motorcycle license and riding within the speed limit. One-half of respondents demonstrated safe riding behavior. These behaviors were caused by an average of four years of riding experience, with nearly 50% of respondents riding for five or more days weekly. Related research found that a higher number of years of riding experience equaled three years and over was a protective factor against road traffic injuries among motorcyclists [26]. In addition, having a motorcycle license influenced safe motorcyclist riding behavior. These caused the regulations regarding obtaining a motorcycle license from authorized offices of the Department of Land Transport Office, Ministry of Transport, Thailand. Respondents with a motorcycle license indicated that they passed a written examination and practiced a riding motorcycle skill test. The approach stated that they gained sufficient experience to perform safety behaviors.

The study revealed three significant variables based on a critical construct of TPB influencing safe motorcyclist riding behavior among university students. TPB states that individuals will perform a behavior if they intend to do so, and the success of the behavior depends on their intention. The critical variable influencing the safety behaviors based on the coefficient of stepwise multiple regression analysis was BI. BI refers to an individual's readiness or effort to exhibit a behavior in a favorable direction [20]. BI drove them to be determined and enthusiastic in demonstrating behavior and making an earnest effort to perform it. Therefore, individuals with a higher intention are likelier to engage in the behavior [27]. Moreover, most respondents had not experienced any accidents over the past year (86.8%). It indicated that experiences serve as a mediating variable influencing the relationship between intention and actual behavior. The past automatically triggers present behavior [28]. Most respondents also had no motorcycle license (81.3%), so they had increased vigilance and intention to drive safely to avoid accidents leading to legal consequences.

Referring to PBC, the statistically significant variable influenced safe motorcyclist riding behavior. It could be explained that most respondents drove motorcycles daily (32.2%), had an average riding experience of 4.4 years, and never drove when drunk. Consequently, they perceived they possessed the ability and sufficient skill to control the motorcycle while riding. Additionally, most rode a motorcycle for a short distance, averaging 10.2 km. The surrounding areas, from the respondent’s house to the university areas, were limited to no more than 80 km per hour because Phranakhon Si Ayutthaya Province, the capital of Siam (the name before reverting to the Kingdom of Thailand) in the mid-13th century, comprised ancient historical sites [11].

A subjective norm reflects society's view of a particular behavior, whether they approve or disapprove of it [29]. The study found friends were significantly influenced by the target behavior, safe motorcyclist riding behavior (β = 0.137, p = 0.019). The respondents, university students with an average age of 20.0 years, spent their daily lives with friends of the same age group. They commonly acquired a sense of belonging. Therefore, friends influenced them to adopt similar behaviors, both positive and negative, such as academic performance [30], smoking [31] and alcohol consumption [32]. Knowledge-related traffic rules were not influenced by safe motorcyclist riding behavior, causing riding a motorcycle to become routine behavior.

The study's limitation was due to a specific setting, focused on the largest university in Phranakhon Si Ayutthaya Province. Although the survey used probability sampling to select the sample, the generalizability to other locations might be limited. On the other hand, the data collection approach assigned by a self-administration questionnaire was another study limitation. The respondents might have been biased in their responses to a socially acceptable question rather than being truthful. Respondents may have been distracted or rushed to finish the questionnaire due to a friend's loud chatter after completing it.

Most respondents exhibit a safe level of motorcyclist riding behavior. The possession of a motorcycle license and the fundamental construct of TPB, including PBC, subjective norms influenced by friends and BI, were all significant predictors of a motorcycle's safety riding behavior. Therefore, educational institutions should create and maintain an up-to-date registration database for their students with motorcycles and driving licenses. These institutions should also strive to make it easier for students to obtain a motorcycle license by increasing accessibility to government services within the university area. Student clubs focusing on safe motorcyclist riding behavior should be established, where students can develop their leadership skills, analyze the causes of motorcycle accidents among university students, and plan to solve the problems. These clubs should also organize multiple campaigns and activities to promote safe motorcyclist riding behaviors. The quasi-experimental research should also be conducted by designing the health education program following the predicted variables for achieving a motorcyclist's safety riding.

PT wrote the initial manuscript draft, coordinated the data collection, developed the data entry system, performed statistical analysis and screened the identified citations. BS participated in the study design, supervised data cleaning and analysis, critically revised and edited the manuscript and contacted the journal. KM contributed to the manuscript review. All authors have read and approved the final version of the manuscript.