Background
   Research over the past few decades has shown that smoking has become a possible fatal risk to health. According to the World Health Organization (WHO), one person dies of smoking-related diseases every ten seconds worldwide. In 1994, the U.S. Department of Health and Human Services published the report named "Preventing Tobacco Use Among Young People", pointed out six major conclusions：1) nearly all first use of tobacco occurs before high school; 2) most adolescent smokers are addicted to nicotine and report that they want to quit but are unable to do so; 3) they 
experience relapse rates and withdrawal symptoms similar to those reported by adults; tobacco is often the first drug used by those young people who use alcohol, marijuana, and other drugs; 4) Adolescents with lower levels of school achievement, with fewer skills to resist pervasive influences to use tobacco, with friends who use tobacco, and with lower self-images are more likely than their peers to use tobacco; 5) Cigarette advertising appears to increase young people's risk of smoking by affecting their perceptions of the pervasiveness, image, and function of smoking; 6) Communitywide efforts that include tobacco tax increases, enforcement of minors' access laws, youth-oriented mass media campaigns, and school-based tobacco-use prevention programs successfully reduce adolescent use of tobacco.

     WHO, The U.S. Centers for Disease Control and Prevention (US CDC), Canadian Public Health Association(CPHA) collaborated to develop the Global Tobacco Surveillance System(GTSS),  which aims to enhance country capacity to design, implement, and evaluate tobacco control interventions.
      The GTSS includes the collection of data through four surveys: the Global Youth Tobacco Survey (GYTS), Global School Personnel Survey (GSPS), Global Health Professions Student Survey (GHPSS), and Global Adult Tobacco Survey (GATS). GYTS focuses on youth aged 13-15 and collects information in schools. GSPS surveys teachers and administrators from the same schools that participate in the GYTS. GHPSS focuses on 3rd- year students pursuing degrees in dentistry, medicine, nursing, and pharmacy. GATS is a nationally representative household survey that monitors tobacco use among adults aged 15 years and older.

        In 2004, Taiwan HPA cooperated with the US CDC, adopted the GYTS in Taiwan. It is a survey on tobacco use among junior and senior high school students. Initially, GYTS was administered for junior high school and senior/ vocational high school students on a rotational basis. To provide more timely results for policy reference, the GYTS has been conducted annually since 2011 with entire country and city-county representative samples of junior high school and senior/vocational high school students. To better reflect the changes in prevalence rates of the health behaviors, the annual survey changed to be a biennial survey since 2019. GYTS collected data such as tobacco use behavior, perceptions of and attitude toward tobacco hazards, exposure to secondhand smoking, and any changes and tendencies as references for relative authorities to plan and evaluate the campaign of preventing tobacco hazards on campus.

● To understand the tobacco use behavior, perceptions and attitude among youth aged 12 to 18 and the implementation of smoking bans on campus in every county and city.
● To compile data of tobacco use behavior, perceptions and attitude among youth, and any changes and tendencies at county / city / national levels, which can constitute concrete evidence for HPA and local health bureaus to follow up and monitor the implementation and effectiveness of intervening and preventing tobacco hazards from young people.
● To implement International Collaborative Research Project on Global Youth Tobacco Survey, increase Taiwan’s publicity and to promote academic exchanges at international level. 

Sampling Design
    The population for the survey was students in junior high schools, local senior high schools, vocational schools, and 1st to 3rd years in junior colleges enrolled in the survey year. However, the administrative costs and the effectiveness of analysis in the future were taken into account. Therefore, the total number of sample students in every county and city was predetermined (about 800 to 2000) and then sampling was conducted. During the process of selecting sample schools, schools were first categorized into regular schools and evening schools. Secondly, the estimated number of sample schools in each category was calculated. Thirdly, the total number of sample students was estimated. Then, the result of the formula below served as the sampling interval: ‘Total Number of Students divided by Estimated Number of Sample Students’. Then the system randomly selected sample schools with such sampling interval. Next, sample classes in each sample school were selected according to (in the following order) types of classes, grades and disciplines. As a result, the system randomly sampled three to six classes. The students in these classes, about 40,000 to 50,000, were selected as the survey samples.

Survey Methods
      A self-administered questionnaire was answered by students anonymously. Three to six staff from the local health bureaus, after receiving a one-day training of hosting a questionnaire survey provided by HPA, coordinated with sample schools and decided the conducting dates and times. The survey was conducted in the classroom of the sample schools. If local health bureaus were short of human resources, HPA would support and designate its staff. To avoid sample students from discussing with each other about the questionnaire, which might have affected the result, all sample 
students in each school took the survey simultaneously. To make sample students respond to the questionnaire willingly and truthfully, all school staff were not present on the site during the survey. Staff from local health bureaus or HPA provided sample students with standardized instructions. This included explaining the purpose of the survey, the importance of their representation of the selected sample students and how to complete the questionnaire. They were assured anonymity in the survey, that the survey’s sole purpose was for statistical and analytical purpose, and they were guaranteed confidentiality of personal details. In order to enhance the confidentiality of the data and increase the data-scanning speed, open-ended questions were avoided at the designing stage and answers to be circled and filled in black by respondents with 2B-lead pencils provided by HPA. Since the COVID-19 pandemic in 2021, GYTS collected onsite and online data to allow eligible students in remote learning settings. Eligible students could read the standardized instructions instead of staff from local health bureaus or HPA provided and explained.

Survey Result
Results of the survey are published in the format of survey reports, press releases and conference or journal paper. Some of the statistical indicators are collected and can be accessed from the Statistical Yearbook of Health Promotion.