[The National Survey on Addiction Problems in Hungary 2019 (NSAPH): Methodology and sample description].

Objectives: This paper presents the methods and main methodological fi ndings of the third wave of data collection of the National Survey on Addiction Problems in Hungary (NSAPH). In addition to mapping the addiction problems of the Hungarian population and monitoring the general population addiction trends in Hungary, the research formulates objectives for the development of the methodology used in addiction research. The study presents the measurement tools used in the research, the sampling and data collection strategy, as well as the methodological results related to sample access and the reliability and validity of the applied measurement tools used.
Methods: The research was carried out on a nationally representative sample of the Hungarian adult population aged 16-64 yrs (gross sample 1800, net sample 1385 persons). The extent of the theoretical margin of error in the sample is ±2,6%, at a reliability level of 95%. Sample attrition was corrected by matrix weighting by layer categories. The research aimed at studying the different type substance use behaviours (smoking, alcohol use and other psychoactive substances) as well as several behavioural addictions (problematic internet use, problematic online gaming, problematic social media use, problematic gambling, exercise addictions, eating disorders, work addiction, compulsive buying, problematic mobile phone use).
Results: Based on the analysis of reliability and the extent of non-sampling errors within the context of the database we concluded that valid and reliable statements can be formulated on the basis of the research data regarding the current characteristics and patterns of the examined addiction behaviours. At the same time, in case of psychoactive substance use the observed trends of non-sampling errors indicate that the analysis of the changes requires outstanding attention whilst interpreting them; in some cases correction procedures might become necessary during estimating and interpreting the tendencies. This wave of the research was the first which examined the reliability and validity of prevalence data related to behavioural addictions. Based on this analysis it can be said that these data are less exposed to non-sampling errors than the ones related to substance use.