non-existence of true vaccines, ineffective vaccination in certain races, being dangerous in elderly people, the ability of vaccines to alter human DNA or to control the behavior of persons, and up to the death of individuals after receiving vaccines [25]. In a large study the knowledge, attitude, and practices of 215,731 participants from 45 countries were assessed. The estimated overall correct answers for knowledge, good attitude and good practice were 75%, 74% and 70%, respectively [26]. Therefore, a serious collaborative initiative has started to remove the misleading claims to counteract the spread of misinformation about COVID-19 vaccines in social media platforms (i.e., Facebook, Twitter, and You Tube) [25]. In this study we aimed to determine opinions and attitudes toward COVID-19 vaccines through analyzing reactions and comments of social media users to the COVID-19 posts released by health authorities. 2. Materials and Methods We conducted quantitative content analysis to analyze the attitude of social media users to COVID-19 vaccination [27]. We analyzed social media posts from official health institutions regarding COVID-19 vaccination and associated social media users’ comments for 24 countries. Researchers identified the main social media platforms, number of social media users, and the social media pages of the official health institutions in each country via Statista website [28]. Based on this information, and by considering the population of each country, the sample size of included posts and comments from each country was determined. For each country, the population number was determined and then after conducting a power analysis, sample size was determined in the form of the minimum required number of posts to be collected for this study. We determined the number of comments we should include from each social media platform based on the percentage of usage of each social media platform in each country [28]. Int. J. Environ. Res. Public Health 2022, 19, 5737 4 of 14 Quantitative content analysis was used to fulfil the goal of this study. Content analysis is a research method to analyze verbal or visual communication/message [29]. There are two forms of content analysis: qualitative and quantitative. Quantitative content analysis is a branch of content analysis that analyzes messages by quantifying the occurrence of words, expressions, phrases, and so on [30]. The process of a quantitative content analysis begins with determining the different expressions to be extracted from the text. A coding process preceded this step, and a codebook was designed and used to quantify the occurrence of the words that provide inference to the needed expressions. Since this process has a human factor (i.e., during the coding process), multiple coders coded same content to ensure the reliability of the coding process. In this study, two waves of coding were carried out on the collected data. A codebook was created as the main instrument used in this study. The codebook aimed to code social media users’ comments and reactions on COVID-19 vaccine posts identified in the previous step. In general, when selecting posts on social media, researchers analyzed this same post on the different social media platforms where it was posted. For Facebook posts, the data collected was screenshot/link to the post, date, time, location, source, language, number of comments, and number of reactions to the post. 2.1. Codebook 2.1.1. Position of the Comment “With vaccination” meant that the comment supported the vaccine’s existence and use, “Against vaccination” meant that the comment was refusing the vaccine in any means whether industry or intake, “Neutral” meant that the comment was simply a comment, not showing a directional attitude. 2.1.2. Tone of the Comment “Serious” meant that the comment was literal in its meaning, “Humorous” meant that the comment was funny “Sarcastic” when the comment has the character of sarcasm, “Opinion” meant that the comment was explaining the person’s point of view about the vaccine. To analyze the comments, the codebook included the following codes: post being analyzed, comment screenshot, tone of the comment (serious, humorous, opinion), opinion position of the comment (with vaccination, against vaccination, neutral), and the number of reactions to the comment. For Twitter, the following codes were included for each tweet: screenshot, date, time, location, source, language, number of replies, number of retweets, number of quote tweets, and number of likes. The following codes were used to analyze the replies: tweet ID, reply ID, screenshot, tone of the reply (serious, humorous, opinion), and comment position (with vaccination, against vaccination, neutral). Before the start of the coding process, all the data collectors were trained on how to code comments. During the training, data collectors got a deep understanding of the concepts included in the codebook and the spectrum of meanings that can be included under each code. The training also involved sample coding activities of real comments. The training was carried out once all the data collectors had a mutual agreement when coding the sample comments. The tone and position codes were entered during the collection process from the interpretation of the researcher to the language of the comment and