The “Anti-Vax” Movement: A Quantitative Report on Vaccine Beliefs and Knowledge across Social Media

Principal Investigator(s):  View help for Principal Investigator(s) Staci L Benoit

Version:  View help for Version V1

Name File Type Size Last Modified
Belief Descriptives.xlsx application/vnd.openxmlformats-officedocument.spreadsheetml.sheet 159 KB 08/03/2020 02:44:PM
Knowledge Descriptives.xlsx application/vnd.openxmlformats-officedocument.spreadsheetml.sheet 20.8 KB 01/17/2019 12:24:PM
Survey Summary.xlsx application/vnd.openxmlformats-officedocument.spreadsheetml.sheet 323 KB 01/17/2019 09:27:AM

Project Citation: 

Benoit, Staci L. The “Anti-Vax” Movement: A Quantitative Report on Vaccine Beliefs and Knowledge across Social Media . Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor], 2020-08-03. https://doi.org/10.3886/E120505V1

Project Description

Summary:  View help for Summary This cross sectional research explored the relationship between the spread of information regarding vaccines and social media use. A sample of 2515 people over the age of 18 around the world completed the survey via a link found on Twitter, Facebook and Instagram. A series of questions on vaccine knowledge and beliefs were compounded to create an individual's "knowledge score" and a "belief score". Knowledge scores were ranked from low knowledge to high knowledge with increasing scores. Belief scores were ranked from belief in myths to disbelief in myths with higher scores. This score was then analysed, using a Welch test and post hoc testing when applicable, across demographics and questions relating to social media use.


Scope of Project

Subject Terms:  View help for Subject Terms vaccines; social media; anti-vaccination
Data Type(s):  View help for Data Type(s) survey data

Methodology

Sampling:  View help for Sampling Snowball sampling of social media users was used. Snowball sampling was used to help perpetuate the survey through social media, where social media is the quality of referral. The study population was aimed at being as demographically diverse as possible among people who use social media. 
Data Source:  View help for Data Source Subjects were recruited through the three largest social media platforms (Facebook, Twitter, Instagram) via a shareable web link and asked to consent before completion of the research survey.  A web survey was chosen due to its ease, speed, cost, and ability to obtain a geographically diverse population (Fricker and Schonlau, 2002). 
Scales:  View help for Scales The first half of the survey consisted of demographics and questions pertaining to use of social media and its relation to vaccine information. The latter half of the survey had six questions relating to vaccine knowledge and six questions relating to vaccine myths.           The six vaccine knowledge questions were scored on a two point scale. Questions were scored by awarding two points for the answer of belief in the vaccine statement, one point for uncertainty, and zero points for the answer of disbelief in the statement. All questions were then totaled for a score on a 12 point scale. Higher values suggesting adequate vaccine knowledge and lower values suggesting inadequate vaccine knowledge. This score could then be appropriately analyzed.           The six vaccine belief questions were scored on a two point scale. Two points were given for the answer choice “disbelief in the vaccine statement”, one point for uncertainty, and zero points for the answer of belief in the statement. All questions were then totaled for a score on a twelve-point scale. Higher values of disbelief in common myths, whereas lower values indicated a belief in common myths. This score could then be appropriately analyzed.
Weights:  View help for Weights All data analysis was conducted using IBM’s SPSS. Significance testing was performed using the Welch test. This test was chosen based on the negatively skewed data distribution with non-homogeneity of variances and sample sizes (Fagerland and Sandvik, 2009). The Welch test has historically been shown to better control Type 1 error for these parameters compared to other tests (Tomarken and Serlin, 1986). Post hoc analysis was completed with Games Howell due to its robustness and utility in non-normal distributions (Hilton and Armstrong, 2006). 

Related Publications

This study is un-published. See below for other available versions.

Published Versions

V2 [2020-12-23]

Export Metadata

Dublin Core

DDI 2.5

Report a Problem

Found a serious problem with the data, such as disclosure risk or copyrighted content? Let us know.

This material is distributed exactly as it arrived from the data depositor. ICPSR has not checked or processed this material. Users should consult the investigator(s) if further information is desired.