This paper is in the following e-collection/theme issue:

Data Sources and Open Data for Infodemiology (15) Public Health Informatics (93) JMIR Theme Issue: COVID-19 Special Issue (2452) Medicine 2.0: Social Media, Open, Participatory, Collaborative Medicine (1641) Infodemiology and Infoveillance (1349) Development of Instruments and Surveys (86) Infoveillance, Infodemiology, Digital Disease Surveillance, Infodemic Management (1041) Vaccination and Immunization in the Digital Age (374) Electronic/Mobile Data Capture, Internet-based Survey & Research Methodology (832)

Published on in Vol 3 (2023)

Preprints (earlier versions) of this paper are available at https://preprints.jmir.org/preprint/43700, first published .
Using COVID-19 Vaccine Attitudes Found in Tweets to Predict Vaccine Perceptions in Traditional Surveys: Infodemiology Study

Using COVID-19 Vaccine Attitudes Found in Tweets to Predict Vaccine Perceptions in Traditional Surveys: Infodemiology Study

Using COVID-19 Vaccine Attitudes Found in Tweets to Predict Vaccine Perceptions in Traditional Surveys: Infodemiology Study

Authors of this article:

Nekabari Sigalo1 Author Orcid Image ;   Vanessa Frias-Martinez1 Author Orcid Image
Article Authors Cited by (2) Tweetations (3) Metrics

    Original Paper

    College of Information Studies, University of Maryland, College Park, MD, United States

    Corresponding Author:

    Nekabari Sigalo, BSc, MPH, PhD

    College of Information Studies

    University of Maryland

    4130 Campus Dr

    College Park, MD, 20742

    United States

    Phone: 1 301 405 2033

    Email: nsigalo@terpmail.umd.edu


    Background: Traditionally, surveys are conducted to answer questions related to public health but can be costly to execute. However, the information that researchers aim to extract from surveys could potentially be retrieved from social media, which possesses data that are highly accessible and lower in cost to collect.

    Objective: This study aims to evaluate whether attitudes toward COVID-19 vaccines collected from the Household Pulse Survey (HPS) could be predicted using attitudes extracted from Twitter (subsequently rebranded X). Ultimately, this study aimed to determine whether Twitter can provide us with similar information to that observed in traditional surveys or whether saving money comes at the cost of losing rich data.

    Methods: COVID-19 vaccine attitudes were extracted from the HPS conducted between January 6 and May 25, 2021. Twitter’s streaming application programming interface was used to collect COVID-19 vaccine tweets during the same period. A sentiment and emotion analysis of tweets was conducted to examine attitudes toward the COVID-19 vaccine on Twitter. Generalized linear models and generalized linear mixed models were used to evaluate the ability of COVID-19 vaccine attitudes on Twitter to predict vaccine attitudes in the HPS.

    Results: The results revealed that vaccine perceptions expressed on Twitter performed well in predicting vaccine perceptions in the survey.

    Conclusions: These findings suggest that the information researchers aim to extract from surveys could potentially also be retrieved from a more accessible data source, such as Twitter. Leveraging Twitter data alongside traditional surveys can provide a more comprehensive and nuanced understanding of COVID-19 vaccine perceptions, facilitating evidence-based decision-making and tailored public health strategies.

    JMIR Infodemiology 2023;3:e43700

    doi:10.2196/43700

    Keywords

    social media; Twitter; COVID-19; vaccine; surveys; SARS-CoV-2; vaccinations; hesitancy 


    Background

    The implementation of successful COVID-19 vaccine rollouts is essential for COVID-19 to remain under control globally. Although vaccines are essential in the global battle against COVID-19, vaccine hesitancy continues to be a barrier to effective and consistent vaccine rollout programs. According to the US Census Bureau’s Household Pulse Survey (HPS), individuals who reported being hesitant about receiving a COVID-19 vaccine cited concerns about side effects, safety, and a lack of trust in the vaccine or the government [1]. Although the number of vaccine-hesitant individuals continues to decline, the fact that vaccine hesitancy still exists interferes with infection control through vaccination.

    Vaccine hesitancy has been fueled in part by the spread of vaccine misinformation, both in the media and on the web. In fact, the COVID-19 vaccine became a popular topic of discussion among social media users, with many individuals expressing their concerns about taking the vaccine on social media platforms [2]. Amid the new normal of self-quarantine and lockdown, Twitter (subsequently rebranded X) quickly emerged as an important means of COVID-19 communications and discussion [3]. This is in part due to the real-time availability of social media messaging, compared with traditional news reporting methods [4,5]. Twitter users often not only take to the platform to announce their own experiences and opinions about the pandemic but also see Twitter as a source of up-to-date information about the pandemic [6].

    The COVID-19 vaccine conversation on social media platforms has been both beneficial and detrimental to vaccination efforts across the world [7]. Although the exact effect of social media on this unprecedented pandemic is difficult to quantify, there has been a constant battle between facts and misinformation, trust and fearmongering, and hope and anger [8]. Research has shown that social media use plays a role in the low acceptance of vaccines [9,10]. Therefore, studying the public COVID-19 vaccine–related discussion on social media can help researchers better understand attitudes related to the vaccine [9].

    Traditionally, surveys are conducted to understand attitudes related to public health. For example, many studies leveraged surveys to examine COVID-19 vaccine hesitancy and compliance. In April 2020, Ward et al [11] administered 4 web-based, nationally representative surveys to adults in France to identify the reasons why individuals would or would not take the COVID-19 vaccine once it became available. Researchers found that nearly a quarter of the respondents refused to take the COVID-19 vaccine once it was made available to them, citing reasons such as not trusting vaccines in general; concerns about the expedited vaccine development process; and a lack of fear of COVID-19, deeming the vaccine unnecessary.

    In the study by Wang et al [12], a cross-sectional, self-administered survey was conducted to evaluate the COVID-19 vaccine intent among nurses in Hong Kong, China. Researchers found higher rates of vaccine hesitancy compared with vaccine acceptance, with nurses citing concerns about the safety and efficacy of the vaccines. A web-based survey conducted in the study by Ruiz and Bell [13] attempted to identify the predictors of the intent to vaccinate against COVID-19 among Americans. Nearly 40% of the survey respondents refused to be vaccinated against COVID-19. Among the survey respondents, male, older, White, and married individuals and individuals with higher socioeconomic status were more likely to be vaccinated against COVID-19. Researchers also found that Republicans and Fox News viewers were less likely to get vaccinated, whereas individuals who were previously vaccinated for influenza were more likely to be vaccinated for COVID-19.

    Although several studies have examined COVID-19 vaccine attitudes through surveys, to our knowledge, no study has evaluated the ability of Twitter, a newer data source, to predict the attitudes reflected in traditionally collected surveys, such as the HPS. In recent years, researchers have looked at social media as a data source, citing the availability of more readily available data and no- or low-cost data collection efforts [14,15]. Traditional mail, field, and telephone interviewer surveys come with high costs of administration, and even though web-based surveys eliminate the costs of postage, paper, printing, and data entry, the newer web-based survey services may still cost up to thousands of dollars for 1 survey [16]. Although relatively inexpensive compared with traditional surveys, web-based surveys are not always cost-effective [16]. Evaluating the ability of information extracted from social media to predict information found in traditional surveys would suggest whether researchers may use this more cost-effective data source to provide similar rich information to that seen in traditional surveys or whether saving money comes at the cost of losing rich data.

    Study Overview

    The main objective of this study was to examine whether aggregate attitudes extracted from social media can predict vaccine attitudes collected via surveys. We hypothesized that social media may contain attitudes similar to those found in traditional surveys, with the added benefits of more readily available data and no- or low-cost data collection efforts. Predictive models of vaccine attitudes at the metropolitan level can be useful for 2 purposes. First, predictions can be used to identify metropolitan areas where vaccine hesitancy is high and create targeted campaigns to increase vaccination. Second, the relationships between sentiments and emotions and vaccine attitudes can be used to understand human perceptions of vaccines and create effective social media messages for vaccination campaigns. Specifically, we hypothesize that there is a direct, positive relationship between (1) positive sentiments and emotions found in Twitter data and the HPS and (2) negative sentiments and emotions found in Twitter data and the HPS survey.


    Data Collection and Preprocessing

    HPS Data

    In April 2020, the US Census Bureau began releasing a cross-sectional nationally representative survey, the HPS, in an effort to assess the social and economic impacts of the COVID-19 pandemic on American households [17]. The data from this survey were made publicly available in near real time with the purpose of informing federal and state response and recovery planning [18]. The HPS sample was selected using a stratified random sampling method [19]. Data were collected via computer-assisted telephone interviewing, a data collection method in which surveyors use computer software to conduct telephone interviews with respondents [19].

    On January 6, 2021, the US Census Bureau added COVID-19 vaccine–related questions to the HPS with the goal of understanding the factors contributing to vaccine hesitancy and compliance among Americans [20] (Table 1). These questions assessed COVID-19 vaccine receipt, whether respondents received or planned to receive all required doses, intentions to get vaccinated, and reasons why respondents refused to get vaccinated.

    Table 1. COVID-19–related Household Pulse Survey questions.
    ConditionQuestionResponses
    Age >18 yearsHave you received a COVID-19 vaccine?(1) Yes
    (2) No
    Answered “yes” to “have you received a COVID-19 vaccine?”Did you receive (or do you plan to receive) all required doses?(1) Yes
    (2) No
    Answered “no” to “have you received a COVID-19 vaccine?”Once a vaccine to prevent COVID-19 is available to you, would you...(1) Definitely get a vaccine
    (2) Probably get a vaccine
    (3) Be unsure about getting a vaccine
    (4) Probably not get a vaccine
    (5) Definitely not get a vaccine

    Measures of vaccine compliance and hesitancy were assessed for each survey wave overall and in the metropolitan areas in Textbox 1. At the start of the survey period, January 2021, vaccine rollout in the United States had just begun, and with most people unvaccinated at that point, the intent to vaccinate was the only option. For the purposes of this analysis, individuals who answered that they would “definitely get a vaccine” or “probably get a vaccine” once available were considered vaccine compliant, and individuals who answered that they would “probably not get a vaccine,” or “definitely not get a vaccine” once available were considered vaccine hesitant.

    The HPS refers to the data collection cycles as weeks for consistency with earlier phases, even though the cycles actually span a 2-week collection period. For this study, we used the HPS microdata from weeks 22 to 30, which were collected between January 6 and May 25, 2021, with response rates ranging from 6.4% to 7.5% (Table 2).

    Textbox 1. Targeted metropolitan areas for data collection (January to May 2021).
    • New York–Newark–Jersey City, New York–New Jersey–Pennsylvania metropolitan area
    • Los Angeles–Long Beach–Anaheim, California metropolitan area
    • Chicago-Naperville-Elgin, Illinois-Indiana-Wisconsin metropolitan area
    • Dallas–Fort Worth–Arlington, Texas metropolitan area
    • Houston–The Woodlands–Sugar Land, Texas metropolitan area
    • Washington-Arlington-Alexandria, District of Columbia–Virginia–Maryland–West Virginia metropolitan area
    • Miami–Fort Lauderdale–Pompano Beach, FL metropolitan area
    • Philadelphia-Camden-Wilmington, Pennsylvania–New Jersey–Delaware–Maryland metropolitan area
    • Atlanta–Sandy Springs–Alpharetta, Georgia metropolitan area
    • Phoenix-Mesa-Chandler, Arizona metropolitan area
    • Boston-Cambridge-Newton, Massachusetts–New Hampshire metropolitan area
    • San Francisco–Oakland–Berkeley, California metropolitan area
    • Detroit-Warren-Dearborn, Michigan metropolitan area
    • Seattle-Tacoma-Bellevue, Washington metropolitan area
    Table 2. Household Pulse Survey data collection schedule.
    Collection datesWeekResponse rate (%)
    January 6-January 18, 2021226.4
    January 20-February 1, 2021237.5
    February 3-February 15, 2021247.3
    February 17-March 1, 2021257.3
    March 3-March 15, 2021267.4
    March 17-March 29, 2021277.2
    April 14-April 26, 2021286.6
    April 28-May 10, 2021297.4
    May 12-May 24, 2021306.8
    Twitter Data

    To align with the HPS data collection period outlined in Table 2, the Twitter Streaming application programming interface, which provides access to a random sample of 1% of publicly available tweets, was used to collect tweets from the metropolitan areas represented in the HPS (Textbox 1) from January to May 2021. All tweets had place information (usually city and state). The place information found in tweets was used to determine the metropolitan area associated with each tweet. Next, to extract tweets related to COVID-19 vaccines, tweets were further filtered by matching variations of vaccine-related keywords, such as vaccine, pfizer, moderna, johnson & johnson, and dose. The tweet sample was further preprocessed to minimize noise resulting from tweets that matched our vaccine-related keywords but did not necessarily reflect the thoughts and opinions of individual Twitter users. For example, companies often promote job postings and advertisements on Twitter using targeted hashtags in hopes of reaching their target audience. To prevent these tweets from adding noise to the sample, tweets related to job postings and advertisements were removed by excluding tweets with hashtags and keywords such as #jobs, #hiring, and #ad.

    Sentiment and Emotion Analysis of Tweets

    To capture the attitudes found in COVID-19 vaccine–related tweets, a sentiment and emotion analysis of all tweets was conducted using the Natural Language Understanding Research Consortium (NRC) lexicon from the Syuzhet package in R (R Core Team) [21]. The NRC lexicon, developed by Saif Mohammad, contains a list of manually labeled English words and their associations with negative and positive sentiments and common human emotions, such as trust, fear, sadness, surprise, and disgust [22]. The Syuzhet package applies the NRC lexicon by independently evaluating and rating each word or expression within a tweet [23]. The get_nrc_sentiment function was applied to all tweets to calculate the valence of 8 different emotions (fear, joy, anticipation, anger, disgust, sadness, surprise, and trust), along with the overall positive and negative sentiments, toward the COVID-19 vaccine. To assess the accuracy of the sentiment classifier, a random sample of 1000 tweets was selected for manual classification as having a positive or negative sentiment. Among the 1000 tweets in the random sample, 734 (73.4%) were accurately classified by the automated sentiment classifier.

    The percentage of tweets expressing the 8 emotions, along with the percentage of tweets expressing a positive or negative sentiment, was calculated at the metropolitan level. For the purposes of this analysis, we used the proportion of tweets with a positive sentiment and positive emotions toward vaccines as a proxy to capture vaccine compliance among Twitter users, and the proportion of tweets with a negative sentiment and negative emotions toward vaccines was used as a proxy to capture vaccine hesitancy among Twitter users.

    Data Analysis

    Statistical analysis was conducted using the R software packages betareg and GLMMadaptive [24,25]. To determine whether COVID-19 vaccine attitudes on Twitter can predict the proportion of COVID-19 vaccine perceptions ultimately expressed in the HPS (unweighted), both generalized linear models (GLMs) and generalized linear mixed models (GLMMs) were constructed (Table 3). The models were developed using a total of 126 data points, including proportional vaccine compliance and hesitancy proxies from 14 metropolitan areas across 9 survey waves.

    Table 3. Regression models evaluating the relationship between Twitter sentiments and emotions and HPSa vaccine hesitancy and compliance.
    ModelFeaturesOutcome
    Model 1a (GLMb)
    • Percentage of positive sentiment
    • Percentage of joy
    • Percentage of surprise
    • Percentage of trust
    • Percentage of anticipation
    • Percentage of survey week (fixed effect)
    • Percentage of metropolitan area (fixed effect)
    		Percentage of vaccine-compliant HPS respondents
    Model 1b (GLMMc)
    • Percentage of positive sentiment
    • Percentage of joy
    • Percentage of surprise
    • Percentage of trust
    • Percentage of anticipation
    • Percentage of percentage of survey week (fixed effect)
    • Percentage of metropolitan area (random effect)
    		Percentage of vaccine-compliant HPS respondents
    Model 2a (GLM)
    • Percentage of negative sentiment
    • Percentage of anger
    • Percentage of disgust
    • Percentage of sadness
    • Percentage of fear
    • Percentage of anticipation
    • Survey week (fixed effect)
    • Metropolitan area (fixed effect)
    		Percentage of vaccine-hesitant HPS respondents
    Model 2b (GLMM)
    • Percentage of negative sentiment
    • Percentage of anger
    • Percentage of disgust
    • Percentage of sadness
    • Percentage of fear
    • Percentage of anticipation
    • Survey week (fixed effect)
    • Metropolitan area (random effect)
    		Percentage of vaccine-hesitant HPS respondents

    aHPS: Household Pulse Survey.

    bGLM: generalized linear model.

    cGLMM: generalized linear mixed model.

    GLMs were implemented with both time and geographic location as fixed effects, whereas the GLMMs were a multilevel approach with time as a fixed effect and metropolitan area as a random effect. The main reason behind this dual modeling choice is that random effects can capture the latent variation in the data that cannot be explained by fixed effects or the error term [26]. Random effects represent factors with multiple levels, such as geographic location, and possess distinct components that vary across these levels. Random effects prove especially valuable when dealing with hierarchical or nested data structures, where observations are not independent but grouped at a higher level, and enable us to account for similarities within these groups and prevent overfitting [25]. However, if there is limited variability across locations, including location as a random effect may lead to unstable estimates or unreliable inferences [27]. In such cases, it is often better to treat location as a fixed effect or aggregate the data at a higher level. Thus, we constructed both GLMs and GLMMs to assess the prediction power of using metropolitan areas as either fixed or random effects. We fit beta regression models with the logit link, which is the most appropriate for modeling proportional data [28]. In beta regression, the outcome variable is assumed to follow a beta distribution. Prior to evaluating the models, we conducted assumption checks and checked for multicollinearity and outliers to determine whether the necessary conditions were met. These model diagnostics are presented in Multimedia Appendix 1.

    As shown in Table 3, we constructed 2 models to predict vaccine compliance (models 1a, GLM, and 1b, GLMM) and 2 models to predict vaccine hesitancy (models 2a, GLM, and 2b, GLMM). In model 1a, we fit a GLM in which the predictor variables were each of the 5 positive Twitter-derived sentiment and emotion features and the outcome variable was the proportion of vaccine-compliant HPS respondents. This model controlled for survey week (time) and metropolitan area as fixed effects. In model 1b, to account for variations in time and location, we fit a GLMM with each of the 5 positive Twitter-derived sentiment and emotion features and survey week (time) as fixed effects and metropolitan area as a random effect.

    By contrast, in model 2a, we fit a GLM in which the predictor variables were each of the 6 negative Twitter-derived sentiment and emotion features and the outcome variable was the proportion of vaccine-hesitant HPS respondents. This model controlled for survey week (time) and metropolitan area as fixed effects. In model 2b, we fit a GLMM with each of the 6 negative Twitter-derived sentiment and emotion features and survey week (time) as fixed effects and metropolitan area as a random effect. As anticipation can be perceived as both positive and negative, this emotion was included as a feature in all models.

    Ethical Considerations

    This project does not meet the definition of human participant research under the purview of the University of Maryland Institutional Review Board according to federal regulations, section 45CFR46.102(e) [29].


    Descriptive Statistics

    There were a total of 92,453 tweets from 32,645 users across the 14 metropolitan areas in this study (Table 4). The Los Angeles–Long Beach–Anaheim metropolitan area had the largest representation of tweets (21,500/92,453, 23.26%), whereas the New York–Newark–New Jersey metropolitan area had the largest representation of users (18,400/32,645, 56.36%). The maximum number of tweets by a single individual was 274 (from a user in the New York–Newark–New Jersey metropolitan area). There were a total of 240,242 respondents to the HPS across the 14 metropolitan areas and 9 waves in this study, with the largest sample being the sample from the Washington-Arlington-Alexandria metropolitan area (Table 5).

    Table 4. Number of tweets (N=92,453) and users (N=32,645) by metropolitan area (January to May 2021).
    Metropolitan area, stateTweets, n (%)Users, n (%)Weekly number of tweets, mean (SD)Weekly number of users, mean (SD)
    Atlanta–Sandy Springs–Alpharetta, Georgia4234 (4.58)1542 (4.72)470 (186)254 (106)
    Boston-Cambridge-Newton, Massachusetts–New Hampshire3019 (3.27)1298 (3.98)335 (133)218 (83)
    Chicago-Naperville-Elgin, Illinois-Indiana-Wisconsin5821 (6.3)2561 (7.84)647 (252)426 (160)
    Dallas–Fort Worth–Arlington, Texas6203 (6.71)2299 (7.04)689 (265)371 (133)
    Detroit-Warren-Dearborn-Michigan1082 (1.17)518 (1.59)120 (56)84 (40)
    Houston–The Woodlands–Sugar Land, Texas5125 (5.54)2421 (7.42)569 (234)388 (145)
    Los Angeles–Long Beach–Anaheim, California21,500 (23.26)5429 (16.63)2389 (983)891 (344)
    Miami–Fort Lauderdale–Pompano Beach, Florida1954 (2.11)849 (2.6)217 (74)131 (40)
    New York–Newark–Jersey City, New York–New Jersey–Pennsylvania18,400 (19.9)7259 (22.24)2044 (683)1272 (400)
    Philadelphia-Camden-Wilmington, Pennsylvania–New Jersey–Delaware-Maryland3652 (3.95)1406 (4.31)406 (156)250 (88)
    Phoenix-Mesa-Chandler, Arizona4778 (5.17)1573 (4.82)531 (183)260 (81)
    San Francisco–Oakland–Berkeley, California6376 (6.9)2008 (6.15)708 (261)347 (116)
    Seattle-Tacoma-Bellevue, Washington3089 (3.34)1333 (4.08)343 (157)227 (103)
    Washington-Arlington-Alexandria, District of Columbia–Virginia–Maryland–West Virginia7220 (7.81)2419 (7.41)802 (313)436 (155)
    Table 5. Number of survey respondents (N=240,242) by city.
    Metropolitan area, stateRespondents, n (%)Weekly number of respondents, mean (SD)
    Atlanta–Sandy Springs–Alpharetta, Georgia12,611 (5.25)1261 (48)
    Boston-Cambridge-Newton, Massachusetts–New Hampshire20,078 (8.36)2008 (121)
    Chicago-Naperville-Elgin, Illinois-Indiana-Wisconsin16,044 (6.68)1604 (89)
    Dallas–Fort Worth–Arlington, Texas15,859 (6.6)1586 (88)
    Detroit-Warren-Dearborn-Michigan12,149 (5.06)1215 (88)
    Houston–The Woodlands–Sugar Land, Texas14,179 (5.9)1418 (125)
    Los Angeles–Long Beach–Anaheim, California17,006 (7.08)1701 (101)
    Miami–Fort Lauderdale–Pompano Beach, Florida11,641 (4.85)1164 (67)
    New York–Newark–Jersey City, New York–New Jersey–Pennsylvania19,730 (8.21)1973 (124)
    Philadelphia-Camden-Wilmington, Pennsylvania–New Jersey–Delaware–Maryland20,240 (8.42)2024 (162)
    Phoenix-Mesa-Chandler, Arizona14,027 (5.84)1403 (106)
    San Francisco–Oakland–Berkeley, California17,787 (7.4)1779 (78)
    Seattle-Tacoma-Bellevue, Washington18,615 (7.75)1862 (106)
    Washington-Arlington-Alexandria, District of Columbia–Virginia–Maryland–West Virginia30,276 (12.6)3028 (227)

    Attitudes Toward COVID-19 Vaccines in Twitter Data

    A sentiment analysis classified most tweets (50,415/92,453, 54.53% of tweets overall) across all metropolitan areas as having a positive sentiment (Table 6). The Washington-Arlington-Alexandria metropolitan area had the largest proportion of tweets with a positive sentiment (53,715/92,453, 58.1%), whereas the Miami–Fort Lauderdale–Pompano Beach metropolitan area had the lowest proportion of tweets with a positive sentiment (47,059/92,453, 50.9%). Tweets with a negative sentiment held the smallest proportions across all metropolitan areas (13,970/92,453, 15.11% of tweets overall). The Los Angeles–Long Beach–Anaheim metropolitan area had the largest proportion of tweets with a negative sentiment (15,162/92,453, 16.4%), whereas the Miami–Fort Lauderdale–Pompano Beach metropolitan area had the lowest proportion of tweets with a negative sentiment (11,926/92,453, 12.9%).

    Table 6. Distribution of sentiments and emotions found in COVID-19 vaccine tweets (N=92,453; January to May 2021).
    Sentiment or emotionTweets, n (%)
    Positive sentiment50,415 (54.53)
    Trust41,317 (44.69)
    Anticipation32,127 (34.75)
    Fear27,227 (29.45)
    Sadness24,935 (26.97)
    Joy24,241 (26.22)
    Anger21,671 (23.44)
    Surprise20,562 (22.24)
    Disgust14,746 (15.95)
    Negative13,970 (15.11)

    The emotion analysis revealed trust as the predominantly expressed emotion in COVID-19 vaccine tweets across all metropolitan areas (41,317/92,453, 44.69%). The most perceived negative emotion across all metropolitan areas was fear (27,227/92,453, 29.45%). The least perceived positive emotions were joy (24,241/92,453, 26.22%) and surprise (20,562/92,453, 22.24%), whereas the least perceived negative emotions were anger (21,671/92,453, 23.44%) and disgust (14,746/92,453, 15.95%). Examples of tweets expressing positive, neutral, and negative sentiments are presented in Textbox 2.

    Textbox 2. Examples of tweets expressing a positive or negative sentiment toward COVID-19 vaccines.

    Positive sentiments

    • “Feeling blessed to be healthy this birthday. My two biggest presents are coming in the next week: Inauguration and my second vaccine.”
    • “Hubby received his first vaccine does this morning-the sense of relief is for real, folks. #vaccinated”
    • “With my granddaughter Aurora, Andy, and Elliot. I can see them again and give them a hug now that I am fully Covid 19 vaccinated. I have had both shots plus over 2 weeks since shot two. Thank you President Biden.”
    • “My mom gets her second dose Sunday, big relief!”
    • “I love so much that I got vaccinated today.”
    • “Proud to work for you @bswhealth-my parents received their COVID vaccines this week at BUMC and said it was so quick and easy and the staff were so friendly! Thank you for taking care of them.”

    Negative sentiments

    • “This is from the Pfizer v-a-c-c-i-n-e. Please understand these shots cause harm. Injury is REAL & not rare. It’s a shame these poor people are being gaslighted, & media giants are censoring them.”
    • “They way my people been bugging me about this d*mn vaccine, I’m not getting that s*it.”
    • “No way!! No more lockdowns!! No vaccines!!! Oh and if your so concerned about the virus how about no illegals!!! Thank goodness for New Hampshire and Florida!! Go out.”
    • Clearly you are ignorant of the fact that they said even if you get the vaccine you still have to wear a mask, social distance & deal with all the same bull shit draconian orders. Even after blatant evidence you still want to get it. Heres 100% evidence of brain wash mind control.”
    • “I am 80. You can have my vaccine. I refuse to get one. I take 2 grams of vitamin C hourly. That makes me IMMUNE. Read: Linus Pauling. No mask. I am out every day working & walking in the park. Paul Kangas 4 Governor.”
    • “F the stupid vaccine.”

    Attitudes Toward COVID-19 Vaccines in the HPS Data

    Most survey respondents (127,833/240,242, 53.21%) across all metropolitan areas indicated that they received a COVID-19 vaccine, ranging from 50.2% (7041/14,027) of the survey respondents in the Phoenix-Mesa-Chandler metropolitan area to 56.4% (10,032/17,787) of the survey respondents in the San Francisco–Oakland–Berkeley metropolitan area (Table 7). Among the respondents who indicated that they received a COVID-19 vaccine, the majority (65,195/127,833, 51%) also indicated that they received or planned to receive all required doses. Among the respondents who indicated that they had not received a COVID-19 vaccine, the majority (89,759/112,409, 79.85% combined) indicated that they probably or definitely would get vaccinated), ranging from 48% (6733/14,027) of the survey respondents in the Phoenix-Mesa-Chandler metropolitan area to 75.2% (13,376/17,787) of the survey respondents in the San Francisco–Oakland–Berkeley metropolitan area. For the purposes of this analysis, individuals who answered they that would “definitely get a vaccine” or “probably get a vaccine” once available were considered vaccine compliant, and individuals who answered that they would “probably not get a vaccine” or “definitely not get a vaccine” once available were considered vaccine hesitant.

    Table 7. Distribution of survey responses (unweighted; January to May 2021).
    Question and responsesRespondents, n (%)
    Have you received a COVID-19 vaccine? (N=240,242)

    		Yes127,833 (53.21)

    		No or did not answer112,409 (46.79)
    Did you receive (or do you plan to receive) all required doses? (n=127,833)

    		Yes65,195 (51)

    		No or did not answer62,638 (49)
    Once a vaccine to prevent COVID-19 is available to you, would you... (n=112,409)

    		Definitely get a vaccine69,233 (61.59)

    		Probably get a vaccine20,526 (18.26)

    		Be unsure about getting a vaccine3114 (2.77)

    		Probably not get a vaccine10,836 (9.64)

    		Definitely not get a vaccine8700 (7.74)

    Predicting HPS Vaccine Attitudes Using Twitter-Based Attitudes

    We evaluated the performance of each GLM in terms of R-squared value and root mean square error (RMSE). Model 1a revealed significant associations (P<.001) between the percentage of vaccine-compliant HPS respondents and the percentage of tweets expressing a positive sentiment and trust (Table 8). The R-squared value for the vaccine-compliant GLM (model 1a) was 94.11%, and the RMSE was 0.053, which suggests that we can predict vaccine compliance in the HPS fairly well using positive sentiments and emotions found on Twitter. The GLM coefficients showed that an increase in the percentage of tweets expressing a positive sentiment (P<.001) was significantly associated with an increase in the percentage of vaccine-compliant HPS respondents. By contrast, an increase in the percentage of tweets expressing trust (P<.001) was significantly associated with a decrease in the percentage of vaccine-compliant HPS respondents.

    Table 8. Model results.
    Model and featuresβ coefficient (SE)P valueR-squared (%)Root mean square error
    Model 1a (GLMa)94.10.053

    		Percentage of positive sentiment5.007 (.865)<.001b


    		Percentage of joy.043 (1.482).98


    		Percentage of surprise1.084 (1.163).35


    		Percentage of trust−4.696 (.865)<.001b


    		Percentage of anticipation.930 (1.197).44

    Model 1b (GLMMc)82.50.062

    		Percentage of positive sentiment4.791 (.865)<.001b


    		Percentage of joy.271 (1.483).86


    		Percentage of surprise.942 (1.157).42


    		Percentage of trust−4.529 (.906)<.001b


    		Percentage of anticipation1.239 (1.107).26

    Model 2a (GLM)93.20.01

    		Percentage of negative sentiment−1.340 (.581).02b


    		Percentage of anger.382 (.556).49


    		Percentage of disgust−.356 (.612).56


    		Percentage of sadness−1.011 (.625).11


    		Percentage of fear.715 (.532).18


    		Percentage of anticipation−.382 (.369).30

    Model 2b (GLMM)9.40.032

    		Percentage of negative sentiment−1.334 (.617).03b


    		Percentage of anger.425 (.587).47


    		Percentage of disgust−.312 (.657).64


    		Percentage of sadness−1.015 (.662).13


    		Percentage of fear.643 (.567).26


    		Percentage of anticipation−.456 (.388).24

    aGLM: generalized linear model.

    bStatistically significant results (α=.05).

    cGLMM: generalized linear mixed model.

    When we compared the vaccine-compliant GLM (model 1a) with the vaccine-compliant GLMM with metropolitan area as a random effect (model 1b), we observed a lower R-squared value (82.5%) and higher RMSE (0.062). Model 1b results also showed that an increase in the percentage of tweets expressing a positive sentiment (P<.001) was significantly associated with an increase in the percentage of vaccine-compliant HPS respondents. By contrast, an increase in the percentage of tweets expressing trust (P<.001) was significantly associated with a decrease in the percentage of vaccine-compliant HPS respondents.

    Model 2a revealed significant associations (P<.05) between the percentage of vaccine-hesitant HPS respondents and the percentage of tweets expressing a negative sentiment. The R-squared value for the vaccine-hesitant GLM (model 2a) was similar to that of the vaccine-compliant GLM (93.17%). However, the vaccine-hesitant GLMM showed a much lower R-squared value (9.4%) and slightly higher RMSE (0.032). When compared with the vaccine-hesitant GLM (model 2a), whose only difference from the GLMM was the use of metropolitan area as a random effect, these results revealed that metropolitan area, as a fixed effect, and negative tweet sentiment (statistically significant in both model 2a and 2b) contributed to the majority of the variation in the percentage of vaccine-hesitant HPS respondents. When looking at the regressors, the results of both the vaccine-hesitant models, model 2a and 2b, showed that an increase in the percentage of tweets expressing a negative sentiment (P=.02 and P=.03, respectively) was associated with a decrease in the percentage of vaccine-hesitant HPS respondents.


    Principal Findings

    In this study, we sought to determine whether the sentiments and emotions found in COVID-19 vaccine tweets can predict the vaccine hesitancy and compliance expressed in the US Census Bureau’s HPS. Depending on the model, GLMs and GLMMs showed significant relationships between (1) the percentage of vaccine-compliant HPS respondents and percentage of tweets expressing a positive sentiment and trust and (2) the percentage of vaccine-hesitant HPS respondents and percentage of tweets expressing a negative sentiment. Positive perceptions expressed on Twitter performed well in predicting positive perceptions in the survey for both GLMs and GLMMs, whereas negative perceptions expressed on Twitter performed well in predicting negative perceptions in the survey only for the GLM.

    Study Findings in Context

    The main objective of this study was to examine whether aggregate attitudes extracted from social media can predict vaccine attitudes collected via surveys. Specifically, we hypothesized that there is a direct, positive relationship between (1) positive sentiments found in Twitter and the HPS survey and (2) negative sentiments found in Twitter and the HPS survey. We expected to see a positive relationship between positive sentiments and emotions on Twitter and vaccine compliance in the HPS, as suggested in a previous study that showed a positive relationship between positive sentiment scores in COVID-19 vaccine–related tweets and an increase in vaccination rates [30]. The results of both vaccine-compliant models revealed, as expected, significant positive relationships between the percentage of vaccine-compliant HPS respondents and percentage of tweets expressing a positive sentiment. However, in both vaccine-compliant models, the direction of one of the statistically significant relationships that were revealed was not what we expected. Both vaccine-compliant models revealed a significant inverse relationship between the vaccine-compliant measure in the HPS and percentage of tweets expressing trust.

    We also expected to see a positive relationship between negative sentiments and emotions on Twitter and vaccine hesitancy in the HPS. Although not significant, the percentages of tweets expressing anger and fear were, as expected, positively related to increases in vaccine hesitancy for both GLM and GLMM. However, the vaccine-hesitant GLM and GLMM both revealed a significant inverse relationship between the vaccine-hesitant measure in the HPS and percentage of tweets expressing a negative sentiment.

    The unexpected direction of the relationship between some vaccine perceptions in the survey and vaccine perceptions found on Twitter might be explained via ecological fallacy. An ecological fallacy occurs when an inference is made about the relationship between variables at an aggregate or a group level (eg, the level of a neighborhood, city, or country) based solely on the relationship between the same variables at an individual level [31]. This can be problematic because the relationship between variables at the group level may differ from the relationship between the same variables at the individual level because of factors such as aggregation bias and other contextual factors.

    Another possible explanation for the unexpected findings on some of the relationships between vaccine perceptions in the survey and vaccine perceptions found on Twitter could be the possibility that some individuals’ web-based personas may not match their reality. For example, an individual might be obligated to get a vaccine because of their job or an upcoming travel, making them vaccine compliant, but rant about it on the web. In our sample data, this type of person would be classified as provaccine instead of antivaccine in the HPS but would also contribute to the negative perceptions found on Twitter. These findings also align with prior research that suggested an individual’s web-based persona may differ from their offline identity [32-34]. This offline identity is often limited by physical, emotional, and financial circumstances that may be beyond an individual’s control [33,35-37]. However, individuals have complete control over the identity they choose to present on the web [32-34]. The inverse relationship between the vaccine-hesitant measure in the HPS and percentage of tweets expressing a negative sentiment may have also been due to the use of sarcasm in tweets, where the text itself contradicts what is actually meant by the user [38].

    The findings of this study contribute to the literature in 2 ways. First, although many studies have examined COVID-19 vaccine acceptance by extracting information from either surveys or social media, to our knowledge, no study has evaluated the relationship between these vastly different data sources. Unlike social media data collection, surveys come with postage, paper, printing, interviewer, and data entry costs, making them costly to administer [39]. Evaluating the relationship between the attitudes found in surveys and those found on social media allows researchers to determine whether social media data can be trusted to reveal the same information that can be extracted from traditional surveys or whether there is a risk of losing important information in exchange for cutting costs. In this study, we found that COVID-19 vaccine attitudes in the HPS, measured as vaccine compliance and hesitancy, can be predicted using social media attitudes toward vaccines, measured via sentiments and emotions toward vaccines. The results of this study support the efforts of researchers, who over the past few years have looked at social media as a data source, citing the availability of readily available data and no- or low-cost data collection efforts [40,41].

    This study makes further contributions by revealing the sentiments and emotions found in tweets across different metropolitan areas. This builds upon several other studies that leveraged natural language processing (NLP) methods, such as sentiment analysis, emotion analysis, and topic modeling, to examine vaccine-related perceptions [42-44]. In this study, we found that most tweets expressed a provaccine sentiment, across all metropolitan areas. However, many tweets also expressed negative feelings and anticipations. This supports previous work, where researchers found many discussions about vaccine hesitancy but ultimately found most tweets to have a positive sentiment [45]. This study also revealed trust as the dominant emotion found in tweets. This supports the results of a prior study that also found trust to be the dominant emotion expressed in tweets during an earlier period [46]. A comparison of these results shows that the vaccine conversation on Twitter remained relatively consistent over time.

    Comparing COVID-19 vaccine perceptions on Twitter with attitudes in traditional public health surveys offers several benefits. Twitter serves as a platform for immediate and widespread dissemination of information. Analyzing vaccine perceptions on Twitter can help identify emerging issues or concerns related to COVID-19 vaccines at an early stage. This early detection allows public health authorities to address misconceptions, respond to emerging challenges, and promptly adapt their communication strategies. For example, the study results suggest that both models may be beneficial when deciding which cities to implement vaccine campaigns in, and the vaccine-compliant model can be used to better understand the role sentiments play in vaccination behaviors. This type of model can be used to craft effective social media messages related to COVID-19 vaccination.

    Twitter provides a platform for a wide range of voices and opinions, including those of individuals with varying backgrounds, beliefs, and experiences. Comparing Twitter data with survey data allows for the exploration of diverse perspectives and can uncover viewpoints that may not be captured through traditional surveys alone. This broader range of perspectives enhances the understanding of the complexities surrounding vaccine perceptions. Twitter data also allow for the real-time monitoring of public sentiment and reactions toward COVID-19 vaccines. This timely information can provide valuable insights into evolving trends, emerging concerns, and the impact of specific events or interventions. By comparing Twitter discussions with survey responses, researchers can identify shifts in public opinion and monitor the effectiveness of public health communication strategies in real time.

    Comparison with Twitter data can complement the findings of traditional surveys, providing a more comprehensive understanding of vaccine perceptions. Twitter data can provide contextual information, qualitative insights, and real-world examples that enrich the analysis of survey responses. The combination of both sources offers a more nuanced and holistic understanding of public attitudes toward COVID-19 vaccines.

    This study provides further evidence for the benefits of using social media data for public health research. The overarching contribution of this work suggests the adoption of alternative data sources and NLP techniques to assist in public health decision-making.

    Limitations and Future Work

    Considering the limitations of this study may lead to future, related work. This study emphasizes the use of Twitter as a data source, but the lack of representation among Twitter users leads to bias in the sample and contributes to sampling errors. For example, Twitter users tend to be younger, be more educated, have higher incomes, and be more liberal [47]. The lack of representation among Twitter users suggests the limited generalizability of the results to the larger population. Adding to this lack of representation is the limited sample of tweets available to the public via the Twitter Streaming application programming interface, which makes available a random sample of 1% of all tweets made by Twitter users at any given time [48]. In addition, in studies assessing COVID-19 vaccine perceptions using social media data, individuals who do not have access to social media are systematically excluded from the analysis sample.

    The lack of demographic information on Twitter users is also a limitation to using Twitter as a data source. The absence of demographic information, such as age, gender, income, and education makes it challenging to understand the characteristics of the Twitter users who generate the data. This lack of information may lead to biased or incomplete analyses and limit the generalizability of the findings. In addition, the absence of demographic data makes it difficult to compare Twitter data with data from other sources that do contain demographic information, such as survey data. Despite these limitations, Twitter data can still be useful in certain contexts.

    It should also be acknowledged that the HPS data are also subject to sampling errors due to sample design, nonresponse, weighting adjustments, and measurement errors [49]. As a result, the true relationship between aggregate attitudes extracted from social media and vaccine attitudes collected via surveys may be different from what was revealed in this study.

    Future studies should endeavor to use other NLP approaches, such as topic modeling, to compare public perceptions of the COVID-19 vaccine on social media with those found in surveys. The survey used in this study, the HPS, presented respondents with in-depth questions related to why they were vaccine hesitant, so applying topic models to tweets may reveal some of the same attitudes and themes as those expressed in the survey. Future studies may also involve pulling data from other social media platforms, such as Facebook, and comparing the overall perceptions reflected across all media.

    Conclusions

    The ongoing COVID-19 pandemic requires consistent monitoring and data-driven public health policies. To slow the spread of the virus, public health officials have stressed that vaccines are essential in the worldwide battle against COVID-19. However, vaccine hesitancy continues to be a barrier to effective and consistent vaccine rollout programs. Prior efforts have used surveys to gauge attitudes toward the COVID-19 vaccine, but this study suggests that these public perceptions may also be extracted from a readily available, low-cost data source, social media. In this study, we validated social media as a data source by evaluating the relationship between the attitudes expressed among Twitter users and attitudes expressed among respondents to the HPS as well as the ability of attitudes expressed among Twitter users to predict vaccine compliance and hesitancy among the HPS respondents. Leveraging Twitter data alongside traditional surveys can provide a more comprehensive and nuanced understanding of COVID-19 vaccine perceptions, facilitating evidence-based decision-making and tailored public health strategies.

    Conflicts of Interest

    None declared.

    Multimedia Appendix 1

    Model diagnostics.

    DOC File , 3561 KB
    1. Household pulse survey COVID-19 vaccination tracker. United States Census Bureau. Dec 22, 2021. URL: https:/​/www.​census.gov/​library/​visualizations/​interactive/​household-pulse-survey-covid-19-vaccination-tracker.​html [accessed 2022-01-13]
    2. Garett R, Young SD. Online misinformation and vaccine hesitancy. Transl Behav Med. Dec 14, 2021;11(12):2194-2199. [FREE Full text] [CrossRef] [Medline]
    3. Pershad Y, Hangge PT, Albadawi H, Oklu R. Social medicine: Twitter in healthcare. J Clin Med. May 28, 2018;7(6):121. [FREE Full text] [CrossRef] [Medline]
    4. Chunara R, Andrews JR, Brownstein JS. Social and news media enable estimation of epidemiological patterns early in the 2010 Haitian cholera outbreak. Am J Trop Med Hyg. Jan 2012;86(1):39-45. [FREE Full text] [CrossRef] [Medline]
    5. Hong L, Fu C, Torrens P, Frias-Martinez V. Understanding citizens' and local governments' digital communications during natural disasters: the case of snowstorms. In: Proceedings of the 2017 ACM on Web Science Conference. Presented at: WebSci '17: ACM Web Science Conference; June 25-28, 2017, 2017; Troy, NY. [CrossRef]
    6. Ball-Rokeach SJ, DeFleur ML. A dependency model of mass-media effects. Commun Res. Jan 1976;3(1):3-21. [CrossRef]
    7. Marston HR, Ko PC, Girishan Prabhu V, Freeman S, Ross C, Sharaievska I, et al. Digital practices by citizens during the COVID-19 pandemic: findings from an international multisite study. JMIR Ment Health. Mar 06, 2023;10:e41304. [FREE Full text] [CrossRef] [Medline]
    8. Charquero-Ballester M, Walter JG, Nissen IA, Bechmann A. Different types of COVID-19 misinformation have different emotional valence on Twitter. Big Data Soc. Sep 22, 2021;8(2):205395172110412. [FREE Full text] [CrossRef]
    9. Bonnevie E, Gallegos-Jeffrey A, Goldbarg J, Byrd B, Smyser J. Quantifying the rise of vaccine opposition on Twitter during the COVID-19 pandemic. J Commun Healthc. Dec 15, 2020;14(1):12-19. [CrossRef]
    10. Deiner MS, Fathy C, Kim J, Niemeyer K, Ramirez D, Ackley SF, et al. Facebook and Twitter vaccine sentiment in response to measles outbreaks. Health Informatics J. Sep 2019;25(3):1116-1132. [FREE Full text] [CrossRef] [Medline]
    11. Ward JK, Alleaume C, Peretti-Watel P, COCONEL Group. The French public's attitudes to a future COVID-19 vaccine: the politicization of a public health issue. Soc Sci Med. Nov 2020;265:113414. [FREE Full text] [CrossRef] [Medline]
    12. Wang K, Wong EL, Ho KF, Cheung AW, Chan EY, Yeoh EK, et al. Intention of nurses to accept coronavirus disease 2019 vaccination and change of intention to accept seasonal influenza vaccination during the coronavirus disease 2019 pandemic: a cross-sectional survey. Vaccine. Oct 21, 2020;38(45):7049-7056. [FREE Full text] [CrossRef] [Medline]
    13. Ruiz JB, Bell RA. Predictors of intention to vaccinate against COVID-19: results of a nationwide survey. Vaccine. Mar 12, 2021;39(7):1080-1086. [FREE Full text] [CrossRef] [Medline]
    14. Arigo D, Pagoto S, Carter-Harris L, Lillie SE, Nebeker C. Using social media for health research: methodological and ethical considerations for recruitment and intervention delivery. Digit Health. May 07, 2018;4:2055207618771757. [FREE Full text] [CrossRef] [Medline]
    15. Sigalo N, Awasthi N, Abrar SM, Frias-Martinez V. Using COVID-19 vaccine attitudes on Twitter to improve vaccine uptake forecast models in the United States: infodemiology study of Tweets. JMIR Infodemiology. Aug 21, 2023;3:e43703. [FREE Full text] [CrossRef] [Medline]
    16. Wright KB. Researching internet-based populations: advantages and disadvantages of online survey research, online questionnaire authoring software packages, and web survey services. J Comput Mediat Commun. Apr 1, 2005;10(3):JCMC1034. [FREE Full text] [CrossRef]
    17. Enriquez D, Goldstein A. COVID-19’s socioeconomic impact on low-income benefit recipients: early evidence from tracking surveys. Socius. Nov 25, 2020;6:237802312097079. [FREE Full text] [CrossRef]
    18. Morales DX, Morales SA, Beltran TF. Food insecurity in households with children amid the COVID-19 pandemic: evidence from the household pulse survey. Soc Curr. Apr 23, 2021;8(4):314-325. [FREE Full text] [CrossRef]
    19. Household pulse survey technical documentation. United States Census Bureau. URL: https://www.census.gov/programs-surveys/household-pulse-survey/technical-documentation.html [accessed 2023-05-09]
    20. Daly M, Jones A, Robinson E. Public trust and willingness to vaccinate against COVID-19 in the US from October 14, 2020, to March 29, 2021. JAMA. Jun 15, 2021;325(23):2397-2399. [FREE Full text] [CrossRef] [Medline]
    21. Silge J, Robinson D. tidytext: text mining and analysis using tidy data principles in R. J Open Source Softw. 2016;1(3):37. [FREE Full text] [CrossRef]
    22. NRC word-emotion association lexicon. NRC Emotion Lexicon. URL: http://saifmohammad.com/WebPages/NRC-Emotion-Lexicon.htm [accessed 2022-01-13]
    23. Kausar MA, Soosaimanickam A, Nasar M. Public sentiment analysis on Twitter data during COVID-19 outbreak. Int J Adv Comput Sci Appl. 2021;12(2) [FREE Full text] [CrossRef]
    24. Zeileis A, Cribari-Neto F, Gruen B, Kosmidis I, Simas AB, Rocha AV. Package ‘betareg’. The Comprehensive R Archive Network. Oct 12, 2022. URL: https://cran.r-project.org/web/packages/betareg/betareg.pdf [accessed 2023-11-10]
    25. Rizopoulos D. GLMMadaptive: generalized linear mixed models using adaptive gaussian quadrature. The Comprehensive R Archive Network. 2023. URL: https://cloud.r-project.org/web/packages/GLMMadaptive/index.html [accessed 2023-05-09]
    26. Li B, Lingsma HF, Steyerberg EW, Lesaffre E. Logistic random effects regression models: a comparison of statistical packages for binary and ordinal outcomes. BMC Med Res Methodol. May 23, 2011;11(1):77. [FREE Full text] [CrossRef] [Medline]
    27. Harrison XA, Donaldson L, Correa-Cano ME, Evans J, Fisher DN, Goodwin CE, et al. A brief introduction to mixed effects modelling and multi-model inference in ecology. PeerJ. May 23, 2018;6:e4794. [FREE Full text] [CrossRef] [Medline]
    28. Chen K, Cheng Y, Berkout O, Lindhiem O. Analyzing proportion scores as outcomes for prevention trials: a statistical primer. Prev Sci. Apr 2017;18(3):312-321. [FREE Full text] [CrossRef] [Medline]
    29. IRB Definitions. Division of Research - University of Maryland URL: https:/​/research.​umd.edu/​resources/​research-compliance/​institutional-review-board-irb/​irb-definitions [accessed 2023-11-27]
    30. Bari A, Heymann M, Cohen RJ, Zhao R, Szabo L, Apas Vasandani S, et al. Exploring coronavirus disease 2019 vaccine hesitancy on Twitter using sentiment analysis and natural language processing algorithms. Clin Infect Dis. May 15, 2022;74(Suppl_3):e4-e9. [CrossRef] [Medline]
    31. Reichert M, Brüßler S, Reinhard I, Braun U, Giurgiu M, Hoell A, et al. The association of stress and physical activity: mind the ecological fallacy. Ger J Exerc Sport Res. May 06, 2022;52(2):282-289. [FREE Full text] [CrossRef]
    32. Huang J, Kumar S, Hu C. A literature review of online identity reconstruction. Front Psychol. Aug 23, 2021;12:696552. [FREE Full text] [CrossRef] [Medline]
    33. Kim HW, Zheng JR, Gupta S. Examining knowledge contribution from the perspective of an online identity in blogging communities. Comput Hum Behav. Sep 2011;27(5):1760-1770. [FREE Full text] [CrossRef]
    34. Hu C, Zhao L, Huang J. Achieving self-congruency? Examining why individuals reconstruct their virtual identity in communities of interest established within social network platforms. Comput Hum Behav. Sep 2015;50:465-475. [FREE Full text] [CrossRef]
    35. Bargh JA, McKenna KY, Fitzsimons GM. Can you see the real me? Activation and expression of the “true self” on the internet. J Soc Issues. Dec 17, 2002;58(1):33-48. [CrossRef]
    36. Donath JS. Identity and deception in the virtual community. In: Kollock P, Smith M, editors. Communities in Cyberspace. London, UK. Routledge; Dec 17, 1998.
    37. Schau HJ, Gilly MC. We are what we post? Self-presentation in personal web space. J Consum Res. Dec 2003;30(3):385-404. [CrossRef]
    38. Govindan V, Balakrishnan V. A machine learning approach in analysing the effect of hyperboles using negative sentiment tweets for sarcasm detection. J King Saud Univ Comput Inf Sci. Sep 2022;34(8):5110-5120. [FREE Full text] [CrossRef]
    39. Morland K, Filomena S. Disparities in the availability of fruits and vegetables between racially segregated urban neighbourhoods. Public Health Nutr. Dec 2007;10(12):1481-1489. [CrossRef] [Medline]
    40. Lewis LB, Sloane DC, Nascimento LM, Diamant AL, Guinyard JJ, Yancey AK, et al. African Americans' access to healthy food options in South Los Angeles restaurants. Am J Public Health. Apr 2005;95(4):668-673. [CrossRef] [Medline]
    41. Sigalo N, St Jean B, Frias-Martinez V. Using social media to predict food deserts in the United States: infodemiology study of tweets. JMIR Public Health Surveill. Jul 05, 2022;8(7):e34285. [FREE Full text] [CrossRef] [Medline]
    42. Sattar NS, Arifuzzaman S. COVID-19 vaccination awareness and aftermath: public sentiment analysis on Twitter data and vaccinated population prediction in the USA. Appl Sci. Jun 30, 2021;11(13):6128. [FREE Full text] [CrossRef]
    43. Liu S, Liu J. Public attitudes toward COVID-19 vaccines on English-language Twitter: a sentiment analysis. Vaccine. Sep 15, 2021;39(39):5499-5505. [FREE Full text] [CrossRef] [Medline]
    44. Hussain A, Tahir A, Hussain Z, Sheikh Z, Gogate M, Dashtipour K, et al. Artificial intelligence-enabled analysis of public attitudes on Facebook and Twitter toward COVID-19 vaccines in the United Kingdom and the United States: observational study. J Med Internet Res. Apr 05, 2021;23(4):e26627. [FREE Full text] [CrossRef] [Medline]
    45. Yousefinaghani S, Dara R, Mubareka S, Papadopoulos A, Sharif S. An analysis of COVID-19 vaccine sentiments and opinions on Twitter. Int J Infect Dis. Jul 2021;108:256-262. [FREE Full text] [CrossRef] [Medline]
    46. Lyu JC, Han EL, Luli GK. COVID-19 vaccine-related discussion on Twitter: topic modeling and sentiment analysis. J Med Internet Res. Jun 29, 2021;23(6):e24435. [FREE Full text] [CrossRef] [Medline]
    47. Freedman DA. Local food environments: they're all stocked differently. Am J Community Psychol. Dec 2009;44(3-4):382-393. [CrossRef] [Medline]
    48. Nguyen QC, Li D, Meng HW, Kath S, Nsoesie E, Li F, et al. Building a national neighborhood dataset from geotagged Twitter data for indicators of happiness, diet, and physical activity. JMIR Public Health Surveill. Oct 17, 2016;2(2):e158. [FREE Full text] [CrossRef] [Medline]
    49. Bradley VC, Kuriwaki S, Isakov M, Sejdinovic D, Meng XL, Flaxman S. Unrepresentative big surveys significantly overestimated US vaccine uptake. Nature. Dec 2021;600(7890):695-700. [FREE Full text] [CrossRef] [Medline]

    ‎
    GLM: generalized linear model
    GLMM: generalized linear mixed model
    HPS: Household Pulse Survey
    NLP: natural language processing
    NRC: Natural Language Understanding Research Consortium
    RMSE: root mean square error

    Edited by R Cuomo; submitted 20.10.22; peer-reviewed by F Kreute, D Steel, Y Mejova, H Mondal; comments to author 16.02.23; revised version received 18.05.23; accepted 14.10.23; published 30.11.23.

    Copyright

    ©Nekabari Sigalo, Vanessa Frias-Martinez. Originally published in JMIR Infodemiology (https://infodemiology.jmir.org), 30.11.2023.

    This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Infodemiology, is properly cited. The complete bibliographic information, a link to the original publication on https://infodemiology.jmir.org/, as well as this copyright and license information must be included.