MétaCan
Menu
Back to cohort
Record W2508320532 · doi:10.2196/resprot.4542

Twitter Strategies for Web-Based Surveying: Descriptive Analysis From the International Concussion Study

2016· article· en· W2508320532 on OpenAlexvenueno aff
Sharief Hendricks, Peter Düking, Stephen D. Mellalieu

Bibliographic record

VenueJMIR Research Protocols · 2016
Typearticle
Languageen
FieldSocial Sciences
TopicSocial Media in Health Education
Canadian institutionsnot available
FundersUniversity of Cape TownNational Research Foundation
KeywordsSnowball samplingSocial mediaPublicityPsychologyDescriptive statisticsSocial network analysisSocial network (sociolinguistics)Applied psychologyMedical educationComputer scienceAdvertisingWorld Wide WebInternet privacyMedicineBusinessMarketing

Abstract

fetched live from OpenAlex

BACKGROUND: Social media provides researchers with an efficient means to reach and engage with a large and diverse audience. Twitter allows for the virtual social interaction among a network of users that enables researchers to recruit and administer surveys using snowball sampling. Although using Twitter to administer surveys for research is not new, strategies to improve response rates are yet to be reported. OBJECTIVE: To compare the potential and actual reach of 2 Twitter accounts that administered a Web-based concussion survey to rugby players and trainers using 2 distinct Twitter-targeting strategies. Furthermore, the study sought to determine the likelihood of receiving a retweet based on the time of the day and day of the week of posting. METHODS: A survey based on previous concussion research was exported to a Web-based survey website Survey Monkey. The survey comprised 2 questionnaires, one for players, and one for those involved in the game (eg, coaches and athletic trainers). The Web-based survey was administered using 2 existing Twitter accounts, with each account executing a distinct targeting strategy. A list of potential Twitter accounts to target was drawn up, together with a list of predesigned tweets. The list of accounts to target was divided into 'High-Profile' and 'Low-Profile', based on each accounts' position to attract publicity with a high social interaction potential. The potential reach (number of followers of the targeted account), and actual reach (number of retweets received by each post) between the 2 strategies were compared. The number of retweets received by each account was further analyzed to understand when the most likely time of day, and day of the week, a retweet would be received. RESULTS: The number of retweets received by a Twitter account decreased by 72% when using the 'high-profile strategy' compared with the 'low-profile strategy' (incidence rate ratio (IRR); 0.28, 95% confidence interval (CI) 0.21-0.37, P<.001). When taking into account strategy and day of the week, the IRR for the number of retweets received during the hours of 12 AM to 5:59 AM (IRR 2.98, 95% CI 1.88-4.71, P>.001) and 6 PM to 11:59 PM (IRR 1.48, 95% CI 1.05-2.09, P>.05) were significantly increased relative to 6 AM to 11:59 AM. However, posting tweets during the hours of 12 PM to 5:59 PM, decreased the IRR for retweets by 40% (IRR 0.60, 95% CI 0.46-0.79, P<.001) compared with 6 AM to 11:59 AM. Posting on a Monday (IRR 3.57, 95% CI 2.50-5.09, P<.001) or Wednesday (IRR 1.50, 95% CI 1.11-1.11, P<.01) significantly increased the IRR compared with posting on a Thursday. CONCLUSIONS: Surveys are a useful tool to measure the knowledge, attitudes, and behaviors of a given population. Strategies to improve Twitter engagement include targeting low-profile accounts, posting tweets in the morning (12 AM-11:59 AM) or late evenings (6 PM-11:59 PM), and posting on Mondays and Wednesdays.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.010
metaresearch head score (Gemma)0.039
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesMetaresearch
Consensus categoriesnone
DomainCandidate signal: Methods · Consensus signal: none
Study designCandidate signal: Observational · Consensus signal: Observational
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.990
Threshold uncertainty score0.053

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0100.039
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.001
Bibliometrics0.0040.007
Science and technology studies0.0010.001
Scholarly communication0.0020.002
Open science0.0010.002
Research integrity0.0010.001
Insufficient payload (model declined to judge)0.0020.001

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.640
GPT teacher head0.636
Teacher spread0.005 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

Study designObservational
DomainMethods
GenreEmpirical

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations14
Published2016
Admission routes1
Has abstractyes

Explore more

Same venueJMIR Research ProtocolsSame topicSocial Media in Health EducationFrench-language works237,207