Dark Mode
DMO Twitter Engagement Dataset
Social Media and Posts
Tags and Keywords
Trusted By
"No reviews yet"
Free
About
This dataset is designed for studying the social media content strategy of Destination Marketing Organisations (DMOs). It aims to understand how linguistic features in DMOs' social media content evolve across different phases: pre-COVID, during lockdown, and post-lockdown, and their subsequent impact on user engagement on social media [1]. The data focuses on key variables such as the confidence and positive engagement expressed in tweets, the amount of cognitive content, media type, and tweet metrics like likes and retweets [1]. The timelines for these phases are determined by the Government of India's lockdown declarations [1].
Columns
The dataset includes 41 columns capturing various aspects of tweets from DMOs. Here are some of the key columns:
- Sno: A serial number for each record.
- State: The Indian state associated with the DMO, with Rajasthan and Odisha being frequently observed [2].
- conversation_id: A unique identifier for each tweet conversation [2].
- retweet_count: The number of times a tweet has been retweeted, ranging from 0 to 3519, with an average of 18.9 [3].
- reply_count: The number of replies to a tweet, varying from 0 to 4580, with an average of 3.6 [4].
- like_count: The number of likes a tweet received, ranging from 0 to 38,244, averaging 140 [4].
- quote_count: The number of times a tweet was quoted, typically low, with a mean of 1.47 [5].
- Buzz: A measure of tweet engagement, ranging up to 48,900, with an average of 203 [5, 6].
- id: Another unique identifier for each record [6].
- Date1: The date when the tweet was posted, with 1043 unique dates recorded [6].
- OpnHours: Indicates if the tweet was posted during Working Hours (57%) or Non-working Hours (43%) [6, 7].
- DateDay: The specific day of the week the tweet was posted, with Friday and Thursday being common [7].
- Day: Categorises the day as weekday (74%) or weekend (26%) [7].
- Time: Categorises the time of posting as Non-business hours (57%) or Business hours (43%) [7].
- 3-Phase: Classifies the tweet period into Post Lockdown (40%), Pre-Covid (32%), and an 'Other' category which likely includes the lockdown phase [7, 8].
- 4-phase: Provides a similar classification of tweet periods as 3-Phase [8].
- Followers: The number of followers the DMO account had, ranging from 5,164 to 1.97 million, with an average of 508,000 [8, 9].
- Status text: The actual text content of the tweet, with 23,001 unique texts [9].
- Vividness: The type of media used in the tweet, primarily photo (79%) or video (15%) [9, 10].
- ContentType: Categorises content as Information (71%) or Interaction (29%) [10].
- WC: Word Count of the tweet, averaging 38.7 words per tweet [10, 11].
- Clout: A linguistic feature, averaging 63.9 [11, 12].
- Cognition: Reflects the amount of cognitive content in the tweet, with a mean of 4.71 [12, 13].
- Affect: A linguistic feature indicating emotional tone, with an average of 7.37 [13, 14].
- emotion: Overall emotional score, averaging 4.43 [14, 15].
- emo_pos: Represents positive emotion, with a mean of 1.09 [15, 16].
- emo_neg: Represents negative emotion, with a mean of 3.26 [16].
- we: A linguistic feature, averaging 0.63 [16, 17].
- tentat: A linguistic feature related to tentativeness, averaging 0.65 [17, 18].
- Drives: A linguistic feature, averaging 3.19 [18, 19].
- i: A linguistic feature, with a low mean of 0.05 [19].
- they: A linguistic feature, averaging 0.19 [20, 21].
- insight: A linguistic feature related to insight, averaging 1.43 [21, 22].
- cause: A linguistic feature, averaging 0.48 [22].
- discrep: A linguistic feature related to discrepancy, averaging 0.52 [23].
- certitude: A linguistic feature related to certitude, averaging 0.22 [23, 24].
- Positive: A sentiment score, ranging from 1 to 5, with a mean of 1.27 [24].
- Negative: A sentiment score, ranging from -5 to -1, with a mean of -1.06 [25].
- Total_Sentiment: The combined sentiment score, ranging from 0 to 6, averaging 0.33 [25].
Distribution
The dataset comprises 21,677 tweets posted between 25th March 2019 and 31st January 2022 [1]. It is provided as a CSV file named "Data LIWC 01 02 23.csv" with a size of 11.6 MB [26]. The dataset contains 41 columns and all records are valid, with no missing or mismatched values reported [2-25, 27].
Usage
This dataset is ideal for:
- Analysing social media content strategies of Destination Marketing Organisations [1].
- Studying the evolution of linguistic features in online communication during different crisis phases (pre-COVID, lockdown, post-lockdown) [1].
- Researching the impact of content strategy elements (e.g., confidence, positive engagement, cognitive content, media type) on user engagement (likes, retweets) [1].
- Understanding patterns of DMO social media activity based on factors like time of day, day of week, and content type [6, 7, 10].
- Developing models to predict social media user engagement based on tweet characteristics [1].
Coverage
The dataset covers tweets from 23 Destination Marketing Organisations (DMOs) located in India, with specific timelines chosen based on the Government of India's lockdown declarations [1, 2]. The time range spans from 25th March 2019 to 31st January 2022 [1]. Geographical representation includes states like Rajasthan and Odisha, along with 21 other states grouped as "Other" [2].
License
CC0: Public Domain
Who Can Use It
- Marketing Researchers: To study digital marketing strategies, content marketing, and consumer engagement in the tourism sector.
- Linguists and Computational Linguists: To analyse changes in language use in public communication during crisis periods.
- Data Scientists and Analysts: For sentiment analysis, time-series analysis of social media data, and predictive modelling of engagement metrics.
- Tourism Boards and DMOs: To gain insights into effective social media communication during and after major disruptions.
- Academics and Students: For academic research, dissertations, and projects related to social media, tourism, and crisis communication.
Dataset Name Suggestions
- DMO Twitter Engagement Dataset
- COVID-19 DMO Social Media Study
- India Tourism Twitter Data
- Linguistic Features of DMO Tweets
- Destination Marketing Social Media Analytics
Attributes
Original Data Source: DMO Twitter Engagement Dataset
QUALITY