Table 2 Tasks respondents currently do, or would like to do on a social media analysis platform (n = 173)
From: Informing social media analysis for public health: a cross-sectional survey of professionals
Tasks N (%) | Currently do | Want to, not possible with current tools | Want to, don’t know how | Don't want to |
|---|---|---|---|---|
Develop own boolean search strings | 68 (39.5) | 36 (20.9) | 53 (30.8) | 15 (8.7) |
Analyze narratives based on a taxonomy | 76 (43.9) | 39 (22.5) | 50 (28.9) | 8 (4.6) |
Use filters for gender or user-type | 69 (40.1) | 41 (23.8) | 44 (25.6) | 18 (10.5) |
Use filters for sentiment or post intent (ie questions / complaints) | 78 (46.4) | 33 (19.6) | 44 (26.2) | 13 (7.7) |
Filter data to a specific country | 116 (65.5) | 22 (12.4) | 34 (19.2) | 5 (2.8) |
Filter data to a specific language | 104 (59.1) | 27 15.3) | 34 (19.3) | 11 (6.3) |
Filter data for specific platform | 103(59.9) | 24 (14) | 34 (19.8) | 11 6.4) |
Group posts to themes | 86 (50) | 38 (22.1) | 39 (22.7) | 9 (5.2) |
Export data | 105 (61.4) | 25 (14.6) | 32 (18.7) | 9 (5.3) |
Annotate narrative themes and content for easier thematic analysis | 68 (41.5) | 39 (23.8) | 49 (29.9) | 8 (4.9) |
Compare and integrate with other data sources | 70 (41.5) | 41 (24.3) | 48 (28.4) | 10 (5.9) |
Examine narrative themes and labels by changes in velocity as well as volume | 52 (31.7) | 38 (23.2) | 59 (36) | 15 (9.1) |
Compare data over time | 120 (68.2) | 26 (14.8) | 27 (15.3) | 3 (1.7) |
Identify common misinformation narratives | 115 9 (67.3) | 24 (14) | 28 (16.4) | 4 (2.3) |
Do social network analysis | 97 (56.1) | 35 (20.2) | 38 (22) | 3 (1.7) |
Trace a narrative through social network model and time | 55 (33.1) | 45 (27.1) | 55 (33.1) | 11 (6.6) |
Annotate a narrative shift or mutation over time and social connections | 49 (29.2) | 38 (22.6) | 63 (37.5) | 18 (10.7) |
Identify bots and bot-generated content | 31 (18.9) | 40 (24.4) | 79 (48.2) | 14 (8.5) |
Import list of trusted users or sources for automated annotation/whitelisting | 32 (20) | 41 (25.6) | 69 (43.1) | 18 (11.3) |
Import list of dubious users or pages for flagging in the analysis as untrustworthy | 33 (20.4) | 42 (25.9) | 68 (42) | 19 (11.7) |
Perform analysis of links to internet web sites or other platforms to detect cross platform switching of content and users | 35 (21.6) | 48 (29.6) | 64 (39.5) | 15 (9.3) |
Automatic scoring of credibility/authority of links in the social media content | 27 (16.6) | 41 (25.2) | 81 (49.7) | 14 (8.6) |
Identification, filtering and analysis of duplicated content from humans and bots | 27 (16.6) | 43 (25.7) | 76 (45.5) | 21 (12.6) |
Ability to filter between boosted and unboosted/organic content | 38 (22.8) | 45 (26.9) | 65 (38.9) | 19 (11.4) |
Compare text based to image or video-based content | 54 (31.8) | 41 (24.1) | 57 (33.5) | 18 (10.6) |
Compare data between countries or communities | 77 (44.8) | 37 (21.5) | 43 (25) | 15 (8.7) |