Skip to main content

Table 2 Data analysis features of included studies (n=36)

From: What methods are used to examine representation of mental ill-health on social media? A systematic review

Methodological Feature

N (%)

Study citation/s

Manual Analysis

  

 Deductive content analysis

9 (25%)

[19, 24, 29, 39, 46, 47, 49, 50, 54]

 Deductive/inductive content analysis

6 (16.7%)

[31, 35, 40, 43, 51, 55]

 Inductive content analysis

8 (22.2%)

[26,27,28, 32, 33, 36, 56, 57]

 Inductive thematic analysis

1 (2.8%)

[52]

 Deductive content analysis & inductive thematic analysis

1 (2.8%)

[34]

Computer Assisted Analysis

  

 Manual content analysis and machine learning

5 (13.9%)

[20, 25, 30, 41, 53]

 Linguistic analysis

3 (8.3%)

[44, 45, 48]

 Sentiment analysis and topic modelling

2 (5.6%)

[37, 38]

 N-gram language modelling

1 (2.8%)

[42]

Unit of Analysis

  

 Text

28 (77.8%)

[19, 20, 24,25,26,27,28,29,30,31,32,33,34,35, 37,38,39,40,41,42,43,44,45,46,47,48, 53, 57]

 Image

3 (8.3%)

[33, 53, 56]

 Caption

2 (5.6%)

[53, 56]

 Comments

8 (22.2%)

[35, 36, 47, 52,53,54,55,56, 57]

 Video

4 (11.1%)

[49,50,51, 55]

 Likes, shares, etc.

15 (41.7%)

[20, 24, 25, 32, 34, 39, 41, 43, 47,48,49,50, 53, 55, 56]

 Links

3 (8.3%)

[28, 33, 46]

 Social media profile/demographics

14 (38.9%)

[19, 20, 24, 26, 28, 33, 39, 41, 44,45,46,47, 49, 51]

 Co-occurring hashtags

1 (2.8%)

[41]

 Follower count

1 (2.8%)

[26]