Table 2 Demographic and Twitter use characteristics of sample.

From: Using language in social media posts to study the network dynamics of depression longitudinally

 

Full Sample N = 946

No Depressive Episode N = 388

Depressive Episode N = 558

p-value

Twitter Behaviour

    Tweets

356.1 (581.4)

306.1 (517.2)

390.8 (620.2)

0.03*

    Retweets

276.0 (494.1)

246.7 (468.0)

296.5 (510.8)

0.13

    Likes

1097.7 (1062.7)

983.4 (996.6)

1177.2 (1100.2)

0.006**

Recruitment Type (% Paid)

680 (71.9%)

274 (70.6%)

406 (72.8%)

0.52a

Word count per day

131.6 (127.7)

115.2 (114.8)

143.1 (134.9)

0.001**

Age (years)

29.6 (10.6)

30.8 (11.3)

28.7 (10.1)

0.002

Gender

    Male

304 (32.1%)

147 (37.9%)

157 (28.1%)

<0.001***a

    Female

617 (65.2%)

240 (61.9%)

377 (67.6%)

–

    Transgender Male

6 (0.6%)

0 (0%)

6 (1.1%)

–

    Transgender Female

1 (0.1%)

1 (0.3%)

0 (0%)

–

    Non-Binary

15 (1.6%)

0 (0%)

15 (2.7%)

–

    Other

3 (0.3%)

0 (0%)

3 (0.5%)

–

Country

    Ireland

44 (4.7%)

24 (6.2%)

20 (3.4%)

0.15a

    United Kingdom

339 (35.8%)

147 (37.9%)

192 (34.4%)

–

    United States

480 (50.7%)

188 (48.5%)

292 (52.3%)

–

    Canada

42 (4.4%)

13 (3.4%)

29 (5.2%)

–

    Australia

15 (1.6%)

4 (1.0%)

11 (2.0%)

–

    Other

25 (2.6%)

12 (3.1%)

14 (2.5%)

–

Education

    Less than high school

16 (1.7%)

4 (1.0%)

12 (2.2%)

<0.001***a

    High school

190 (20.1%)

66 (17.0%)

124 (22.2%)

 

    Some university

317 (33.5%)

108 (27.8%)

209 (37.5%)

–

    Bachelor’s degree

285 (30.1%)

142 (36.6%)

143 (25.6%)

–

    Master’s degree

104 (11.0%)

53 (13.7%)

51 (9.1%)

–

    Professional degree

15 (1.6%)

4 (1.0%)

11 (2.0%)

–

    Doctorate

19 (2.0%)

11 (2.8%)

8 (1.4%)

–

Currently Employed (% Yes)

458 (48.4%)

204 (52.6%)

254 (45.5%)

0.04*a

Physician diagnosed depression (% Yes)

432 (45.7%)

76 (19.6%)

356 (63.8%)

<0.001***a

  1. Twitter and demographic characteristics of all participants along with differences between participants with and without a depressive episode.
  2. *p < 0.05, **p < 0.01, ***p < 0.001, aChi-square test.