Table 3 Associations between Cluster Membership and Participant Outcomes

From: Identifying therapeutic characteristics of digital social media narratives about suicide: a mixed methods investigation

 

T2 Suicidal Thought

T2 Social Connectedness

T2 Optimism

 

β

CI

p

β

CI

p

β

CI

p

Cluster 1 (vs. 3)

0.15

[0.04, 0.26]

0.006

0.06

[-0.08, 0.20]

0.431

−0.12

[−0.23, −0.01]

0.039

Cluster 2 (vs. 3)

0.07

[−0.01, 0.16]

0.101

0.15

[0.03, 0.26]

.014

−0.06

[−0.15, 0.03]

0.192

T1 Suicidal Thoughts

0.89

[0.86, 0.91]

<.001

-

-

-

-

-

-

T1 Social Connectedness

-

-

-

0.78

[0.74, 0.81]

<0.001

-

-

-

T1 Optimism

-

-

-

-

-

-

0.87

[0.85, 0.90]

<0.001

  1. Note: This table displays the result of three ANCOVA models testing associations between clusters of suicide narratives and the three outcomes of interest, suicidal thoughts, social connectedness, and optimism, each rated after the participant read a narrative (T2). In each analysis, we controlled for the rating of the outcome of interest prior to reading the narrative (T1). We used dummy coding for Cluster membership, with Cluster 3 as the reference level. Thus, Associations between Cluster 1 or 2 and each outcome is relative to the association between Cluster 3 and that outcome. Due to this dummy coding scheme, the pairwise contrast between Cluster 1 and 2 is not calculated. We performed alternative tests with Cluster 1 set as the reference category to test the contrast between Cluster 1 and 2, which was only significant for the outcome T2 Suicidal Thoughts (β = −0.08, CI: [−0.15, −0.01], p = 0.032).
Back to article page