Skip to main content

Table 11 Latent variable structure, direct effects, and covariances the Empath relapse SEM model

From: Analysis of associations between emotions and activities of drug users and their addiction recovery tendencies from social media posts using structural equation modeling

Relationships between variables

Estimate

Standardized estimate

Standard error

Z value

P value

Latent variables

 Positive emotion΄—> 

Emotional

1.000

0.527

–

–

–

 Positive emotion΄—> 

Suffering

1.196

0.630

0.260

4.5

0.000

 Positive emotion΄—> 

Swearing_terms

1.231

0.649

0.268

4.6

0.000

 Career΄—> 

White_collar_job

1.000

0.995

–

–

–

 Career΄—> 

Blue_collar_job

0.951

0.946

0.088

10.8

0.000

Career΄—> 

office

0.134

0.133

0.077

1.7

0.082

Regressions

 Positive emotion΄—> 

Relapse

0.694

0.372

0.204

3.3

0.001

 Career΄—> 

Relapse

0.100

0.101

0.075

1.3

0.183

 Urban΄—> 

Relapse

0.144

0.147

0.068

2.1

0.034

Covaraiances

 Joy΄

Zest΄

0.597

0.937

0.092

6.5

0.000

 Joy΄

Cheerfulness

0.066

0.096

0.022

3.0

0.002

 White_collar_job΄

Blue_collar_job

0.290

0.526

0.085

3.4

0.001

 White_collar_job΄

Office΄

− 0.160

− 0.374

0.079

− 2.036

0.042

 Positive_emotion΄

Career΄

0.198

0.424

0.062

3.2

0.001

  1. The symbol ‘—> ’ is used to represent a path or direct effect in our SEM model. Emotional distress, career, rural, and weakness positively impacted relapse behavior, but the impact of career was statistically insignificant