Skip to main content

Table 2 Multilevel logistic regression of adolescent suicidal ideation

From: The association between school class composition and suicidal ideation in late adolescence: Findings from the Young-HUNT 3 study

 

Model 1

Model 2

Individual-level variables

  

Gender (ref.: boys)

1.28 (1.07-1.54)**

1.10 (0.88-1.37)

Age (ref.: 16)

  
 

17

1.05 (0.83-1.33)

1.30 (0.97-1.75)

 

18

0.88 (0.68-1.14)

1.77 (1.22-2.57)**

 

19

0.89 (0.59-1.34)

1.85 (1.13-3.05)*

Parent education (ref.: low)

0.88 (0.68-1.14)

0.99 (0.80-1.21)

Living arrangement (ref.: both parents)

  
 

One parent

1.40 (1.01-1.91)*

1.41 (1.03-1.94)*

 

Away from home

1.43 (1.08-1.89)**

1.42 (1.07-1.87)*

 

Other

2.45 (1.61-3.72)***

2.38 (1.56-3.62)***

Divorced parents (ref.: not divorced)

1.24 (0.94-1.65)

1.20 (0.91-1.58)

School class-level variables

  

Vocational subject

 

1.46 (1.16-1.84)**

Proportion of parents with higher education

 

1.07 (0.65-1.84)

Proportion of girls

 

1.93 (1.33-2.79)***

School grade (ref.: First grade)

  
 

Second grade

 

0.77 (0.58-1.02)

 

Third grade

 

0.37 (0.25-0.54)***

Variance level 2 (Std. Error)

0.11 (0.06)

0.02 (0.030)

ICC (%)

3.2

0.6

Deviance

3088.1

3043.7

  1. *: P < .05; **: P < .01; ***: P < 0.001. OR (95% CI).