Skip to main content

Table 2 Modelled student and school level characteristics associated with assent to data linkage

From: Sources of potential bias when combining routine data linkage and a national survey of secondary school-aged children: a record linkage study

   Data linkage sub-sample (n = 18,956)  
   Bivariate models Multivariable
model
  Consented to data linkage, n/N (%) OR (95% CIs) Wald test AOR (95% CIs)c Wald test
Student-level
 Gender
  Male 4654/9545 (48.8) 1.00   1.00  
  Female 4465/9064 (49.3) 1.01 [0.95, 1.07]   1.02 [0.95, 1.08]  
  Did not state 113/347 (32.6) 0.52 [0.41, 0.65] X2(2) = 31.55 0.65 [0.50, 0.86] X2(2) = 10.26
 School grade
  Year 7 1980/3787 (52.3) 1.00   1.00  
  Year 8 2125/4122 (51.6) 0.97 [0.89, 1.07]   0.99 [0.90, 1.09]  
  Year 9 1969/4347 (45.3) 0.74 [0.68, 0.81]   0.76 [0.69, 0.83]  
  Year 10 1696/3500 (48.5) 0.88 [0.80, 0.96]   0.93 [0.84, 1.03]  
  Year 11 1462/3200 (45.7) 0.77 [0.70, 0.85] X2(4) = 67.55 0.81 [0.73, 0.90] X2(4) = 51.32
 Weekly smoker
  No 8794/17,656 (49.8) 1.00   1.00  
  Yes 219/571 (38.4) 0.66 [0.55, 0.78]   1.02 [0.81, 1.30]  
 Past month cannabis use
  No 8848/17,799 (49.7) 1.00   1.00  
  Yes 279/699 (39.9) 0.67 [0.57, 0.78]   0.86 [0.70, 1.06]  
 Ever truant
  No 6988/13,666 (51.1) 1.00   1.00  
  Yes 2007/4455 (45.1) 0.78 [0.73, 0.84]   0.91 [0.85, 0.99]  
 Ever excluded
  No 8633/17,216 (50.2) 1.00   1.00  
  Yes 511/1304 (39.2) 0.65 [0.58, 0.74]   0.82 [0.72, 0.94]  
SWEMWBS (sd.) 22.415 (4.23) 1.03 [1.03, 1.04]   1.02 [1.01, 1.03]  
FAS (sd.) 9.457 (2.29) 1.03 [1.02, 1.05]   1.02 [1.01, 1.04]  
School-level
 Mean FAS (sd.) 9.397 (0.50) 1.38 [1.04, 1.85]   1.36 [1.02, 1.83]  
ICC – constant only 0.071     
ICC – Student level 0.071     
ICC – Level 1 & 2 variables 0.064     
  1. Notes: P < 0.05; SWEMWBS, Short Warwick-Edinburgh Mental Wellbeing Scale; FAS, Family Affluence Scale, Intraclass correlation coefficient (ICC)
  2. cAdjusted model: n = 17,256; schools = 39; observations per school: min = 121; max = 1003
\