Skip to main content
. Author manuscript; available in PMC: 2009 Sep 14.
Published in final edited form as: Elem Sch J. 2009 Jan 1;109(3):251–266. doi: 10.1086/592306

Table 2.

Hierarchical Logistic Regression Analyses Predicting Retention from Child and School Variables

Model N ba s(b) Sig.b Effectc R2 totald Δχ2 Sig. (Δχ2) df
1 Academic competence: 769 0.248 134.62e .000 5
    Unit IQ −0.001 0.012 0.934 −0.003
    Woodcock-Johnson
      Reading W −0.048 0.011 0.000 −0.155
      Math W −0.007 0.014 0.617 −0.018
    Teacher-rated:
        Engagement −0.071 0.160 0.657 −0.02
        Achievement −0.512 0.206 0.012 −0.089
2 Sociodemographics: 769 0.304 28.54 .000 3
    Economic
    disadvantage 0.291 0.341 0.393 0.039
2 Sociodemographics:
    % Female 0.020 0.282 0.943 0.003
    Age in months −1.578 0.420 0.000 −0.169
3 Social, emotional and behavioral characteristics: 769 0.262 7.882 .163 5
    Ego resilience 0.134 0.185 0.469 0.033
    Aggression −0.176 0.191 0.357 −0.042
    Hyperactivity 0.052 0.178 0.770 0.013
    Emotional problems −0.077 0.183 0.674 −0.019
    Prosocial behaviors −0.014 0.182 0.939 −0.004
5 School context: 769 0.252 2.112 .550 3
    School % economic Disadvantage 0.009 0.010 0.368 0.041
    Teacher-rated support 0.018 0.199 .928 −0.004
    Teacher-rated conflict −0.119 0 .139 .391 −0.039
6 Home environment: 480 0.344 25.85 .001 6
    Parent-school:
    Communication 0.956 0.338 .005 0.128
    Positive perception 0.963 0.314 .002 0.139
        Sense of shared
        responsibility −0.590 0.508 .245 −0.053
        Home
        involvement 0.538 0.408 .187 0.060
    Educational aspiration 0.024 0.126 .849 0.009
    Mobility 0.832 0.509 .102 0.074
a

Regression coefficient predicting log-odds ratio

b

Asymptotic normal distribution probability for b/s(b)

c

Effect size as correlation computed as ratio of b/s(b) divided by square root of b/s(b) squared plus N

d

R-square computed using Nagelkerke likelihood ratio statistic

e

Model 1 involves only variables in the academic competence block. In each of the other models the variables were added hierarchically and sequentially to the academic competence block. Thus, all models except Model 1 involved two blocks, with academic competence always the first block. For Model 1, the regression weights are based on simultaneous entry of all variables in block 1. For all other models the regression weights are those for the individual variable added to the academic competence block.