Example L3 Continuous response growth model

 

Snijders & Bosker, (1999) analysed the development over time of teacher evaluations.  Starting from the first year of their career, teachers were evaluated on their interpersonal behaviour in the classroom.  This happened repeatedly, at intervals of about one year.  In this example, results are presented about the 'proximity' dimension, representing the degree of cooperation or closeness between a teacher and his or her students.  The higher the proximity score of a teacher, the more cooperation is perceived by his or her students.

 

There are four measurement occasions: after 0, 1, 2, and 3 years of experience.  A total of 51 teachers were studied. The number of observations for the 4 moments decreased from 46 at time i=0 to 32 at time i=3. The non-response at various is treated as ignorable.

 

 

Reference

 

Snijders, T. A. B. and Bosker, R. J. (1999). Multilevel Analysis. London: Sage.

 

 

Data description

 

Number of observations (rows): 153

Number of variables (columns): 8

 

The subset of variables we use are:

 

teacher =  teacher identifier

time = 0,1,2,3 the year at which the teacher evaluation was made

proximity = degree of cooperation or closeness between a teacher and his or her students

gender = 1 if teacher is female, 0 otherwise

d1 =  1 if time =0, 0 otherwise

d2 =  1 if time =1, 0 otherwise

d3 =  1 if time =2, 0 otherwise

d4 =  1 if time =3, 0 otherwise

 

 

The first few lines of growth.dat look like

 

 

 

Sabre commands

 

out growth.log

trace growth.trace

data teacher time proximity gender d1 d2 d3 d4

read growth.dat

case teacher

yvar proximity

family g

constant cons

lfit cons

dis m

dis e

mass 64

scale 0.5

fit cons

dis m

dis e

lfit d1 d2 d3 d4

dis m

dis e

sigma 0.25

scale 0.5

fit d1 d2 d3 d4

dis m

dis e

stop

 

 

Sabre log file

 

<S> trace growth.trace

<S> data teacher time proximity gender d1 d2 d3 d4

<S> read growth.dat

 

        153 observations in dataset

 

<S> case teacher

<S> yvar proximity

<S> family g

<S> constant cons

<S> lfit cons

 

    Iteration       Log. lik.       Difference

    __________________________________________

        1          -88.715008

 

<S> dis m

 

    X-vars            Y-var

    ______________________________

    cons              proximity

 

    Univariate model

    Standard linear

 

    Number of observations             =     153

 

    X-var df           =     1

 

    Log likelihood =     -88.715008     on     152 residual degrees of freedom

 

<S> dis e

 

    Parameter              Estimate         Std. Err.

    ___________________________________________________

    cons                   0.63856          0.35048E-01

    sigma                  0.43352

 

<S> mass 64

<S> scale 0.5

<S> fit cons

 

    Initial Homogeneous Fit:

 

    Iteration       Log. lik.       Difference

    __________________________________________

        1          -88.715008

 

 

    Iteration       Log. lik.         Step      End-points     Orthogonality

                                     length    0          1      criterion

    ________________________________________________________________________

        1          -84.581007        1.0000    fixed  fixed       136.94

        2          -75.772623        0.5000    fixed  fixed       1298.9

        3          -64.493890        1.0000    fixed  fixed       741.33

        4          -61.916672        1.0000    fixed  fixed       606.92

        5          -61.737341        1.0000    fixed  fixed       1198.4

        6          -61.702731        1.0000    fixed  fixed       661.65

        7          -61.688023        1.0000    fixed  fixed       1557.1

        8          -61.688018        1.0000    fixed  fixed       1843.3

        9          -61.688018        1.0000    fixed  fixed

 

<S> dis m

 

    X-vars            Y-var             Case-var

    ________________________________________________

    cons              proximity         teacher

 

    Univariate model

    Standard linear

    Gaussian random effects

 

    Number of observations             =     153

    Number of cases                    =      51

 

    X-var df           =     1

    Sigma df           =     1

    Scale df           =     1

 

    Log likelihood =     -61.688018     on     150 residual degrees of freedom

 

<S> dis e

 

    Parameter              Estimate         Std. Err.

    ___________________________________________________

    cons                   0.64795          0.53346E-01

    sigma                  0.27155          0.19025E-01

    scale                  0.34388          0.42213E-01

 

<S> lfit d1 d2 d3 d4

 

    Iteration       Log. lik.       Difference

    __________________________________________

        1          -87.536537

 

<S> dis m

 

    X-vars            Y-var

    ______________________________

    d1                proximity

    d2

    d3

    d4

 

    Univariate model

    Standard linear

 

    Number of observations             =     153

 

    X-var df           =     4

 

    Log likelihood =     -87.536537     on     149 residual degrees of freedom

 

<S> dis e

 

    Parameter              Estimate         Std. Err.

    ___________________________________________________

    d1                     0.58652          0.64064E-01

    d2                     0.72395          0.70485E-01

    d3                     0.64378          0.71431E-01

    d4                     0.60594          0.76810E-01

    sigma                  0.43450

 

<S> sigma 0.25

<S> scale 0.5

<S> fit d1 d2 d3 d4

 

    Initial Homogeneous Fit:

 

    Iteration       Log. lik.       Difference

    __________________________________________

        1          -87.536537

 

 

    Iteration       Log. lik.         Step      End-points     Orthogonality

                                     length    0          1      criterion

    ________________________________________________________________________

        1          -64.095200        1.0000    fixed  fixed       94.548

        2          -63.262786        0.5000    fixed  fixed       773.48

        3          -60.084462        1.0000    fixed  fixed       507.61

        4          -59.421825        1.0000    fixed  fixed       182.88

        5          -59.263588        1.0000    fixed  fixed       310.69

        6          -59.204119        1.0000    fixed  fixed       204.81

        7          -59.146517        1.0000    fixed  fixed       335.13

        8          -59.146452        1.0000    fixed  fixed       348.43

        9          -59.146452        1.0000    fixed  fixed

 

<S> dis m

 

    X-vars            Y-var             Case-var

    ________________________________________________

    d1                proximity         teacher

    d2

    d3

    d4

 

    Univariate model

    Standard linear

    Gaussian random effects

 

    Number of observations             =     153

    Number of cases                    =      51

 

    X-var df           =     4

    Sigma df           =     1

    Scale df           =     1

 

    Log likelihood =     -59.146452     on     147 residual degrees of freedom

 

<S> dis e

 

    Parameter              Estimate         Std. Err.

    ___________________________________________________

    d1                     0.58508          0.62624E-01

    d2                     0.71760          0.66132E-01

    d3                     0.67158          0.66631E-01

    d4                     0.63893          0.69505E-01

    sigma                  0.26500          0.18573E-01

    scale                  0.34532          0.41967E-01

 

<S> stop