This data set is about the evolution of a friendship network and delinquent behavior of pupils in school classes, collected in the Dutch Social Behavior study, a two-wave survey in classrooms (Houtzager and Baerveldt, 1999).

These data are from classrooms of the *MAVO * track,
the lower middle level of the Dutch secondary school system,
in which the pupils filled in a
questionnaire in the 3^{d} and 4^{th} years,
with about one year in between.

A further description of the data can be found here.

The network data files for groups 1, 3, 4, and 6 are included in the
*RSiena* package as objects N3401, N3403, N3404, N3406 (wave 1)
and HN3401, HN3403, HN3404, HN3406 (wave 2).

Script *RscriptMultipleGroups.R* at the Siena scripts website
uses this data set (groups 1, 3, 4, 6).

** Data files**:

The data files are for 19 schools, numbered
**h** = 1, 3, 4, 6, 7, 8, 9, 10, 11, 13, 14, 15, 16, 18, 19, 20, 21, 22, 23.
For each of these values of **h**, the following files are available.
Thus for the first school the files are
N34_1.DAT, HN34_1.DAT, CBE1.DAT, cbc1.dat, and cbe1.sim;
and similarly for the other school numbers.

- Network data files (adjacency matrices):

wave 1:`N34_`

(grade 3)**h**.DAT

wave 2:`HN34_`

(grade 4)**h**.DAT

The relation is defined as giving and receiving emotional support: there is a tie from pupil*i*to pupil*j*if*i*says that he/she receives and/or gives emotional support from/to pupil*j*. - Actor attributes

`CBE`

**h**.DAT

with the variables, respectively:- gender (1 = girls, 2 = boys);
- a measure of delinquent behavior, measured at the first wave;
number of minor offences which the respondent states to have committed,
transformed by the formula ln(1+
*x*) to correct the skewness of the distribution; - importance of school friends (varying from 1 = very important to 4 = unimportant).

- Changing actor behavior

`cbc`

**h**.dat

the same measure of delinquent behavior, measured at waves 1 and 2, transformed by ln(1+*x*), but now also rounded to integer values. - Dyadic covariate

`cbe`

**h**.sim

defined as 1 if the pupils have the same ethnic background, and 0 otherwise. (For group h = 23, the number of cases in cbe23.sim was 72 instead of 73; one last row and column with zeros has been added to allow this matrix to be used in example analyses.)

Chris Baerveldt, Beate Völker, and Ronan Van Rossem (2008).
Revisiting selection and influence: an inquiry into the friendship networks of high school students
and their association with delinquency.
*Canadian Journal of Criminology and Criminal Justice*, 50, 559-587.

http://dx.doi.org/10.3138/cjccj.50.5.559.

(Note: the quadratic shape function has been mistakenly omitted from the specification of the behavior objective function in the analyses reported here.)

Houtzager, B. & Baerveldt, C. (1999). Just like Normal. A Social Network
Study of the Relation between Petty Crime and the Intimacy of Adolescent
Friendships. *Social Behavior and Personality 27(2)*, 177-192.

Snijders, Tom A.B, and Baerveldt, Chris (2003).
A Multilevel Network Study of the Effects
of Delinquent Behavior on Friendship Evolution.
*Journal of Mathematical Sociology* 27, 123-151.

Back to the Siena data sets page