The Strange Situation procedure was designed to assess the quality of infant--mother attachment and is considered appropriate for infants between 12 and 18 months (for details see, Ainsworth, 1978). The procedure consists of seven three-minute episodes arranged to continuously increase the stress , so that the infant's attachment system with respect to the mother is activated. The crucial episodes, the reunion episodes (R1 and R2), are those in which the mother returns after having been away, and the infant was left alone with a stranger. On the basis of the infant's behavior, especially during these episodes, the quality of the infant--mother attachment relationship is generally categorized as insecure-avoidant (A), secure (B),or insecure-resistant (C). According to the classification instructions (see Patterns of attachment, Ainsworth, M. D. S. 1978, pp. 59--63) the scores on five 7-point scales in the two reunion episodes play a crucial role in the clinical classification, that is, the Proximity Seeking (PS), Contact Maintaining (CM), Resistance (RS), Avoidance (AV), and Distance Interaction (DI) scales, where high scores on Avoidance are especially indicative for an A classification, and high scores on Resistance for a C classification. The aim of a three-mode cluster analysis on these data may be assessing whether the clinical A,B,C typology corresponds to the statistical grouping derived via the mixture method of clustering. A full three-mode analysis of these data was given by Kroonenberg, Basford, & Van Dam (1995).
De data set consists of 326 infants' scores on five 'interactive' scales during the two reunion episodes when the infant is reunited with its mother after having been alone with a stranger. Ths scales run from 1 to 7 with 7 indicating the highest intensity of the reaction.
|1.||R1||First reunion episode (often designated with M5)|
|2.||R2||Second reunion episode (often designated with M8)|
A three-way data array X = (x(i,j,k)) has the following form
|-----|i=1 |-----| |i=2 |-----| | |.. | | | |.. | | |____|i=I=326 k=K=2 | |____| k=2 |_____| k=1 j=1,.,J=5
The actual data file has the following form:
j=1,.,J=5 |-----|i=1 | |i=2 | |.. k=1 | |.. |_____|i=I=326 |-----|i=1 | |i=2 | |.. k=2 | |.. |_____|i=I=326
Thus the first mode (i) is nested in the third mode (k) and there are 326 (Infants) times 2 (Episodes) rows and 5 (Scales) columns.
In mixture method of clustering the means and (co)variances are estimated, so that there is not preprocessing. For details see Kroonenberg (2008). Applied multiway data analysis. Hoboken NJ: Wiley (Chapters 6 and 16).
[Download the zipped Strange Situation Data]