DeKlyen, M. (1996). Disruptive Behavior Disorder and Intergenerational Attachment Patterns: A Comparison of Clinic-Referred and Normally Functioning Preschoolers and Their Mothers. Journal of Consulting and Clinical Psychology, 64, 357-365.

This study examined the linkages between child disruptive behavior disorder (DBD), quality of mother-child interactions, and mothers' recollections of and attitudes toward their own parents.  Twenty-five preschool boys referred to a psychiatric clinic were matched with normally functioning boys.  Mothers and sons were videotaped during a separation-reunion sequence, the Adult Attachment Interview was administered to mothers, and mothers completed questionnaires assessing family environment.

Mothers with secure internal models of attachment were predicted to differ from insecure mothers on several family and environmental variables, on the basis of hypothesized associations between attachment and intimate relationships, caregiving attitudes, and psychopathology. 

These predictions were tested in two stages.  First, one-way analysis of variance (ANOVAs) and chi-square values were generated to explore relations between a dichotomous measure of mothers' attachment classification (secure vs. insecure) and family-environmental variables.  Analyses revealed an unexpected relationship between mothers' education level and attachment, F(1, 48) = 7.5, p < .01.  Secure maternal attachment was expected to be associated with marital satisfaction, with social supports, and general life satisfaction. Both support from intimates and general life satisfaction were related to maternal attachment, F(1, 47) = 6.7, p = .01, and F(1, 45) = 9.0, p < .005, respectively.  The score for Depression was significantly higher for insecure mothers, F(1, 47) = 5.5, p = .02, whereas Hostility was in the predicted direction but failed to reach significance.  As predicted, mothers with insecure attachments reported more negative life stresses within the past year, F(1, 46) = 4.6, p = .04.

The analyses show that there was a difference between maternal attachment and other family and environmental variables, including depression. 

---

Thanks for the example, Debra.  I have a few comments.  First, note that the grouping ("independent") variable is dichotomous -- 'secure' and 'insecure' attachment.  The ANOVAs are thus absolutely equivalent to t tests, which normally would be the statistic reported.  Of greater concern is that the analyses reported here are independent samples analyses (as I can tell from the reported df respective to reported sample sizes), but the authors state that their samples were matched (but on what I cannot tell).  If they had matched samples, they should have employed analyses appropriate for matched samples (correlated t tests or the ANOVA equivalent).  There are multiple "dependent" (continuous) variables (family-environmental variables), but no effort was made to control familywise error (such as making a Bonferroni adjustment).  I think such adjustments are generally too conservative, but others insist on them.  A multivariate analysis (MANOVA or Hotellings T) would be, IMHO, a more appropriate analysis here.  Note that inferential statistics are given for those associations which were declared statistically "significant," but not for those that failed the .05 test -- apparently these authors believe that if p > .05 then the effect is nil.  The statistics reported in this summary tell us next to nothing about the magnitude of the effects and we are left guessing with respect to some of the directions of effect.  I assume that the authors provided a table of means and standard deviations from which the reader could determine these all-important characteristics of the research (?).  Also, I wonder about the grouping variable -- did the authors originally have a continous variable which they converted to a dichotomous variable?  If so, why degrade the information in the continuous variable to a mere dichotomy?

Contact Information for the Webmaster,
Dr. Karl L. Wuensch



This page most recently revised on 18. February 2008.