Correction to: Changing intergenerational patterns of emotional dysregulation in families with perinatal borderline personality disorder

Correction to: Archives of Women’s Mental Health (2021) 24:641–648

https://doi.org/10.1007/s00737-021-01119-2

The original version of this article, unfortunately, contained several mistakes. A data error was made when Wilcoxon Signed Rank effect sizes were calculated using the formula r = z / √N. In each calculation, N was incorrectly entered as the number of cases in each analysis (e.g., N = 58), rather than the number of observations (i.e., N = 116, given the study was a pre-post design). This resulted in incorrect effect sizes being generated, however, the positive impact of the MI-DBT intervention on mothers and their infants was maintained in all instances. The result of this error was that the reported clinical significance of the intervention was generally reduced slightly (i.e., effect sizes decreased).

Additionally, the effect size that was calculated in the paper (r = z / √N) was incorrectly described in the manuscript text as a Pearson correlation, however should have been described as an effect size for the Wilcoxon Signed Rank test.

• 1. Page 644, Statistical analysis section: "Pearson correlation effect sizes were calculated in order to determine the magnitude of the change between pre- and post- group scores, with r = 0.10, 0.30, and 0.50 indicating small, medium, and large changes, respectively (Cohen 1992). Should be changed to: "Effect sizes for the Wilcoxon Signed Rank test (r = z / √N) were calculated in order to determine the magnitude of the change between pre- and post-group scores, with r = 0.10, 0.30, and 0.50 indicating small, medium, and large changes, respectively (Cohen 1992).

• 2. Page 644, Maternal mental health section: “Moreover, moderate to large effect sizes for each of these measures (r_range -0.36 to -0.52) were found.” Should be changed to: “Moreover, moderate effect sizes for each of these measures (r_range 0.25 to 0.36) were found.”

• 3. Page 644, Use of DBT skills section: “The findings for each of the three subscales of the DBT Ways of Coping checklist (adap-tive skills use, maladaptive general dysfunction, blaming others) were statistically significant and, moreover, a large effect ranging from -0.59 to -0.65 was found for each (see Table2).” Should be changed to: “The findings for each of the three subscales of the DBT Ways of Coping checklist (adaptive skills use, maladaptive general dysfunction, blaming others) were statistically significant and, moreover, a moderate-large effect ranging from 0.42 to 0.46 was found for each (see Table 2).”

• 4. Page 644, Parenting capability measures section: “Similarly, results on measures of parental capability as assessed by the Parenting Sense of Competence scale were significant, with moderate to large effects found for the over-all measure and each of the three subscales.” Should be changed to: “Similarly, results on measures of parental capability as assessed by the Parenting Sense of Competence scale were significant, with moderate effects found for the over-all measure and each of the three subscales.”

• 5. Page 644, Parenting capability measures section: “Moth-ers made fewer malevolent attributions about their child’s behavior (p = 0.002; r = -0.41) and were more inter-ested in the way their child behaved and felt (p = 0.000; r = -0.59).” Should be changed to: “Mothers made fewer malevolent attributions about their child’s behavior (p = 0.002; r = 0.29) and were more interested in the way their child behaved and felt (p = 0.000; r = 0.42).”

• 6. Page 644, Parenting capability measures section: “The extent of parent–child dysfunctional interactions was the only subscale on the Parenting Stress Index which was significant (p = 0.016), indicating that the mother viewed her child as less of a negative element in her life, although the effect size was small (r = -0.06). Nevertheless, parental distress has reduced with findings indicating a moderate-large effect (r = -0.43); however, it is possible that statistical power was limited by the small sample (n = 32), with this measure being introduced later in the project.” Should be changed to: “The extent of parent–child dysfunctional interactions was the only subscale on the Parenting Stress Index which was significant (p = 0.016), indicating that the mother viewed her child as less of a negative element in her life, with a moderate effect size (r = 0.30). Nevertheless, parental distress has reduced with findings indicating a small-moderate effect (r = 0.20); however, it is possible that statistical power was limited by the small sample (n = 32), with this measure being introduced later in the project.”

• 7. Page 645, Mother-infant interactions section: “There were no significant differences between pre- and post-group ratings of mother-infant interaction quality as measured by the NCAST, with effect sizes ranging from negligible (r = -0.02) to small (r = -0.18) (see Table 2).” Should be changed to: “There were no significant differences between pre- and post-group ratings of mother-infant interaction quality as measured by the NCAST, with effect sizes ranging from negligible (r = 0.01) to small (r = 0.13) (see Table 2).”

• 8. Page 645, Table 2:

Table 2 Comparison of pre- & post-group scores for women completing MI-DBT

Should be changed to:

Table 2 Comparison of pre- and post-group scores for women completing MI-DBT

The original article has been corrected.