The psychometric properties of the parent-child relationship questionnaire.
Conduct disorders in children are very common. The consequences of such problems
are far-reaching not least because they impact upon the quality of the parent-child
relationship. Parent Management Training (PMT) is a widely recognised treatment for
childhood conduct disorders although the outcomes for many children are
disappointing. Theoretically, the current study considers why outcomes are not more
favourable and illustrates that attention should be paid to the role of maternal
cognition in the quality of mother-child relationships. The aims of this study were to:
• orientate the reader to the nature of the problem
• describe the development of a tool that measures the quality of the mother-child
relationship, The Parent-Child Relationship Questionnaire (PCRQ)
• assess the psychometric properties of the PCRQ
• discuss the findings, strengths and weaknesses of the study and consider future
Control mothers completed the PCRQ once. Referred mothers completed the PCRQ
and a measure of depression pre and post PMT. These mothers also completed a
measure of parenting stress pre group only. A further sample of mothers completed
the PCRQ twice.
Principal components analysis was used to determine a) the factor structure of the
PCRQ and b) whether this differed between a control and referred sample. One factor
was extracted from the control sample and two from the referred. The two factors
extracted were named Negative Cognition and Emotion (NC-E) and Behavioural
Reliability analysis revealed adequate levels of alpha (Cronbach's) for the PCRQ and
NC-E subscale although only barely adequate for BM subscale in both samples. Testretest
reliability analysis on a small sample revealed a high Pearson's correlation
coefficient. Intra-class correlation coefficients for total PCRQ and the BM subscale
were reasonable whereas for NC-E the coefficient was weak.
Criterion validity was explored by comparing the control and referred sample means.
The differences between the means were significant.
Concurrent and convergent validity were investigated by computing correlations
between the PCRQ and the Parenting Stress Index Short-Form (PSI-SF; Abidin,
1995). This showed moderate correlations between the measures. The correlations
between two PSI-SF dimensions and the BM subscale were low. Discriminant validity
was investigated by computing correlations between the PCRQ and the Beck
Depression Inventory (BDI; Beck & Steer, 1993). This showed a moderate
relationship. Sensitivity to change following PMT was investigated by computing a
paired samples t test in the referred sample. This test was also used to compare the pre
and post group scores for the individual subscales. The changes were significant.
There was no relationship between PCRQ scores and either age or gender.
Normative values for the PCRQ and its subscales were calculated using a theoretical
The results suggest that the PCRQ can be used to investigate the extent of negative
maternal cognition underlying the quality of the mother-child relationship. Its
usefulness is enhanced by employing a measure of maternal depression.