Abstract
We investigate the degree of correspondence between parents’ reports on child behavioral and educational outcomes using wave four of a rich Danish longitudinal survey of children (the DALSC). All outcomes are measured at age 11 when the children are expected to be in fifth grade. Once discrepancies are detected, we analyze whether they are driven by noisy evaluations or by systematic bias, focusing on the role of parental characteristics and response heterogeneity. We then explicitly assess the relative importance of the mother’s versus the father’s assessments in explaining child academic performance and diagnosed mental health to investigate whether one parent is systematically a better informant of their child’s outcomes than the other. Our results show that parental psychopathology, measured as maternal distress, is a source of systematic misreporting of child functioning, that the parent–child relationship matters, and that mothers are not necessarily a better informant of child functioning than fathers. This last finding should not only be valid for Denmark but also for many other countries, where the father’s role in childcare has been growing.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
A literature review on the use of parental subjective assessments and on the analysis of parental discrepancies is reported in the on-line appendix.
This is especially true for internalizing or emotional problems, e.g. child anxiety and depression, which are less observable compared to the child externalizing problems, e.g. hyperactivity and oppositional behavior.
The DALSC is a repeated survey of the primary parent of, initially, 6000 children born between 15 September and 31 October 1995. Five waves are available, from 1996 (6 months), 1999 (3 years), 2003 (7 years), 2007 (11 years) and 2011 (15 years). The fathers of these children were surveyed separately in 1999 and 2007.
Specifically, we include three variables summarizing the beliefs of both parents about the level of parental disagreement: whether both parents report high frequency of parental discussions, whether only the mother reports parental discussions and whether only the father reports parental arguments.
The number of psychiatric diagnoses has been collected according to the ICD-10 Classification of Mental and Behavioral Disorders (World Health Organization 1992).
We include three variables summarizing the parents’ closeness to the child: whether both parents report having a very close relationship with the child, whether only the mother reports having a very close relationship with the child and whether only the father reports it.
The following activities are included in the survey: cooking, reading, doing homework and playing together.
In a more complete specification, we have also included: iv) whether the father is the breadwinner; v) whether the child was enrolled in a municipality provided daycare program at age 3 and vi) the hours of non-parental care. For the father to be the breadwinner, both his average labor market experience and his income over the period 2000-2006 have to be higher than those of the mother. The results from this augmented specification are reported in the on-line appendix and are very similar to the ones reported in the paper. We have decided to report the main results without these additional controls in a parsimonious specification as these characteristics may have a direct influence on the outcome of interest, i.e. child development scores.
The main descriptive statistics of all the variables used in our empirical analysis, separately by child gender and by whether disabled kids and non-intact households are included or not in the sample are reported in Tables A1 and A2 of the on-line appendix.
Responses range from “Very well” (coded 1) through “Badly” (coded 5), i.e. a higher score indicates worse child outcome as for the SDQ.
As in the case of the SDQ ratings, there is strong evidence of dissimilarity between mother and father evaluations in the pairwise correlations between respondents’ evaluations, as reported in Table A2 of the on-line appendix.
We have also estimated Eq. (1) separately for girls and boys. Results are very similar to the ones reported in the paper and are available on request from the authors.
As an additional attempt to partly address the issue of unobserved heterogeneity, we also estimate Eq. (1) by using quantile regression techniques. The coefficients obtained from the latter can in fact be interpreted as a sort of latent “propensity” to behave differently along the distribution of parental discrepancies, which are implicitly indexed by each quantile (Doksum 1974). See table A5 of the on-line appendix.
Child school performance is measured using parents' assessments of how well their child fares academically. Relatively few replied “average”, “not so well” and “badly” so we combined these three categories together and proceeded with 3 categories: (1) very well, (2) well, (3) average, not so well and badly.
For the full list of child-specific psychiatric disorders included, see the online appendix.
We perform a number of robustness checks and additional analyses on all three specifications, discussed thoroughly in the on-line appendix.
Several diagnostics indicates that multicollinearity is not a problem in our estimations. The variance inflation factor (VIF) is, for example, around 1.5 in the most complete specification, much below the threshold suggested by Allison (1999).
This result is obtained by standardizing the dependent variable. As the mean and the standard deviation of the dependent variable are respectively 0.651 and 3.705, an additional symptom is associated with approximately an 8 percent increase in the reporting gap between parents. Not all the y-standardized coefficients are reported in the paper, but all are available on request from the authors.
Consistently with these findings, we find that only mother’s symptoms are positively associated with their own assessments of the child behavior (see Table A9 of the on-line appendix).
We find that parental perception of conflicts modifies both mother’s and father’s ratings in the same direction (see Table A9 of the on-line appendix).
We find that parental perception of the relationship to the child modifies both mother’s and father’s ratings in the same direction (see Table A9 of the on-line appendix). However the association of this variable with father’s assessment is significantly larger compared to the one estimated on mother’s side. This may explain the reason why we still observe an independent association between father’s relationship to the child and the difference in parental evaluations.
Other results available on request also indicate that none of the variables related to the parents’ socio-economic status, such as differences in educational and ethnic background or household’s income, is significantly associated with the level of parental discrepancies. Finally, we generally do not find that the gender of the child, whether the child is the first one within the household, whether the child has a low birth weight, or the age difference within the couple influence parental discrepancies. The same results hold for the father’s and mother’s ratings separately (see Table A9 of the on-line appendix).
As mentioned in Sect. 4, child academic performance is proxied by the respondent’s perception of how well the child fares academically, recalling that the perception score ranges from 1 (“Very well”) to 5 (“Badly”) so that a positive coefficient means that mothers rate academic performance harsher than fathers.
Looking at the probability that the mother assesses the child’s academic performance worse or better than the father provides qualitatively similar results (columns 5 and 6 of Table A7 of the on-line Appendix). Furthermore, parental assessments of academic performance disaggregated by subject corroborate these findings, especially strong for Science but also for the other subjects (see Table A8 of the on-line Appendix).
We are not able to report results by child gender as models separately estimated for boys and girls did not converge.
Note that in all specifications we include the square and the cube of the objective measures.
Note that the dependent variable ranges from “Very well” (coded 1) through “Badly” (coded 5). Hence a negative coefficient estimated on the objective scores is interpreted as a positive correlation.
Very similar results are obtained for the remaining threshold, i.e., the one for the category “Well” and are reported in Table A10 of the online appendix.
Very similar results are obtained by entering the mother’s and the father’s reports separately (see table A11 of the on-line appendix).
As a robustness check, we have also checked whether this assumption holds using both parents’ assessments of the child’s academic performance in Math and Danish, instead of the general ones. The results from this additional analysis indicate that the mother’s assessments do not receive a higher and statistically different weight in explaining child school functioning compared to those of the father’s.
References
Allison, P. D. (1999). Multiple regression: A primer. Thousand Oaks, CA: Pine Forge Press.
Almond, D., & Currie, J. (2010). human capital development before five. In D. Card & O. Ashenfelter (Eds.), Handbook of labor economics (Chapter 15) (Vol. 4). Amsterdam: Elsevier Science B.V.
Bago d’Uva, T., Lindeboom, M., O’Donnell, O., & van Doorslaer, E. (2011). Slipping anchor?: Testing the vignettes approach to identification and correction of reporting heterogeneity. Journal of Human Resources, 46(4), 875–906.
Becker, G. S., & Tomes, N. (1986). human capital and the rise and fall of families. Journal of Labor Economics, 4(3), S1–S39.
Blanden J., Gregg, P., & Macmillan, L. (2006). Explaining Intergenerational income persistence: Non-cognitive skills, ability and education. CMPO Working Paper Series 06/146, Department of Economics, University of Bristol, UK.
Borghans, L., Duckworth, A. L., Heckman, J. J., & ter Weel, B. (2008). The economics and psychology of personality traits. Journal of Human Resources, 43(4), 973–1059.
Case, A., Fertig, A., & Paxson, C. (2005). The lasting impact of childhood health and circumstance. Journal of Health Economics, 24(2), 365–389.
Dalton, D., & Ortegren, M. (2011). Gender differences in ethics research: The importance of controlling for the social desirability response bias. Journal of Business Ethics, 103(1), 73–93.
Doksum, K. (1974). Empirical probability plots and statistical inference for nonlinear models in the two sample case. Annals of Statistics, 2(2), 267–277.
Fisher, R. J. (1993). Social desirability bias and the validity of indirect questioning. Journal of Consumer Research, 20(2), 303–315.
Goodman, R. (1997). The Strengths and Difficulties Questionnaire: A research note. Journal of Child Psychology and Psychiatry, 38(5), 581–586.
Kalenkoski, C., Ribar, D., & Stratton, L. (2007). The effect of family structure on parents’ child care time in the United States and the United Kingdom. Review of Economics of the Household, 5(4), 353–384.
Kan, M. L., Sullivan, O., & Gershuny, J. (2011). Gender convergence in domestic work: Discerning the effects of interactional and institutional barriers from large-scale data. Sociology, 45(2), 234–251.
Kerkhofs, M., & Lindeboom, M. (1995). Subjective health measures and state dependent reporting errors. Health Economics, 4(3), 221–235.
Lindeboom, M., & van Doorslaer, E. (2004). Cut-point shift and index shift in self-reported health. Journal of Health Economics, 23(6), 1083–1099.
McLeod, J. D., & Kaiser, K. (2004). Childhood emotional and behavioral problems and educational attainment. American Sociological Review, 69(5), 636–658.
Prendergast, C. (2002). Consumers and agency problems. The Economic Journal, 112(478), 34–51.
Prendergast, C., & Topel, R. (1996). Favoritism in organizations. Journal of Political Economy, 104(5), 958–978.
Sayer, L. C., Bianchi, S. M., & Robinson, J. P. (2004). Are parents investing less in children? Trends in mothers’ and fathers’ time with children. American Journal of Sociology, 110(1), 1–43.
Selfridge, A. (2007). Equal protection and gender preference in divorce contests over custody. Journal of Contemporary Legal Issues, 16, 165.
Silverman, W. K., & Rabian, B. (1995). Test–retest reliability of DSM-III-R anxiety disorders symptoms using the Anxiety Disorders Interview Schedule for Children. Journal of Anxiety Disorders, 9(2), 139–150.
Spearman, C. (1904). “General intelligence”, objectively determined and measured. American Journal of Psychology, 15, 201–293.
Thurstone, L. L. (1931). Multiple factor analysis. Psychological Review, 38, 406–427.
Treutler, C. M., & Epkins, C. C. (2003). Are discrepancies among child, mother, and father reports on children’s behavior related to parents’ psychological symptoms and aspects of parent-child relationships? Journal of Abnormal Child Psychology, 31(1), 13–27.
Tversky, B., & Marsh, E. J. (2000). Biased retellings of events yield biased memories. Cognitive Psychology, 40(1), 1–38.
World Health Organization. (1992). The ICD-10 classification of mental and behavioural disorders: Clinical descriptions and diagnostic guidelines. Geneva: World Health Organization.
Youngstrom, E. A., Izard, C., & Ackerman, B. (1999). Disphoria-related bias in maternal ratings of children. Journal of Consulting and Clinical Psychology, 67(6), 905–916.
Author information
Authors and Affiliations
Corresponding author
Electronic supplementary material
Below is the link to the electronic supplementary material.
Rights and permissions
About this article
Cite this article
Datta Gupta, N., Lausten, M. & Pozzoli, D. Does mother know best? Parental discrepancies in assessing child behavioral and educational outcomes. Rev Econ Household 16, 407–425 (2018). https://doi.org/10.1007/s11150-016-9341-1
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11150-016-9341-1