Using machine learning to study parenting styles

How should we raise our children? Research has shown that the amount of time invested by the parents is not the only crucial element for children’s skills development (Del Boca et al. 2014, Attanasio et al. 2016); parenting style is also important (Fiorini and Keane 2014). Parenting style is a strategic choice linked to incentives (Doepke and Zilibotti 2014). To study the relationship between parenting style and child development, researchers rely on ad hoc perceptions or previous research to limit the complexity of parenting to certain core actions. For example, reading to children appears to be highly predictive of the development of skills in children (Kalb and Jan van Ours 2013). But how can we measure parenting styles without relying on previous beliefs?

In a recent article (Rauh and Renée 2022) we opt for an approach that lets the data speak. We use a computational linguistics model to identify parenting styles that exist and are most prominent. The original purpose of the machine learning algorithm is to learn from the simultaneous occurrence of words and to form subjects around them. In the context of parenting styles, the idea is that parents engaged in one activity are more likely to engage in another particular activity, so the algorithm learns from the simultaneous occurrence of parental actions to define two or more parenting styles.


We rely on the Québec Longitudinal Study of Child Development (QLSCD), a detailed panel of a representative sample of families from Québec, a province in Canada, with a baby born between October 1997 and July 1998. We use information about maternal behavior collected over three waves when the target children were 5, 17 and 29 months old.

An advantage of the dataset is that the behavior of mothers towards their children is not recorded via self-reported survey questions, but via the enumerator. At the end of each interview, the enumerator records ten variables that classify whether the mother takes certain actions during the interview. This makes the data less prone to bias that can result from mothers misreporting how they behave towards their children.

Table 1 shows the list of ten activities and the percentage of mothers who indulged in them during the three interviews. Two things stand out. First, parents are generally more likely to be supportive by confirming progress or kissing and hugging the child rather than yelling or expressing annoyance at the child. Second, while this imbalance occurs in all three waves of research, there is a significant shift towards more punitive and less supportive measures as the child ages.

table 1 Parental actions in different survey waves

Comments:: The table describes the behavior of the respondents and their interactions with their children during the annual QLSCD interview in percentage. Behavior is assessed by the enumerator during the interview.

Parenting Styles

The model we use to classify parenting styles is the Latent Dirichlet Allocation (LDA), developed by Blei et al. (2003) for classifying text by subject. In recent work, economists have used LDA to determine conspicuous CEO behavior (Bandiera et al. 2020) and political ideologies (Draca and Schwarz 2021). In our case, each parenting style is determined by the probability of each parental action, and parents are classified into two styles. The classification is not ‘strict’ in the sense that parents can be a combination of the two types: the algorithm assigns shares of each style to parents in the dataset.

In Figure 1, we show the resulting distribution of parental actions when classifying parenting into two styles, A and B. The red bars indicate the prevalence of a particular action under parenting style A and the blue bars under parenting style B. The left panel shows the probability of each of ten actions within a subject. A parenting style should be imagined as an urn from which a parent draws actions. A mother who follows parenting style A is going to close her eyes and draw an action from the red bars. She will most likely draw the largest bars; for example, she may give a pedagogical toy to her child or respond to the sounds of her baby. A parenting style B mother is most likely not going to draw any action when drawing the blue bars in the urn because they are so small. In other words, parenting style B is characterized by passivity. Based on these distributions of actions, we follow developmental psychologists McCoby and Martin (1983) and label parenting style A as ‘warm’ and parenting style B as ‘cold’. In the right panel of Figure 1 we see the standardized relative prevalence of a particular act within a parenting style. We see that it is relatively unlikely that warm mothers will reprimand or yell at their child. Cold parents, if anything, just check on their kid.

Figure 1 Distribution of actions by parenting style

Comments:: The left pane describes the topic share of actions for each of the two parenting styles. The right panel shows the standardized importance of an action within a style by setting the mean to 0 and the standard deviation to 1.

Are all parents equally likely to follow a warm parenting style? To answer this question, we look at the relationship between warm parenting and parental and family characteristics. In figure 2 we see the distribution of parenting styles according to the mother’s upbringing. In the left panel we see that the cold parenting style is quite probable among low-educated mothers with at most a secondary education. This is indicated by the mass to the left of the graph. Looking at the right panel, we see the opposite for college-educated moms. For highly educated mothers, there is a shift in the distribution to the right; they are more likely to follow a warm parenting style. We also find that younger mothers and those with more children are more likely to have a cold parenting style.

Figure 2 Distribution of styles by mother’s education, averaged over the waves

Comments:: The transparent bars represent the discarded probabilities of the probability of participating in a warm rather than a cold parenting style, while the solid line is the core density. The sample is the pooled sample in which each parent appears three times.

Table 1 indicated that styles change over time. While there is some persistence – that is, warm style mothers are more likely to continue warm style and cold style mothers are more likely to continue cold style – there are also systematic shifts as the child ages. Parenting styles are getting colder on average. The other notable shift is that while there is no difference between parenting styles for boys and girls at 5 months, boys are statistically more likely to have a cold parenting style by the time the child reaches 29 months.

Do parenting styles influence skill development?

Although we cannot provide a definitive answer to this question due to a lack of exogenous variation, we can look at whether children exposed to certain parenting styles achieve higher skill levels later in life. More specifically, we look at summary measures of cognitive test scores (such as math or logic) and non-cognitive scores (such as conduct disorder or hyperactivity) at age 6.

In Figure 3 we show the impact of warm parenting on the standardized measures of ability at age 6. We show the relationship for parenting styles measured at each age individually (top) and for an aggregated style measure calculated across the three ages (bottom). We see that children exposed to a completely warm parenting style instead of an absolutely cold parenting style after 5 months achieve cognitive skills (left) that are 0.3 standard deviation greater and non-cognitive skills (right) that are more than 0.2 standard deviations. be higher . The effect sizes of parenting styles at 5 months of age are larger than at 17 and 29 months of age, even though parenting style is furthest from the outcome when the child is 6 years old. This lends some support to the idea that early childhood investments are of particular importance.

figure 3 Coefficients of Warm Parenting Regression on Cognitive and Noncognitive Skills at Age 6

Comments:: The dependent variable is calculated by taking the first factor of six measures of each, cognitive and non-cognitive ability, at age 6. The score is standardized with a mean of zero and a standard deviation of one. The thin lines represent the 90% confidence interval.

The aggregate measure of parenting styles shows an even higher correlation with outcomes. Warm parenting is associated with more than half a standard deviation of higher cognitive skills and a third standard deviation of higher non-cognitive skills.

Looking forward

Given the exponential growth of available data, we need to leverage machine learning to better understand what kind of parenting “works.” By letting the data speak, we can challenge traditional assumptions and discover patterns of success. This could help policy makers and researchers design interventions and support programs to help parents navigate the complexities of raising children and avoid ‘parental trap’ (Hilger 2022).


Attanasio, O, S Cattan and S Krutikova (2016), “Early Childhood Development Policy: The Evidence and Research Agenda”,, June 09.

Bandiera, O, A Prat, A Hansen, and R Sadun (2020), “CEO Behavior and Company Performance”, Political Economy Magazine 128(4): 1325-1369.

Blei, DM, AY Ng, and MI Jordan (2003), “Latent Dirichlet Assignment”, Journal of Machine Learning Research 3:993-1022.

Del Boca, D, C Flinn and M Wiswall (2014), “Household Choices and Child Development”, Assessment of economic studies 81(1): 137-185.

Draca, M and C Schwarz (2021), “How polarized are citizens? Measuring ideology from the ground up”, SSRN, May 11.

Doepke, M and F Zilibotti (2014), “Tiger Moms and Helicopter Parents: The Economy of Parenting Style”,, October 11.

Fiorini, M and MP Keane (2014), “How Children’s Time Allocation Affects Cognitive and Non-Cognitive Development”, Magazine for Labor Economics 32(4): 787-836.

Hilger, N. G. (2022), The parental trap: how to stop overburdening parents and solve our inequality crisis?MIT Press.

Kalb, G and J of Ours (2013), “Reading to children: a head start in life”,, June 10.

McCoby, E and J Martin (1983), “Socialization in the context of the family: parent-child interaction”, Child Psychology Handbook 4:1-101.

Rauh, C and L Renee (2022), “How to measure parenting styles?”, CEPR Discussion Paper 17326.

Leave a Comment

Your email address will not be published. Required fields are marked *