Using machine learning to study parenting styles

How should we raise our children? Research has shown that the amount of parental time invested is not the only crucial element for children’s skills development (Del Boca et al. 2014; Attanasio et al. 2016); parenting style also matters (Fiorini and Keane 2014). Parenting style is a strategic choice linked to incentives (Doepke and Zilibotti 2014). In order 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 key actions. For example, reading for children has been found to be highly predictive of children’s skills development (Kalb and Jan van Ours 2013). But how can we measure parenting styles without relying on past beliefs?

In a recent paper (Rauh and Renée 2022) we follow an approach that makes the data speak. We adopt a model of computer linguistics to determine what kind of parenting styles exist and are the most striking. The original goal of the machine learning algorithm is to learn from the simultaneous occurrence of words and form topics around them. In the context of parenting styles, the idea is that parents involved in one activity are more likely to engage in another given 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 mothers’ behaviors collected. performed over three waves when the target children were 5, 17 and 29 months old.

One advantage of the dataset is that mothers’ behavior toward their children is not captured by self-reported survey questions, but by the adder. At the end of each interview, the adder records ten variables that classify whether the mother is involved in certain actions during the interview. This makes the data less susceptible to a prejudice that may arise from mothers who misrepresent how they behave towards their children.

In Table 1 we see the list of ten activities and what percentage of mothers get involved 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 embracing the child rather than yelling at the child or expressing annoyance. Second, despite this imbalance occurring in all three admissions waves, there is a significant shift toward more punitive and less supportive action as the child gets older.

Table 1 Parental action on recording waves

Notes: The table describes the behavior of the respondents and their interactions with their children during the annual QLSCD interview in terms of percentages. Behavior is evaluated by the breeder 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 the classification of text into topics. In recent work, economists have used LDA to determine striking CEO behavior (Bandiera et al. 2020) and political ideologies (Draca and Schwarz 2021). In our case, each parenting style is defined by the probability of each parenting action, and parents are classified into two styles. The classification is not ‘strict’ in the sense that parents can be a mixture of the two types: the algorithm assigned shares of each style to parents in the dataset.

In Figure 1 we show the consequent distribution of parental actions when parenting is classified into two styles, A and B. The red bars indicate the occurrence of a given action under parenting style A and the blue bars under parenting style B. The left panel shows the probability of each of the ten actions within a topic. A parenting style should be represented 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 out of the red bars. She will most likely pull the largest rods; she may, for example, give an educational toy to her child or respond to her baby’s sounds. A parenting style B mom will most likely not take any action when you draw out of the blue bars in the urn because they are so small. In other words, parenting style B is characterized by lack of action. Based on these divisions of actions, our developmental psychologists follow McCoby and Martin (1983) and label parenting style A as ‘hot’ and parenting style B as ‘cold’. In the right panel of Figure 1, we see the standardized relative occurrence of a given action within a parenting style. We see that hot mothers are relatively unlikely to reprimand or scream at their child. Cold parents, if anything, simply look after their child.

Figure 1 Distribution of actions according to parenting style

Notes: The left panel 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 inclined to follow a warm parenting style? To answer this question, we look at the correlation between warm parenting and parenting and family characteristics. In Figure 2, we see the spread of parenting styles through maternal education. In the left panel, we see that among less educated mothers with at most a high school education, the cold parenting style is most likely. This is indicated by the mass to the left of the graph. If we look at the panel on the right, we see the opposite for mothers with a college degree. 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 adopt a cold parenting style.

Figure 2 Distribution of styles through maternal education was average across waves

Notes: The transparent stripes represent the built-in probabilities of the probability of engaging in a warm rather than a cold parenting style, while the solid line is the pit density. The sample is the pooled sample in which each parent occurs three times.

Table 1 indicated that styles change over time. Although there is a degree of perseverance – that is, mothers with a warm style are more likely to continue to follow a warm style and those with a cold style are more likely to continue a cold style sit – there are also systematic shifts as the child gets older. On average, parenting styles get colder. The other remarkable shift is that, while at 5 months there is no difference between parenting styles versus boys and girls, by the time the child reaches 29 months, boys are statistically more likely to face a cold parenting style.

Do parenting styles influence skills development?

Although we can not give a definitive answer to this question due to a lack of exogenous variation, we can see if children who are exposed to certain parenting styles reach higher skill levels at later ages. More specifically, we look at summary measures of cognitive test scores (such as math or logic) and non-cognitive scores (such as behavioral disorders or hyperactivity) at age 6.

In Figure 3, we show the impact of warm parenting on the standardized skill measures at age 6. We show the relationship for parenting styles measured at each age separately (above) and for a total measure of style calculated across the three ages (below). We see that children who are exposed to a fully warm parenting style at 5 months rather than an absolutely cold parenting style achieve cognitive skills (left) that are greater than 0.3 standard deviations and non-cognitive skills (right) that are greater than 0 , 2 standard deviations are higher. The effect sizes of parenting styles at the age of 5 months are larger than at ages 17 and 29 months even though the parenting style is measured the furthest away from the outcome when the child is 6 years old. This provides some support for the idea that very early childhood investments are of particular importance.

Figure 3 Warm parenting regression coefficients on cognitive and non-cognitive skills at age 6 years

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

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

Look ahead

Given the exponential growth of available data, we need to utilize machine learning to gain a better understanding of what kind of parenting ‘works’. By making the data speak, we enable us to challenge traditional assumptions and uncover patterns of success. It can help policymakers and researchers design interventions and support programs to help parents navigate the complexities of raising children and avoid the ‘parental trap’ (Hilger 2022).


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

Bandiera, O, A Prat, A Hansen and R Sadun (2020), “CEO behavior and firm performance”, Journal of Political Economy 128 (4): 1325–1369.

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

Del Boca, D, C Flinn and M Wiswall (2014), “Household Choices and Child Development”, Overview of Economic Studies 81 (1): 137–185.

Draca, M and C Schwarz (2021), “How polarized are citizens? Measuring Ideology from the Ground-Up ”, SSRN, 11 May.

Doepke, M and F Zilibotti (2014), “Tiger moms and helicopter parents: The economy of parenting style”,, 11 October.

Fiorini, M and MP Keane (2014), “How the allocation of children’s time affects cognitive and non-cognitive development”, Journal of Labor Economics 32 (4): 787–836.

Hilger, NG (2022), The parental trap: how to stop overloading parents and correct our inequality crisisMIT Press.

Kalb, G and J van Ours (2013), “Reading for children: an edge in life”,, 10 June.

McCoby, E and J Martin (1983), “Socialization in the Context of the Family: Parent-Child Interaction”, Handbook of Child Psychology 4: 1–101.

Rauh, C and L Renee (2022), “How to Measure Parenting Styles?”, CEPR Discussion Paper 17326.

Related Posts