Parental beliefs, perceived health risks, and time investment in children: evidence

The estimates in the first step are used in the second step to obtain the average mean and variance of the coefficients. [...] We first add race of the respondent, sex and age of the child, and formal childcare quality (Appendix Table B9), and show that the results are qualitatively the same, with the exception of the perceived returns to play with friends, for which we uncover a more pronounced gradient by socio-economic status.36 We also studied whether the quality of the early years settings available to the respondent. [...] We further examined whether perceived returns and risks are associated with the number of COVID-19 cases (measured in the two weeks before the survey) in the Local Authority where the respondent resides, finding that this is generally not the case (Appendix Table B10).38 On the other hand, we find that the relationship between the number of COVID-19 deaths (measured in the two weeks before the sur. [...] The ratings are at the provider level; for the anal- ysis, we aggregate them and use the average rating in the Local Authority where the respondent resided, since we did not ask the name of the early years setting attended. [...] Interestingly, the results in Table 6 also show that the gradients in the time the mother and the father spend in active play with the child go in opposite directions: respondents with higher levels of education report that, in their household, the mother spends less time actively playing with the child (column 3), while the father spends more time in this activity (column 4), as compared to those.