The adjustment of home-based parental literacy involvement to the level of reading literacy of pupils in primary school: a quantitative formalisation and empirical test

ABSTRACT

Previous research has shown that parental involvement at home seems to be socio-economically stratified. To study this stratification, recent research has drawn a distinction between early and late parental literacy activities at home. Early parental literacy activities at home happen before primary school, whereas late parental literacy activities at home occur during the time at primary school. In this article, we focus on the relationships between early and late home-based parental literacy involvement and reading literacy in Western Europe. The goal is to advance our understanding of these relationships by formulating and testing equations that explain the level of parental literacy involvement and reading literacy within a specific range of uncertainty. So far, no efforts have been made in the literature on parental literacy involvement and reading literacy, to formulate, create and test such equations.

KEYWORDS:

• Social class
• children
• parental involvement
• sociology

Acknowledgments

Thank you Nohemi Jocabeth Echeverria Vicente, Joe van ’t Hemelrijk, Rachel Cowler, Mary Graham and anonymous reviewers.

Disclosure statement

No potential conflict of interest was reported by the author.

Notes

1. Although the PIRLS 2006 dataset includes non-Western countries (e.g. Qatar), we decided to solely focus on Western European countries. Future research can include non-Western countries.

2. Reading literacy test scores of school tests are expected to fluctuate relatively more over time than the level of reading ability as measured by PIRLS because of – for example – different school test difficulties and topics of school tests. School tests do not (always) test reading literacy ability in a comprehensive way. However, we expect the reading literacy school test scores to be highly correlated with reading literacy as measured by the PIRLS reading literacy test, although school and the PIRLS reading literacy ability tests will both be affected by – for example – test anxiety and the level of concentration. We acknowledge the fact that the PIRLS reading scores give ‘plausible’ values.

3. Analyses were done with SPSS 21 and Excel.

4. We also checked the linear regression assumptions: we had normally distributed standardised residuals in all models, we checked for heteroscedasticity in the models in plots of the regression standardised residuals and the regression standardised predicted values. No standardised residual went beyond 3.04 or −3.13.

5. We also used another difference score, namely: the mean reading literacy of the transition groups – the general mean reading literacy instead of the mean reading literacy score of the SES group in which the transition group is situated. This general mean is 543.08. As expected, the explained variances in the same type of regression models dropped to respectively 53%, 54%, 75% R-squares for the early, late and reading literacy difference variables. The last variable is the only one that changed in this exercise. These results show that the mean reading literacy scores of the separate SES groups give higher explained variances.

6. This point is important to emphasise if a researcher wants to only use the data of a national sample. The sample size should be big enough to get enough detail.

7. In all models at the individual level, we had normally distributed standardised residuals in all models, we checked for heteroscedasticity in the models in plots of the regression standardised residuals and the regression standardised predicted values. The lowest standardised residual was −6.19 and the highest standardised residual was 6.09.

8. E.g. Sénéchal distinguishes “informal” and “formal” literacy involvement with differential relationships with literacy (2015, p. 398).

9. As Marx noted: “Je tiefer wir in der Geschichte zurückgehen, je mehr erscheint das Individuum, daher auch das producirende Individuum als unselbstständig, einem grössern Ganzen angehörig (…). Erst in dem 18t Jhh., in der ‘bürgerlichen Gesellschaft’ treten die verschiednen Formen des gesellschaftlichen Zusammenhangs dem Einzelnen als bloses Mittel für seine Privatzwecke entgegen, als äusserliche Nothwendigkeit. Aber die Epoche, die diesen Standpunkt erzeugt, den des vereinzelten Einzelnen, ist grade die der bisher entwickelsten gesellschaftlichen (allgemeinen von diesem Standpunkt aus) Verhältnisse. Der Mensch ist (…) ein Thier, das nur in der Gesellschaft sich vereinzeln kann” (Marx, 1976, p. 22).