Does value importance impact value-behaviour relations in children?

• In Event: Values and Behaviors: A Developmental Perspective on the Origins of Character

Abstract

Children’s values systematically relate to prosocial and aggressive behaviour (e.g., Benish-Weisman, 2015, Benish-Weisman, Daniel, Sneddon & Lee, 2019). Specifically, prosocial behaviours are positively associated with values emphasizing others’ welfare (self-transcendence), whereas aggressive behaviours are positively associated with values promoting self-interests (self-enhancement). We propose that the role of values becomes especially meaningful when values are central to one's value system; specifically, when they are either very important or especially low in importance.
We test this with a novel technique, estimating quantile correlations between values and prosocial and aggressive behaviours. This technique is an extension of median and quantile regression that uses the entire data set to estimate correlations at different points along a variable’s distribution by weighing positive and negative residuals differently (Benoit & van den Poel, 2009).
723 Australian primary school children (6-12 years) participated. We measured values using the revised Animated Values Instrument (Lee et al., 2017) which, based on best-worst scaling theory, asks children to choose which of the presented animated values scenarios they “most want to be like” and which they “least want to be like”. Scores for each value-item (seen 5 times and paired with every other item once across sets) were calculated using the square-root ratio procedure. We measured prosocial and aggressive behaviour using peer (classmate) nominations (McDonald, Benish-Weisman, O’Brien, & Ungvary, 2015). Specifically, children were given a list of classmates names and asked to circle the names of all classmates who fit the criterion for prosocial behaviour (“cooperative”, “helpful”, “kind”) and for aggressive behaviour (“hits”, “is mean”, “fights”).
As predicted, associations between children’s values and peer nominations of behaviour were stronger at higher levels of value importance. At higher versus lower levels of value importance, self-enhancement values were more strongly positively associated with aggressive behaviour (Fig. 1a) and self-transcendence values were more strongly negatively associated with aggressive behaviour (Fig. 1b). Further, both Pearson and Spearman correlations between self-enhancement and aggressive behaviour (r = .27; rs = .22) and self-transcendence and aggressive behaviour (r = -.15, rs = -.07) were much weaker than quantile correlations at the .8 (ρ_τ = .49, -.23) and .9 (ρ_τ = .63, -.36) quantiles of value importance. However, the results were not quite as clear for prosocial behaviour. While self-enhancement values were more strongly negatively associated with prosocial behaviour (see Fig.1c) at higher levels of value importance, this was not the case for self-transcendence (Fig. 1d). Further, for self-enhancement and prosocial values Pearson and Spearman correlations were much lower (r = -.28, rs = -.26) than quantile correlations at the .8 (ρ_τ = -.35) and .9 (ρ_τ = -.41) quantiles of value importance.
These results show, for the first time, that the level of value importance influences value-behaviour relations in middle childhood. Important values are more strongly related to value-expressive behaviour than previously thought. However, in the case of the more normative self-transcendence values and prosocial behaviours, other factors may be in play. Further research is needed to examine these factors and to investigate which developmental factors impact these findings.

Authors

• Julie Lee, University of Western Australia
  Presenting Author

• Joanne Sneddon, University of Western Australia
  Non-Presenting Author

• Trish Collins, Edith Cowan University
  Non-Presenting Author