How to Measure Basic Personality Traits in Young Children

Despite their limitations, models of basic personality structure, such as the Big Five or the HEXACO model, have proven useful in summarizing and structuring the plethora of personality traits. While most research and applications using basic personality models have focused on adults, there is a growing interest in considering the basic personality dimensions of children. Herein, I will describe a recent study (led by Elisa Altgassen), in which we introduced the first measures to assess HEXACO personality levels in elementary school children (i.e., approx. 8-10 years old). (An overview of previously introduced measures to assess the personality of children and adolescents can be found here.)

The HEXACO Dimensions

The HEXACO Model of Personality structures and summarizes differences between individuals in six basic dimensions: Honesty-Humility, Emotionality, Extraversion, Agreeableness vs. Anger, Conscientiousness, and Openness to Experience. With regard to adults, research has shown that each dimension maps onto one class of criteria in particular (e.g., levels in Honesty-Humility are often associated with criteria in the realm of exploitation vs. cooperation). When measuring the HEXACO dimensions, the most widely used adult questionnaire also allows for the assessment of four facets per dimension. For instance, Extraversion comprises the facets of Liveliness, Sociability, Social Boldness, and Social Self-Esteem.

Developing the Children's Measures

While HEXACO questionnaires for adults have been applied to hundreds of thousands of participants, for various reasons, one cannot simply use the same questionnaires for children. For instance, some of the item content refers to situations that children do typically not encounter (e.g., behavior at work), and some of the item content is rather abstract for a child (e.g., on the appreciation of unconventional ideas). Following this, other researchers have already introduced the HEXACO-Middle School Inventory. This questionnaire can be administered as a self- or observer report form, and is targeted to assess the HEXACO levels of children around 10 to 14 years old.

Complementing this research, we aimed to provide a measure to assess the HEXACO levels in even younger children, approximately between 8 and 10 years old. To this end, we developed over 120 items, each theoretically corresponding to a HEXACO facet—and consequently, to one HEXACO dimension. We then obtained more than 2,000 parent ratings for these items. Parents – either the mother, the father, or both – assessed the HEXACO levels of their child, with the children being around 8-and-a-half years old on average. In addition, we again collected parent ratings for approx. 1,300 of the children about six months later.

Using the parents’ responses to the items, we employed so-called Ant Colony Optimization (getting Marvel vibes here?) meta-heuristics to identify the 'best' items for assessing the HEXACO dimensions. (Obviously, not all of our developed items were equally suitable.) Ant Colony Optimization is a good approach for selecting items or item sets because it considers multiple characteristics of a set of items simultaneously. For instance, we aimed to identify an item set that resembles the hierarchical HEXACO structure, with each HEXACO dimension comprising several facets. In turn, we selected a subset of items for each dimension such that each item represents a specific facet of the dimension, without capturing much content of the other facets. At the same time, we aimed to select a minimal number of items, because assessment time is typically limited in practice. We prompted the computer multiple times to make a suggestion for an item set based on these criteria, and looked at which items and/or item sets were suggested most often. (We also did a few other things to ensure getting the best results possible.)

The Final Measures

In the end, we felt confident to introduce two versions: First, a short form which includes 54 items and allows assessing the six HEXACO dimensions as well as three facets per dimension (remember that the most prominent HEXACO adult questionnaire comprises four facets per dimension). Second, an ultra-short form which includes 18 items and allows assessing the six HEXACO dimensions, but should not be used when interested in many facets. Examples of the finally chosen items are “My child is sad when they have to say goodbye to their parents or friends.” (Emotionality); “My child gets annoyed with other children when they just don't understand something.” (Agreeableness vs. Anger); or “My child wants to do things very thoroughly.” (Conscientiousness). We also found that the children who were rated as having rather higher (or average or lower) levels in a dimension where also more likely to be among those children who were rated as having rather higher (or average or lower) levels in the same dimension about six months later (i.e., relatively high test-retest reliability in statistical terms).

Relations with Other Variables

Beyond the fact that our measures seem suitable to assess the HEXACO dimensions for elementary-school children when looking at the structure of the measures themselves, we also investigated whether the dimensions were associated with relevant criteria in a theoretically plausible way. For instance, children’s levels in Honesty-Humility should be negatively related to their aggressive or antisocial behavior, or children’s levels in Openness to Experience should be positively related to their (controlling for their age) accumulated knowledge across domains of the culture they live in (so-called crystallized intelligence). While not every single relation turned out as expected, we did find, at large, meaningful relations between each of the HEXACO dimensions and (at minimum) one criterion.

Conclusion

It is important to emphasize that measuring children’s personality traits is a different ball game than measuring those of adults. For instance, the structure of children’s personality might be different (e.g., there might not be the same six dimensions of basic personality that have been observed in many studies with adults); children might show less stable patterns in their affect, cognition, and behavior across situations (e.g., there might not be fully fledged traits); or the assessment might come with certain challenges (e.g., parents might lack opportunities to compare their children with other children across situations, and asking children to fill out self-reports might also come with some limitations). So, one should carefully consider in which circumstances one wants to assess children’s personality traits—and, if so, how.

But if one is interested in measuring children’s basic personality traits — e.g., because this might provide insights into selecting or creating situational features that support a child in their development or because one wants to follow the personality development of children over a long period of time — we now introduced two ways to assess (about) 8-10-year-old children’s HEXACO personality levels: A short and an ultra-short form, which, we hope, supports researchers and practitioners being interested in younger children’s HEXACO levels.