Poster #85 - The Transitional Relationships Between Early Childhood Learning Behaviors, Preschool Fade Out, and Shifting Reference Standards

• In Event: 1-169 - Poster Session 04
  In Poster Session: PS 04 Section - Education, Schooling

Integrative Statement

Head Start and many other early education programs have found it beneficial to fortify the process of learning basic cognitive and social-emotional skills by promoting children’s learning behaviors (Matthews et al., 2010; McDermott et al., 2009). Learning behaviors explain how children learn and encompass behavioral manifestations of competence motivation, endurance in learning, and strategic planning (McDermott et al., 2011). As such, their improvement can lead to improvements in the skills that emerge from them (Hyson, 2008; Kagan et al., 1995).

The early evidence for the growth trajectories of learning behaviors revealed almost as many distinctive patterns of transition as there were different cognitive and social-emotional outcomes (McDermott et al., 2014, 2016). These results are tied to individual growth modeling, and while informative, the method does not identify the growth trajectories that distinguish natural subpopulations (classes) of children.

In this study, we employ growth mixture modeling (GMM; Ram & Grimm, 2009) to identify the characteristic trajectories for learning behaviors. These discoveries would reveal the relative sizes of the resultant classes and enable researchers to relate them to important precursors and outcomes.

A large sample (N = 2,152) of children was assessed twice annually through Head Start, kindergarten and 1st-grade for manifestations of Competence Motivation and Attentional Persistence. Competence Motivation measures phenomena such as interest in learning activities while Attentional Persistence includes frustration tolerance.

GMM revealed that a two-class model fit best for Competence Motivation and a three-class model for Attentional Persistence. Each form of learning behavior featured dominant subpopulations with quite good learning behaviors during Head Start that deteriorated upon kindergarten entry (comprising 63.2% of children for Competence Motivation, 45.0% for Attentional Persistence). Other subpopulations showed children arriving in Head Start with poor or average learning behaviors. See Figures 1 and 2. Membership in a less desirable subclass is associated with preexisting variables (e.g., male) and negative outcomes (e.g., mathematics nonproficiency in 2nd-grade).

Notably, most children’s learning behaviors deteriorated precipitously upon kindergarten entry. In the context of preschool performance fade out, research suggests this may be associated with (a) students transitioning into qualitatively inferior school settings (Currie & Thomas, 2000; Lee & Loeb, 1995); and (b) that the skills taught by teachers may be more associated with teaching-to-the-test (te Nijenhuis et al., 2014).

Teachers’ shifting reference standards may also play a role in this decline. Because Head Start children tend to be struggling learners (Kopack Klein et al., 2013), teachers may calibrate their perceptions such that child performance, perceived as mediocre in a more heterogeneous population (such as kindergarten), will not be assessed that way. The literature supports this, showing teacher evaluations will shift with population shifts and classroom changes (Hamre et al., 2008; Mashburn & Henry, 2004).

Efforts to reduce the readiness gaps affecting at-risk early education populations (e.g., Lipsey et al., 2013) have produced mixed results. Whereas myriad factors may play central roles in the dilemma, we suggest there would be reasonable justification to view faulty learning behaviors as prime suspects leading to more pervasive difficulty in schooling.

Authors

• Paul A. McDermott, University of Pennsylvania
  Non-Presenting Author

• Michael J Rovine, University of Pennsylvania
  Non-Presenting Author

• Roland S Reyes, University of Pennsylvania
  Presenting Author

• Jessica L Chao, University of Pennsylvania
  Non-Presenting Author

• Richard Scruggs, University of Pennsylvania
  Non-Presenting Author

• Katharine W Buek, University of Pennsylvania
  Non-Presenting Author

• John W Fantuzzo, University of Pennsylvania
  Non-Presenting Author