Poster #73 - FLO: A Novel Video-Coding Tool to Evaluate Responsive Caregiver-Child Interactions

• In Event: Poster Session 10

Abstract

Although there is substantial evidence that parenting behaviors are closely linked to child outcomes (Ferguson, 2013; Pinquart, 2016), current standardized protocols for evaluating parent-child interactions struggle to effectively capture responsive caregiving behaviors among high-risk samples, where the amplitude of serve and return behaviors is lower (Lowe et al., 2017). Global rating systems fall short of capturing discrete behaviors that are sensitive to change, while existing microsocial coding schemes focus on specific granular behaviors and are often developed by research teams for use within single published studies rather than broader application (Wan et al., 2017). Publication bias makes it difficult to document examples of failed attempts to capture behavioral changes across intervention periods, but there is a notable absence of research that evaluates parent-child interaction in the context of parenting interventions with high-adversity samples (Peisch et al., 2018).

The FLO Coding Tool is a novel video-coding measure designed to address the limitations of other observational measures by straddling microsocial and global coding methods. This approach, which mimics the natural flow of dyadic interactions, allows researchers to capture the detail necessary to quantify caregiver behaviors (i.e., following, leading, other) in a way that is sensitive to change across an intervention period. The coding protocol utilizes a decision flowchart and accompanying glossary to categorize behaviors; resulting analyses offer a simple caregiver profile based on the percentage of following/leading/other behaviors during a given interaction. This approach is designed to maximize coding efficiency and feasibility across contexts while still capturing a level of insight into the nature of serve and return behaviors within interactions.

This poster will summarize the FLO Coding Tool’s feasibility, reliability, validity, and sensitivity to change using analyses from three datasets that assess the scope and breadth of this measure’s potential:
a) Pilot 1 includes 233 free-play videos, filmed in families’ homes. 91 pre and post intervention sessions will be included in analyses. Children’s ages range from 6 - 36 months; 68.1% (n=62) of these families recorded interactions in Spanish.
b) Pilot 2 includes 380 free-play videos filmed in a research lab. 132 pre and post intervention sessions will be included in analyses. Children all had a suspected or diagnosed developmental disability and ranged in age from 2.5 - 6 years.
c) Pilot 3 includes 45 free-play videos filmed in a research lab. This pilot utilizes a cross-sectional study and does not include pre-post intervention outcomes. Children were all 6 months old; 31% (n=14) of this sample are caregivers in treatment for opioid abuse.

These three pilot trials aim to evaluate the psychometric properties of the FLO Coding Tool to validate its application across varying contexts and ages. We will present an assessment of the core psychometric properties of this observational coding tool: a) inter and intra-rater reliability, b) stability over time, and c) convergent and discriminant validity. Analyses will emphasize understanding the internal properties of the tool and evaluating how caregiver following/leading/other behaviors are correlated with other relevant parenting outcomes (i.e., parenting stress, sense of competence, executive function) and child outcomes.

Authors

• Andrea Imhof, University of Oregon
  Presenting Author

• Shannon Peake, University of Oregon

• Philip Fisher, Stanford University