Measuring Parent–Adolescent Communication and Interaction Using Video: A Pilot Study of the ‘TEENCAM’ Video Observational Methodology in Soweto, South Africa

Abstract

Background: Adolescents in South Africa deal with many challenges pertaining to their development and mental health. Parent-adolescent communication is essential for promoting and fostering strong parent-child relationships, which in turn promote increased supervision and monitoring of adolescent behaviour, which is known to improve adolescent outcomes. Generally, parent-adolescent communication is measured using subjective self-report questionnaires, which are known to have bias and tend to focus more on broader aspects of parenting and adolescent behaviours or attitudes, and less on communication and quality in parent-adolescent relationships. Similarly, qualitative interviews tend to have an individual rather than a didactic focus, relying on audio recording transcripts which provide limited insight into relational interactions and communication. Observational video methodology has for a long time been used in early childhood research to measure parent-child relationships objectively, and video feedback methods are often used as an intervention tool in parenting programs. However, video observation has seldom been applied to parent-adolescent interactions, and potentially offers a new and objective method to measuring communication and interaction between parents and adolescents. Little is known about the acceptability or feasibility of this methodology for this age group or its potential for use in a South African urbanised context, or its relative advantages to other current observational methods. Aim: The aim of this qualitative study was to explore the acceptability and feasibility of using 'Teencam' to observe parent-adolescent communication and interaction in a South African urban setting and, to explore its potential acceptability as a video feedback intervention tool. Methods: The study enrolled 16 parent-adolescent pairs in Soweto, South Africa (SA). The adolescents (60% females), aged 11 to 15 years, were paired with their biological parent, who identified as the adolescent’s current primary caregiver, and the pair needed to be co-resident to be eligible for participation. Communication and interactions between the parents and adolescents were assessed over several time- points, using a novel video observational portable headcam device called ‘Teencam’. Video data were coded using manual (audio and video coding) and automated coding (FaceReader software – capable of detecting basic facial expressions). Feasibility was evaluated by the reliability of the Teencam device to record v interactions that were usable or codable in terms of audio and video content, and by evaluating the Teencams relative performance and added value against other methods of self-report (audio transcripts only) and analysis (automated software). To determine acceptability, 14 parents participated in one-on-one interviews about their perceptions on their adolescent’s behaviour during observation, the comfortability of wearing the Teencam device, and the feasibility of using the observational game tasks during their interactions with their adolescents. Results: The Teencam methodology proved to be feasible and acceptable amongst parent-adolescent pairs in Soweto, South Africa (SA). The Teencam device was able to capture verbal and non-verbal communication and interactions amongst parent-adolescent pairs enhancing the quality of data available for interpretation, and the inferences that could be made, when compared to audio transcript only. Limitations in the Teencam device's capacity to handle participant movement and some operational constraints, such as turning on and off and beginning recording were noted. Furthermore, the quality of lighting in the assessment rooms and the colourful decoration of the headcam device impeded the performance of the FaceReader software in analysing facial expressions. Parents reported the method to be acceptable and engaging, with specific preferences for card game interactions and endorsed that the approach could be helpful in interventions that aimed to improve the parent-adolescent relationship and communication. Conclusion: This study demonstrates that with some technological enhancements, the feasibility and acceptability of measuring parent-adolescent communication and interaction in research using Teencam video observational methods in urban South Africa is promising. The highest quality of analytic output stemmed from manual coding and triangulation of data, which while appropriate for qualitative investigations or pilot studies, would be unfeasible to replicate at a larger scale without significant human and time resources. Advances in technology are needed to use this method on a larger scale.