*2.6. Analysis*

### 2.6.1. Quantitative Data

The data for the 85 observed participants were broken down into one-minute blocks which resulted in a total of 1532 min of observation time which was entered into SPSS version 23 [29]. Descriptive statistics were used to describe the parent/caregiver participant characteristics and the children under the parent/caregiver's supervision. The quantitative analysis explored the coinciding MDU, supervision, interaction and injury risk behaviours within each minute of the observation. A series of cross-tabulations with Pearson chi-square tests with a significance level of *p* < 0.01 were conducted to test the association between MDU, MDU Mode, MDU Duration and the outcome variables Supervision, Interaction and Injury Potential.

### 2.6.2. Qualitative Data Collection and Analysis

The interviews were transcribed verbatim by one researcher (KB) and were divided among the research team for open coding (KB, JL, JJ). The first researcher then collated the coded interviews using NVIVO 12 [30] and identified relationships between the codes to form the emerging themes which were agreed upon by the research team. Data saturation was reached during the data collection process (*n* = 20) and no new concepts emerged. Identified themes were consistent with work previously undertaken by Australian researchers [4]. The general inductive approach is a straightforward easily used, systematic set of procedures for analysing qualitative data and provides reliable findings [31]. Participant quotes to support the themes have been de-identified and presented in the results.
