TREE Mixed Methods Sampling Framework

Description

Make mixed methods sampling decisions in a fixed order, from paradigm to purpose, sample size, relationship between samples and techniques, so they are transparent and defensible.

A meta-framework using a tree metaphor for sampling across strands. The roots are paradigmatic assumptions. The trunk covers the purposes of the study and of mixing, whom to study, the approach and number of phases, and sample size per strand: power analysis for quantitative strands, saturation logic for qualitative ones, tempered by heterogeneity, scope and feasibility. The limbs cross the relationship between samples (identical, parallel, nested, multilevel) with timing (concurrent, sequential) to give eight sampling designs, plus techniques chosen by the kind of generalisation sought (statistical, analytic, case-to-case). The leaves are the data. A flowchart walks the decisions in order and each choice is documented. The framework warns that commonly cited saturation figures, such as Guest et al. (2006), are context-bound rather than universal.

Attributes

Loss type
Siloed
Effort
Hours

Steps

  1. State paradigmatic assumptions for the study.
  2. Set the purposes of the study and of mixing methods.
  3. Define the population and frame, and whether sampling is needed.
  4. Choose the approach and number of phases.
  5. Size each strand: power analysis for quantitative strands, saturation reasoning for qualitative strands, adjusted for heterogeneity and feasibility.
  6. Choose the relationship between samples (identical, parallel, nested, multilevel) and timing (concurrent, sequential).
  7. Choose sampling techniques by the generalisation sought and document every decision.

Sources

  • The Sage Handbook of Mixed Methods Research Design (Poth, 2023)
  • Introduction to Mixed Methods in Impact Evaluation (Bamberger, 2012)

Also known as

  • Mixed methods sampling designs
  • Transparent, Rigorous, Ethical and Equitable sampling
  • TREE sampling meta-framework

Categories

  • Research Operations
  • Research Design & Rigor