← Projects
Completed Research

Remote Work Intensity Survey Instrument

View on GitHub ↗
RlavaanAMOSSPSSQualtricsBibTeX

About this project

Validated psychometric survey instrument developed for PhD research — measuring Remote Work Intensity, Liminal Spaces, Technostress, Work-Life Balance, and employee outcomes. Used in n=717 SEM study.

Background

The survey instrument is the foundation of the empirical work in my PhD thesis. The constructs I needed to measure — Remote Work Intensity, Liminal Spaces, Technostress (three dimensions), Work-Life Balance, and four outcome variables — don't all have off-the-shelf validated scales. Liminal Spaces in a remote work context is a construct I developed specifically for this research; the other constructs were adapted from existing scales with modifications appropriate to the remote work context.

Construct development follows a specific sequence: conceptual definition, item generation, expert review, pilot testing, and then CFA to assess psychometric properties. Each of the eight latent variables has four to six indicator items. The CFA results assess convergent validity (indicators load highly on their intended factor) and discriminant validity (factors are distinct from each other). Both need to be established before you can trust the SEM results that follow.

The n=717 sample was collected via Qualtrics with stratified recruitment to ensure representation across industry sectors and remote work intensity levels. Common method bias is a persistent concern in self-report survey research — I addressed it with Harman's single-factor test and a marker variable technique. The full Lavaan and AMOS code for the CFA and SEM models, along with the cleaned dataset structure, are documented in the repository.

Highlights

  • Construct operationalisation across eight latent variables
  • Confirmatory factor analysis (CFA) with convergent and discriminant validity testing
  • Structural equation modelling (SEM) with full mediation path analysis
  • n=717 Australian knowledge worker sample
  • Common method bias controls: Harman's single-factor test, marker variable technique
← All projects GitHub ↗
← Twitter Sentiment Analysis Leximancer Thematic Analysis Pipeline →