Data Collection Strategy

Source programme

Data come from the K-Hitech digital transformation training platform, operated jointly by the Korea Institute for Advancement of Technology and the Ministry of Trade, Industry and Energy. The programme targets manufacturing SMEs of 1–300 employees and provides both DT training and smart-factory infrastructure support.

Why this source

Three properties make it suitable, and one makes it problematic.

Suitable: it covers four consecutive years, giving temporal spread that a one-shot survey cannot; participation is administratively recorded rather than self-selected into a research study; and the population is exactly the one the research question concerns.

Problematic: the instrument was designed for programme administration, not for this study. The outcome item was not built to support the measurement claims the analysis places on it. This is the source of the study’s principal open problem.

Respondents

One to two senior respondents per firm — chief executive, chief information officer, or chief technology officer. Senior respondents give informed answers about firm-level infrastructure and strategy, at the cost of a single perspective per firm and no independent corroboration.

Structure

Repeated cross-section rather than a designed panel. Firms enter as they enrol; some return in later waves, most do not. The resulting structure supports pooled analysis directly and within-firm analysis only on the returning subset.

Planned augmentation

The pending revision requires an outcome measure that does not depend on the same respondent as the predictors. Two routes are under consideration: matching participants to administrative programme records — attendance, module completion, certification — or constructing a change measure from the infrastructure variable on the subset of firms observed in more than one wave. Whether the first route is available depends on institutional data access that has not yet been tested; the second requires no external permission but applies to a smaller sample.