Panel 03 · Survey design
Katz and Krueger: survey design when alternative arrangements reshape measured employment
Lawrence Katz and Alan Krueger’s research on alternative work arrangements is frequently cited when analysts ask whether nontraditional jobs grew relative to standard employment. For a method desk, the lasting value is less a single headline share and more the survey-design problems their work surfaces: who is counted, under which label, and with which comparison baseline.
Why questionnaire wording is the trellis
Estimates of independent contracting, temporary help or on-call work depend on whether respondents recognize those labels and whether the survey asks about the main job, all jobs or any paid activity in a reference window. Small wording shifts can move people across categories without any change in their week-to-week tasks.
Katz and Krueger’s analyses of Contingent Worker Supplement waves and related survey experiments illustrate how measured prevalence can differ when instruments change even if the underlying labor market is stable. That is a lattice constraint: the openings of the frame determine what can be hung on it.
Secondary earnings and uneven capture
Nontraditional earnings often appear unevenly across instruments. A person with a wage job and occasional platform tasks may look conventional in a main-job module and unconventional in a side-activity module. Researchers composing household income accounts therefore need to know which module—if any—captured the secondary flow.
Administrative sources can contradict household reports for some forms of nontraditional pay. Contested gaps between survey and tax data are documented in the broader measurement literature; they are not resolved by a single paper’s point estimate.
Reading claims carefully
Claims that “alternative work rose” or “fell” should be checked against the exact survey wave, weighting and definition used by the authors. Contested findings belong on the same page as the claim, not in a footnote the reader never sees.
What this note does not claim
This panel describes survey-design lessons associated with Katz and Krueger’s line of work. It does not estimate any reader’s employment form, prescribe contracting choices or forecast earnings. Trellis Pad does not publish testimonials.