Ticket 03 · Survey design
Katz and Krueger: survey design when alternative work arrangements reshape measured employment
Lawrence F. Katz and Alan B. Krueger’s research on alternative work arrangements examined how survey design captures independent contracting, temporary help, on-call work and related forms—and how those forms appear (or fail to appear) beside traditional employment measures. When secondary or nontraditional earnings are discussed in editorial settings, the Katz–Krueger design debate is a useful hang point: the object measured depends on the ticket’s wording.
What the survey design problem looks like
Standard household surveys often organize questions around a main job. Alternative arrangements may be asked in a supplement, in a special module, or not at all. Katz and Krueger’s work—alongside BLS Contingent Worker Supplement documentation—shows that small changes in how temporary status or contractual form is asked can move estimated shares. Later revisions and critiques have also argued that some early readings of rising alternative arrangements were sensitive to sample and question design. Contested series belong on the same page as the critique, not in a separate footnote that readers never reach.
Nontraditional earnings further complicate the rail. A person may hold a wage job and a platform task stream; instruments that only hang the main job leave the secondary stream off the board. Instruments that ask about “any paid work” without distinguishing arrangement type blur categories that labour economists later try to separate. Design is therefore not cosmetic: it decides which flows can be composed into household income accounts.
Composition beside administrative traces
Administrative records (for example tax information returns) sometimes show nontraditional payments more clearly than interviews. Research associated with Katharine Abraham and colleagues on survey–administrative gaps illustrates why interview-based alternative-arrangement totals and tax-based nontraditional earnings can diverge. Cliprail records that divergence as a measurement fact. It does not invent a preferred national rate or prescribe how households should report income.
What this note does not claim
This ticket summarizes survey-design lessons from published academic and statistical work. It does not estimate alternative-arrangement prevalence for any reader, prescribe contracting choices or forecast earnings. Cliprail does not publish testimonials or success narratives. Named authors and agencies remain the sources of record for any figure quoted elsewhere.