作者: Jeffrey Moore , Nancy Bates , Joanne Pascale , Aniekan Okon
DOI: 10.1002/9780470743874.CH5
关键词:
摘要: This chapter examines the impact of dependent interviewing procedures on “seam bias,” a phenomenon peculiar to longitudinal panel surveys. Seam bias refers tendency for estimates change measured across “seam” between two successive survey administrations far exceed within single interview – often by factor 10 or more. The presence seam almost always signals measurement error. Much research over past decades has documented existence in surveys, and also shed light its essential nature too little reference period interview, much at seam. Attempts control have met with some success, but been limited primarily employment-related characteristics.