An estimate of the error in self-reported college major

ABSTRACT

This letter provides an estimate of the extent of error in self-reports of college major. A student’s field of study is an important determinant of labour market outcomes and of increasing interest to labour economists, but little is known about the reliability of survey data on college major. A unique dataset from the United States with both transcript and survey data on major field of study suggests that the error rate of self-reported college major is almost 20%. Error rates are higher for relatively small or obscure majors, and are lower for larger majors or majors closely associated with licensed professions (e.g. health care). Although these error rates are not trivial, they are comparable to prior estimates of error in reporting educational attainment.

KEYWORDS:

• College major
• measurement error
• survey data

JEL CLASSIFICATION:

• I210
• C81
• J240

Disclosure statement

No potential conflict of interest was reported by the author.

Notes

¹ For discussion of assessments of measurement error that do not rely on the a priori assumption that one data sources is authoritative, see Abowd and Stinson (2013).