Table 2.

Results from falsification tests

		Tests for differences between cohorts
Variable description	Name	p	FDR-adjusted p
Trying to start a business	startbiz	0.50	0.75
Number of full-time jobs since 2005	work3yrs	0.67	0.78
Number of years worked for current employer	curempyr	0.54	0.75
Amount of pay change since started job	paychnge	0.40	0.75
Was pay higher/lower/the same in previous job?	pastpay	0.28	0.68
Why did the respondent leave their previous job?	whyleave	0.35	0.75
Does more trade lead to fewer jobs in the U.S.?	moretrde	0.27	0.68
Computer use at work	wkcomptr	0.12	0.66
Can job be done without a computer?	wocomptr	0.82	0.82
Have any co-workers been replaced by computers?	autonojb	0.02	0.22
Frequency of meetings with customers, clients, or patients	meetf2f1	0.15	0.66
Frequency of meetings with co-workers	meetf2f2	0.28	0.68
Frequency of communication with co-workers outside the U.S.	intlcowk	0.22	0.68
Does the respondent receive health insurance from their employer?	emphlth	0.82	0.82
Is there another name for the respondent's insurance or HMO policy?	othplan	0.58	0.75
Gender of sex partners	sexsex18	0.09	0.66
Ever been the target of sexual advances by a co-worker/supervisor?	harsexjb	0.55	0.75
Has respondent been the target of a sexual advance by a religious leader?	harsexcl	0.66	0.78
Do they know others who have been the target of sexual advances?	knwclsex	0.47	0.75

Note: These items were not asked of the 2006 cohort in 2006, but were asked of both cohorts in 2008. Variable names are given in the “name” column. The FDR-adjusted p is the p -value adjusted for the False Discovery Rate. Adjustments were made using the procedures of Benjamini and Hochberg (1995). See text for more details.