Analytic methodology for childhood predictor analyses for wave 1 of the Global Flourishing Study

Table 4 Sensitivity of meta-analyzed childhood predictors to unmeasured confounding

Variable	Category	E-value for estimate	E-value for 95% CI
Relationship with mother	(Ref: very bad/somewhat bad)
Relationship with mother	Very good/somewhat good	1.22	1.09
Relationship with father	(Ref: very bad/somewhat bad)
Relationship with father	Very good/somewhat good	1.15	1.03
Parent marital status	(Ref: parents married)
	No, divorced	1.12	1.00
	Single, never married	1.11	1.00
	No, one or both had died	1.18	1.00
Subjective financial status of family growing up	(Ref: got by)
	Lived comfortably	1.20	1.13
	Found it difficult	1.16	1.09
	Found it very difficult	1.26	1.16
Abuse	(Ref: no)
Abuse	Yes	1.25	1.14
Outsider growing up	(Ref: no)
Outsider growing up	Yes	1.28	1.21
Self-rated health growing up	(Ref: good)
	Excellent	1.37	1.25
	Very good	1.26	1.18
	Fair	1.32	1.22
	Poor	1.36	1.21
Immigration status	(Ref: born in this country)
Immigration status	Born in another country	1.17	1.00
Age 12 religious service attendance	(Ref: never)
	At least 1/week	1.29	1.20
	1–3/month	1.29	1.13
	Less than 1/month	1.19	1.11
Year of birth	(Ref: 1998–2005; age 18–24)
	1993–1998; age 25–29	1.12	1.00
	1983–1993; age 30–39	1.21	1.00
	1973–1983; age 40–49	1.19	1.00
	1963–1973; age 50–59	1.25	1.00
	1953–1963; age 60–69	1.25	1.00
	1943–1953; age 70–79	1.28	1.00
	1943 or earlier; age 80 + ^a	1.34	1.00
Gender	(Ref: Male)
	Female	1.18	1.13
	Other^a	3.70	1.00

Note. N = 202,898; the E-value is the minimum strength of the association an unmeasured confounder must have with both the outcome (sense of mastery) and the predictor, above and beyond all measured covariates, for an unmeasured confounder to explain away an association [36]; and ^aGroup is very small (< 0.1% of the observed sample) within several countries potentially leading to complete separation and large uncertainty in this estimate—be cautious about interpreting this estimate

ISSN: 2731-913X