Table 2 Correlations for educational attainment are consistently higher between dizygotic twins than between non-twin siblings suggesting twin-specific shared environments influence this trait.

From: Conventional twin studies overestimate the environmental differences between families relevant to educational attainment

Sibling studies

Twin studies

Paper

Country (Birth year)

Sample

rSIB

rDZ

Country (Birth year)

Sample

Paper

Gratz et al. (2021)8

Finland (1974–1980)

Registers (Statistics Finland)

0.36

0.52a

Finland (1974–1979)

Finn16

Silventoinen et al. (2020)13

0.60b

Finland (1936–1955)

Finnish Twin Cohort Study

Silventoinen et al. (2004)76

0.62b

Finland (1919–1957)

Finnish Twin Cohort Study

Silventoinen et al. (2000)77

Germany (1976–1989)

SOEP

0.51

0.60a

Germany (1990–1993)

TwinLife

Baier & Lang (2019)60

0.60a

Germany (1914–1969)

Bielefeld Longitudinal Study

Silventoinen et al. (2020)13

0.67a

Germany (1926–1987)

Berlin Twin Register

Norway (1970–1980)

Registers

0.41

0.51b

Norway (1967–1979)

Norwegian Twin Register

Ørstavik et al. (2014)78

0.46

Norway (1967–1979)

Norwegian Twin Register

Lyngstad et al. (2017)79

Sweden (1960–1982)

Multi-generation Registry

0.44

0.55

Sweden (1926–1958)

Swedish Twin Registry

Isacsson (1999)80

0.50c

Sweden (1950–1970)

Multi-generation Registry

Rietveld et al. (2013)37

UK (1954–1989)

UKHLS

0.42

0.47

UK (1951–1985)

TwinsUK

Branigan et al. (2013)12

USA (1954–1986)

PSID

0.51

0.56

USA (1976–1984)

AddHealth

Nielsen & Roos (2015)18

0.52a

USA (1908–1977)

California Twin Programme

Silventoinen et al. (2020)13

0.54a

USA (1910–1976)

Carolina African American

0.56a

USA (1979–1989)

Colorado Twin Registry

0.61a

USA (1894–1987)

Mid Atlantic Twin Registry

   

Twin and sibling studies

   

rSIB

rDZ

Country (Birth year)

Sample

Paper

   

0.39

0.49

Norway (1946-65)

Registers

Björklund & Salvanes (2011)61

   

0.45

0.50

Sweden (1950–1970)

Multi-generation Registry

Rietveld et al. (2013)37

   

0.45

0.56

USA (1976–1984)

AddHealth

Nielsen & Roos (2015)18

  1. rSIB represents the Pearson correlation for years of education between siblings and rDZ for the correlation between dizygotic twins.
  2. arDZ was algebraically derived from ACE estimates using the formula 0.5 A + C.
  3. brDZ was pooled from male and female samples (simple arithmetic mean and inverse variance weighted averages were the same).
  4. crDZ is male only (while rSIB is mixed).