Skip to main content

Table 5 Logistic regression to predict contact on first attempt, by migrants and non-migrants

From: Coverage of the migrant population in large-scale assessment surveys. Experiences from PIAAC in Germany

  Migrants Non-migrants
B SE B SE
Day: evening vs. before lunch .435* .227 .793** .083
Day: evening vs. after lunch .176 .144 .465** .050
Day of the week: Sat./Sun. vs. Mon./Tues. −.212 .193 .466** .071
Day of the week: Sat./Sun. vs. Wed. .439 .235 .496** .079
Day of the week: Sat./Sun vs. Thurs./Fri. .062 .205 .506** .074
Year: school holidays (reference = no) −.405** .138 −.023 .048
Year: religious holidays (reference = no) −.291 .664 −.465 .307
Gender (reference = male) .345** .133 .336** .047
Age .013* .005 .017** .002
  1. Migrant subpopulation: n = 1176. X 2 = 35 (df = 9); −2 log likelihood = 1.341; Cox and Snell R 2 = .03; Nagelkerke R 2 = .04. Non-migrant subpopulation: n = 8947. X 2 = 309 (df = 9); −2 log likelihood = 10.708; Cox and Snell R 2 = .03; Nagelkerke R 2 = .05. The Wald statistic has a Chi square distribution with 1 df. Reference category coded as 0
  2. * Significant at the .01 level
  3. ** Significant at the .001 level