One of the major problems in estimating the mode measurement effect in (mixed-mode) surveys is that isolation of the causal effect of mode on measurement is difficult due to the fact that selection and measurement effects are (potentially) correlated. In this report I use experimental data collected in round 11 of the European Social Survey (ESS) to study the size of mode effects in Great Britain and Finland. Respondents were randomized into a condition in which recruitment and interviewing was done using face-to-face interviewing, or a condition in which a push-to-web design was used. Here respondents were invited by postal mail, and aksed to do the survey online. Paper was used to convert nonrespondents or for respondents requesting to do the survey on paper. The report analyses mode effects, and aims to control for selection effects to assess the size of mode-effects across 111 numeric variables in the European Social Survey.