I've skimmed the entire study (possibly not available to you), published this month in Parliamentary Affairs, and I won't get into all the details. But the authors do a good job of tracing the history of this natural experiment, how they defined good citizenship, and most important -- how to measure their outcomes (are students who got the curriculum better students).
What I really like is among the outcome variables are measures of online participation as well as the traditional means of being a good citizen. That's neat. Table 5 really gets at the heart of the study, a series of multiple regression models predicting participation, knowledge, efficacy, and other key outcome variables. First off, they report the results oddly, leading with their variable of interest and then including all the control variables. But below lemme sum it all up:
- Participation is best predicted by education, parental occupation, but also by being in the good citizenship curriculum even after a boatload of other statistical controls.
- Knowledge is best predicted parental education and trust and also by the curriculum.
- Efficacy (the sense you can make a difference) is affected by a lot of variables listed above and again by the new curriculum.
An end to compulsory citizenship classes will not of course mean an end to citizenship education in Britain, since some schools will continue to deliver it as part of their mission to educate the next generation. But the large differences observed in Figure 4 between England, where citizenship education has been compulsory, and the other UK countries where it has not been is bound to raise questions about the long-term consequences of this change of policy on civil society in general.