From a recent Economics & Politics:
This paper explores the link between compulsory voting and income distribution using a cross-section of countries around the world. Our empirical cross-country analysis for 91 countries during the period 1960-2000 shows that when compulsory voting can be strongly enforced the distribution of income improves as measured by the Gini coefficient and the bottom income quintiles of the population. … Because poorer countries are the ones with relatively more unequal distribution of income it might make sense to promote such voting schemes in developing regions, such as Latin America. … We also use an instrumental variables approach to further check the robustness of our results. … Regardless of the control variables used, we obtain coefficients that are statistically significant in compulsory voting.
I suspect this is the main reason people push others to vote – because added voters tend to favor their side in wealth redistribution debates. To such pushers, the good of this outweighs added voter ignorance making other policies worse.
On voters leading to worse policy: The Myth of the Rational Voter by Bryan Caplan. His point is that common biases, not simple ignorance, lead to worse policies. He argues that if voters were simply ignorant, and errors were randomly distributed, then the miracle of aggregation would work and the well-informed would drive policy. Prof. Caplan's thesis is that errors are not random but skew predictably from what voters with the same values would prefer if they were better informed. Voters can indulge in feel-good "rational irrationality" because of the near-zero chance that any given vote will have a meaningful impact on policy.
"the good of this outweighs added voter ignorance making other policies worse"
Well, the study you quote seems to say that some good does come out of compulsory voting. That added voter ignorance, and therefore worse policies, also come out of it, makes sense to me, but is it true?
If voting is not compulsory, and noone is pushing people to vote, who votes? The best informed? The most partisan?
Are there any studies that support your "added voter ignorance" claim?