It's logically accepted (albeit with some challenges to the conventional wisdom) that smaller class sizes are better for students because teachers can devote more time to each student. Similarly a smaller investment portfolio means that the investor can devote more time to monitoring each investment. There are tradeoffs though. Smaller classrooms mean a higher cost of education per student and smaller portfolios come at the expense of diversification.
Unfortunately there's no scientific way to determine the optimal level that balances these competing forces in classrooms or investments (sorry efficient markets theorists). However, qualitatively, the best size would be the largest size possible without sacrificing the individual attention needed to monitor the progress of each student/investment.
Looking at international data, the prevailing balance in the classroom seems to be in the 20 student range. In the average teacher's portfolio, each student is a 5% position.