I remember hearing about this after Apple did a stock split a little while back; apparently it completely tanked the Dow Jones due to the price per share being lower. I think they decided to add some new tech stocks or something to try to compensate for it, but finding out that they used the raw share price just made me stop paying attention to the Dow Jones ever.
It's a pretty dumb methodology, but it's kind of OK as a basic meter of how 30 big companies are doing and as a proxy for big companies in general. As long as you're mostly comparing between time periods and adjusting for splits and what not.
