A common source of confusion for new investors is whether they should try and build a portfolio themselves of individual stocks or, alternatively, invest in some sort of pooled structure such as a mutual fund, particularly a low-cost index fund.
Though the question itself is somewhat misguided - ultimately even an individual index fund is made up of individual stocks, it brings up an important point in that attempting to select portfolio components and portfolio component weightings without knowing what you are doing could be potentially disastrous to your family's net worth.
Still, for those who want to know whether they should invest directly in individual stocks, my general rule is that successful investing over long periods of time comes down to one sentence: Even a great company is a terrible investment if you pay too much for it. In actuality, you can substitute any asset for the word company; it doesn't matter if you are investing in bonds or investing in real estate. To drive this point home, the single biggest mistake I often see new investors make is not understanding what drives stock prices over decades. To account for this, I developed a five-second test that has served me well in estimating the relative level of knowledge a person has if they want to take the do-it-yourself approach to investment selection.
That test? I ask the following question: "If a company is earning $1 per share, there is no dilution expected, it is growing at 3% annually and expected to do so forever (merely keeping pace with inflation, which is also 3%), and the 30-year Treasury is yielding 5%, how much should you pay for the stock?"
If you can't answer that question in under five seconds, you probably have no business picking stocks for your own portfolio. Instead, you might find a broadly diversified portfolio of low-cost index funds more appropriate.
Why do I feel so strongly about this? Because over time, stock prices are driven by earnings per share and your returns are dependent upon how much you paid for each share of stock relative to those earnings. That is investing, in a nutshell. If you can't tell me how much is a "fair" price for a $1 of earnings at a specific growth rate, you can't hope to make money short of blind luck. You would be entering a game that was stacked against you and when you are dealing with retirement assets and the security of your family, that is not an intelligent course of action.
For those of you who are interested in the answer to the test, it depends on the rate of return you want. In that sense, it is a trick question. Someone experienced enough to know how to value a stock is going to know the price he should pay will depend upon the return they demand. If they wanted to earn 15% on their money - far in excess of the long-term returns earned on equities - they couldn't pay more than $8.58 for the stock. If they wanted to earn 12%, another highly attractive return that exceeds historical averages, they couldn't pay more than $11.44. If he wanted to earn 8%, he couldn't pay more than $20.60. If the stock is trading at, say, $30, the investor should know that he should only expect a rate of return of 6.43% in a world of stable risk-free rates and valuation multiples unless the growth rate turns out to be higher than projected. It adds clarity to your thought process. It imposes discipline on you.
The math involved in calculating intrinsic value goes far beyond the scope of Investing for Beginners. Suffice it to say, this test may serve you very, very well when you are determining how to build your portfolio even if all it does it let you know you would be better off building a widely diversified portfolio of mutual funds.