Help with setting up a statistics hypothesis test please?

[Mrs. Cullen]

The stats question is asking if investors attitudes toward stocks differ from their attitude towards bonds. Can anyone help with how to set up the null and alternative hypothesis? I know they should be written out, i.e. Ho: The attitudes differ; Ha: The attitudes do not differ or Ho:The attitudes are the same; Ha: The attitudes are not the same, but I'm not sure which of these sets is correct and that makes a huge difference in the outcome of the problem. If anyone can help me out, I'll choose it as Best Answer if you can support why your decision is right. Thanks!!

 

[Steppenwolf]

When you do hypothesis testing, you are trying to figure out the likelihood that two sets of observations are coming from the same probability distribution. First figure out what measurable variable represents the attitude of the investor. An example is the length of the holding period for an equity purchase. Your null hypothesis could then be that the mean length of the holding period is the same whether the equity is a bond or a stock. You can then try to see if you can reject the null hypothesis using statistical evidence. Hope that helps. "Attitude" is too vague a word to be useful in statistical testing.