Corporate Bonds Data
July 20th, 2010 | 
Author: 
Question: Statistics Help!?
What type of statistical test should I use for the following data:
7% of mutual fund investors rate corporate stocks "very safe, 58% rate them "somewhat safe", 24% not very safe, 4% not at all safe, and 7% not sure.
A poll asked 529 mutual fund investors how they would rate Corporate Bonds on safety. Their answers are as follow:
48 very safe, 323 somewhat safe, 79 not very safe, 16 not at all safe, and 63 not sure.
I converted the frequencies of the percentage of corporate stocks by the 529 people interviwed on Corporate Bonds to get my data and now I don't know if I should do a chi sqare test for the difference between two proportions for my hypothesis testing.
The end question in the problem is:
Do mutual fund investors attitudes toward Corporate Bonds differ from their attitudes toward corporate stocks? Support answer with statistical test using alpha = 0.01
Answer: If you have converted the stock frequency data back to the actual numbers then you have the counts for the different categories and a chi-squared test would be appropriate but you need to think about what to do with the "not sures"
"not at all", "not very", "somewhat", "Very".......ordinal data and it is possible to do ranking tests such as Kruskal-Wallis modified for contigency tables
DRIVING MR. & MRS. JONES ON MAIN STREET: Are Paulsons $700B Turncoats to buy Wall Street Toxic Bonds or Dole them to Bernanke to cover the FEDS Trillion lent to the Defunct Oil Speculators Corporate America against same Devilish Stock Insurance Circle.
Posted in Corporate Bonds 101 | 
Tags: business, corporate bonds data, corporate bonds database, finance, investing, research, stocks