RobertNiles.com
About Robert
Reporting Help
Finding Data on the Internet
Questions for Candidates
Stats Lessons
Mean
Median
Percent
Per capita
Standard Deviation
Margin of Error
Data Analysis
Sample Sizes
Stats Tests
Help Board
Bookstore


Statistics for the Utterly Confused

This book offers a super-accessible approach to the much-misunderstood subject of statistics.
More information
-->

Home > Statistics Every Writer Should Know > The Stats Board > Discusssion

sample of bivariate data
Message posted by Han on November 22, 1999 at 12:00 AM (ET)

I'm assigned to find a set of bivariate data sample 50, of any topic and analyse it. It is mainly a discussion on using the Pearson's Product Moment Correlation Coefficient. My questions are:

1) Where do I get such a small-sized set of data and yet ensuring the randomness of such a small-sized sample?

2) What sort of bivariate data are being analysed from time to time, due to its high importance? Where can I find a sample of readily analysed and interpreted data?

3) How do I know if if it would give me a positive correlation coefficient (besides calculating it)?

3) How much can the correlation coefficient tell me?

Thanx a lot!!!


READERS RESPOND:
(In chronological order. Most recent at the bottom.)

Re: sample of bivariate data
Message posted by Bill on November 22, 1999 at 12:00 AM (ET)

1) What are the requirements for the sampling process? Could you stop every xth student entering the student union and ask for their height and weight.
2) I'm not sure what we mean in point 2.
3) You don't - that's why you do the project although height and weight should be positively correlated.
4) The obtained correlation will tell you if the obtained linear correlation is significantly greater than 0.


Re: sample of bivariate data
Message posted by JG on November 22, 1999 at 12:00 AM (ET)

Asking people their height and weight is a good way to get bivariate data. You could also go to the financial page in the paper and pick any two columns of numbers, such as opening and closing prices, any two stock indeces, etc.
Before you do any computations you should construct a scatter diagram and see what the data looks like. This will also give you a good idea what correlation coefficient to expect.
Remember that the square of the correlation coefficient measures it's explanatory power. Also, correlation does not imply causality and zero correlation does not imply independence.



Your $5 contribution helps cover part the $500 annual cost of keeping this site online.

Niles Online Sites:RobertNiles.comTheme Park InsiderViolinist.com

RobertNiles.com™, the site, content and services 咀opyright 1996-2002, Robert Niles.
All rights reserved. Questions? Comments? Read my Privacy Policy, or E-mail me!