F03-Corr-Reg

# F03-Corr-Reg - PubH 7405 REGRESSION ANALYSIS Simple Correlation Regression Variables A variable represents a characteristic or a class of

This preview shows pages 1–11. Sign up to view the full content.

PubH 7405: REGRESSION ANALYSIS Simple Correlation & Regression

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
Variables A variable represents a characteristic or a class of measurement . It takes on different values on different subjects/persons. Examples include weight, height, race, sex, SBP, etc. The observed values, also called “observations,” form items of a data set . Depending on the scale of measurement, we have different types of data .
There are “ observed variables ”(Height, Weight, etc… each takes on different values on different subjects/person) and there are “ calculated variables (Sample Mean, Sample Proportion, etc… each is a “statistic” and each takes on different values on different samples). The Standard Deviation of a calculated variable is called the Standard Error of that variable/statistic.

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
A “variable” – sample mean, sample standard deviation, etc… included – is like a “ function ”; when you apply it to a target element in its domain, the result is a “number”. For example, “height” is a variable and “the height of Mrs. X” is 135 lbs; it’s a number.
TYPES OF DATA There are binary or dichotomous outcomes, e.g. Sex/gender (male/female), Morbidity (sick/well) There are categorical or polytomous outcomes, eg. Race (white/black/Hispanics/Asian) There are continuous outcomes, e.g. blood pressure, cholesterol level); of course, you can dichotomized or categorized a continuous outcome to make it binary or categorical – but some information are lost in the process.

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
In most problem involving statistical inference, we investigate one variable at a time . However, in many important investigations, we may have two measurements made on each subject , and the research objective is concerned not with each of them but with the relationship between them .
AN EXAMPLE : IN SEARCH OF AN HONEST EMPLOYEE Shoplifting is a big problem, it costs up to 2 billions dollars a year in America Who done it? Customers? Yes, but customer shoplifting ranks second to employee theft which involves between 2% and 3% of all employees.

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
SOLUTION? One approach to curtailing employee theft is screen job applicants so as not to hire those with “high potential” to theft. How to do it? How about using polygraph test (lie detector)? But who want to apply? you need to treat your future employee with dignity!
AN ALTERNATIVE May be a less visible pencil-and- paper test as part of the application . Need: to device some kind of a questionnaire; but its “ score ” should be “ highly correlated ” to to the result by the polygraph test.

This preview has intentionally blurred sections. Sign up to view the full version.

View Full Document
ANOTHER : RESEARCH IN AN AMUSEMENT PARK? Yes, they do it for business planning : designing questionnaires, selecting samples, conducting interviews, and analyzing data that provide information about visitors’ attitudes, perceptions, and preferences.
This is the end of the preview. Sign up to access the rest of the document.

## This note was uploaded on 11/21/2011 for the course PUBH 7405 taught by Professor Staff during the Fall '08 term at Minnesota.

### Page1 / 75

F03-Corr-Reg - PubH 7405 REGRESSION ANALYSIS Simple Correlation Regression Variables A variable represents a characteristic or a class of

This preview shows document pages 1 - 11. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online