To test whether a third variable is responsible for an observed correlation.

-

To test whether a third variable is responsible for an observed correlation.




Answer: To test whether your X-Y correlation is spurious - you need to first identify what third variable is. (Note: you're allowed to have more than one third variable and you can test all of your possible third variables in one study.) Then you run a new study in which you measure the third variable(s), together with the original two variables, and conduct a covariance analysis (which removes any part of the target relationship that can be explained by the covariates). The new correlation between X & Y is now called a "partial correlation (with respect to the covariates)." If the partial correlation is just as strong as the original, then the third variable(s) did not cause the X-Y correlation. If the partial correlation is zero, the original X-Y correlation was spurious. Anything in the middle is a mix of the two.

Comments

Post a Comment

Popular posts from this blog

What is JAD and how are users involved? What are advantages and disadvantages of JAD?

-
What is JAD and how are users involved? What are advantages and disadvantages of JAD? Joint application development, is a popular fact finding technique that brings users into the development process as active participants. Users have a vital stake in the information system, and should participate fully in the development process. They are/should be involved at every stage of development. Advantages : Users can participate effectively in the requirements modeling stage. More accurate statement of systems requirements, better understanding of common goals and stronger commitment to success of the new project Disadvantages : More expensive, can be cumbersome if the group is too large for the project.
Read more

What are the three dimensions to business problems? Give an example of each.

-
What are the three dimensions to business problems? Give an example of each. Answer: The three dimensions to business problems are: (1) Organizational: This may include outdated business processes, unsupportive culture and attitudes, political conflict, turbulent business environment and change, complexity of task, inadequate resources. (2) Technology: This may include insufficient or aging hardware, outdated software, inadequate data management, insufficient telecommunications capacity, incompatibility of old systems with new technology, and rapid technological change. (3) People: This may include lack of employee training, difficulties in evaluation performance, legal and regulatory compliance, work environment, lack of employee support and participation, indecisive management, or poor management.
Read more

A laboratory assistant prepared solution of 0.8 M, 0.6 M, 0.4 M, and 0.2 M sucrose, but forgot to label them. After realizing the error, the assistant randomly labeled the flasks containing these four unknown solutions as flask A, flask B, flask C, and flask D.

-
A laboratory assistant prepared solution of 0.8 M, 0.6 M, 0.4 M, and 0.2 M sucrose, but forgot to label them. After realizing the error, the assistant randomly labeled the flasks containing these four unknown solutions as flask A, flask B, flask C, and flask D. Design an experiment, based on the principles of diffusion and osmosis, that the assistant could use to determine which of the flasks contains each of the four unknown solutions. Include in your answer (a) a description of how you would set up and perform the experiment: (b) the results you would expect from your experiments: and (c) an explanation of those results based on the principles involved. (Be sure to clearly state the principles addressed in your discussion.)
Read more