Variable-Centered Approach

 

·        A Problem: Often students are confused about the various statistical concepts, especially in inference. One source of this confusion is a result of fuzziness in the use of the terms variable and population.

 

·        A Solution:  By employing a variable-centered approach—where the terms variable and population are used consistently and properly—those statistical concepts can be unified and clarified.

 

·        Details: The following table provides details about employing a variable-centered approach with some commonly used inferential procedures. By specifying the variable(s) and population(s) for any given problem, students can more easily understand the nature of the problem and inference and, as well, better comprehend similarities and differences among the various types of inferences.

 

Type of

inference

Number of populations

Number of variables

Type of variables

Example of population(s)

Example of variable(s)

One mean

1

1

Quantitative

Females

Height

Two means

2

1

Quantitative

Females/Males

Height

ANOVA

k

1

Quantitative

Four U.S. regions

Energy consumption

Simple regression

1

2

Quantitative

Corvettes

Age/Price

Independence test

1

2

Categorical

U.S. adults

Edu level/Income class

Homogeneity test

k

1

Categorical

Four U.S. regions

Political party

 

 

·        Example: The American Association of University Professors (AAUP) conducts salary studies of college professors and publishes its findings in AAUP Annual Report on the Economic Status of the Profession. Suppose that we want to decide whether the mean annual salaries of college faculty in public and private institutions are different.

 

Populations: Here there are two populations—college faculty in public institutions and college faculty in private institutions.

 

Variable: Here there is one variable—annual salary.