STAT 151 Exam 1

Chapters 1 2 3

Research (Definition)

A problem solving activity. Involves a systematic search of information on a specific topic.

Data collected must be: adequate, representative, accurate, and without bias.

Research is used to:

• Establish or confirm facts

• Reaffirm the results of previous work

• Solve new or existing problems

• Support theorems or develop new theories

Research is not:

• Accidental discovery

• Data collection

• Searching out published research results in libraries or the internet

Qualitative Research Type

Observation based research used to gain understanding of latent construct or ideas. Helpful in construction of hypothesis for quantitative research.

Quantitative Research Type

Research that utilizes statistical or computational tools on data extracted through some instrument or scale

Pure Research

Contributes to the theory or philosophy of knowledge

Applied Research

Seeks practical applications of an existing knowledge

Objectives of a research study

• Determine the characteristics of the target universe or population

• Compare two or more groups

• Determine the degree of association between or among variables

• Predict the value of one variable in other variables

Correlation Analysis

Statistical technique used to determine the degree of linear relationship between two or more variables

Two questions for correlation analysis

• Does an association exist? If yes, how strong?

• What is the pattern or direction of the association?

Properties of correlation coefficient (obvious ones not here)

• Correlation treats X and Y symmetrically (P_xy = P_yx)

• Has no unit and should not be expressed as a percentage

• Is not affected by changes in the center, or scale of either variable

• Measures the degree of linear association

  • Two variables can be strongly associated but have small correlation if association is not linear

• Correlation is sensitive to outliers, one outlier can make correlation small or large

Lurking Variables

Hidden variable that stands behind a relationship and determines it by simultaneously affecting the other two variables

Partial Correlation

Between two variables is one that nullifies the effect of a third variable (or more) upon both the variables being correlated