Intro to Marketing Research: Factor Analysis Factor Loadings Communalities Total Variance Explained

What It Is

Factor Loadings, Communalities, and Total Variance Explained are key concepts in factor analysis, a statistical method used to reduce the dimensionality of large datasets. In marketing research, factor analysis helps identify underlying patterns and relationships between variables. A famous example is the work of Paul Meehl and J.P. Guilford, who used factor analysis to identify personality traits in the 1940s. This matters for marketing decision-making as it helps identify the underlying drivers of consumer behavior and preferences, enabling targeted marketing strategies.

Key Terms & Concepts

• Factor Loadings: The correlation between a variable and a factor. A high loading indicates that the variable is strongly related to the factor. (e.g., in a study on consumer preferences, a high loading on the "sustainability" factor might indicate that the variable "environmental concerns" is strongly related to that factor.)
• Communalities: The amount of variance in a variable that is explained by the factors. A high communality indicates that the variable is well-explained by the factors. (e.g., in a study on consumer attitudes, a high communality for the variable "brand loyalty" might indicate that it is well-explained by the factors "brand image" and "customer satisfaction.")
• Total Variance Explained (TVE): The percentage of variance in the data that is explained by the factors. A high TVE indicates that the factors are good at explaining the data. (e.g., in a study on consumer behavior, a high TVE might indicate that the factors "demographics" and "psychographics" are good at explaining consumer behavior.)
• Eigenvalues: The amount of variance explained by each factor. A high eigenvalue indicates that the factor is good at explaining the data. (e.g., in a study on consumer preferences, a high eigenvalue for the factor "sustainability" might indicate that it is good at explaining consumer preferences.)
• Scree Plot: A graphical representation of the eigenvalues, used to determine the number of factors to retain. (e.g., in a study on consumer attitudes, a scree plot might be used to determine the number of factors to retain based on the eigenvalues.)
• Kaiser Criterion: A rule of thumb for determining the number of factors to retain, which suggests retaining factors with eigenvalues greater than 1. (e.g., in a study on consumer behavior, the Kaiser criterion might be used to determine the number of factors to retain based on the eigenvalues.)
• Factor Rotation: A technique used to simplify the factor loadings and improve the interpretability of the factors. (e.g., in a study on consumer preferences, factor rotation might be used to simplify the factor loadings and improve the interpretability of the factors.)
• Varimax Rotation: A type of factor rotation that maximizes the variance of the factor loadings. (e.g., in a study on consumer attitudes, varimax rotation might be used to maximize the variance of the factor loadings.)
• Oblique Rotation: A type of factor rotation that allows for correlations between factors. (e.g., in a study on consumer behavior, oblique rotation might be used to allow for correlations between factors.)
• Factor Score Coefficients: The coefficients used to calculate the factor scores. (e.g., in a study on consumer preferences, factor score coefficients might be used to calculate the factor scores.)
• Factor Scores: The scores calculated for each case based on the factor loadings and factor score coefficients. (e.g., in a study on consumer attitudes, factor scores might be used to calculate the scores for each case based on the factor loadings and factor score coefficients.)

Common Misunderstandings

• Misunderstanding: Factor loadings are the same as factor scores.
• Correction: Factor loadings are the correlations between a variable and a factor, while factor scores are the scores calculated for each case based on the factor loadings and factor score coefficients. (e.g., in a study on consumer behavior, factor loadings might be used to identify the underlying factors, while factor scores might be used to calculate the scores for each case based on those factors.)
• Misunderstanding: The Kaiser criterion is a hard and fast rule for determining the number of factors to retain.
• Correction: The Kaiser criterion is a rule of thumb, and the number of factors to retain should be determined based on the research question and the data. (e.g., in a study on consumer preferences, the Kaiser criterion might be used as a guideline, but the number of factors to retain should be determined based on the research question and the data.)
• Misunderstanding: Factor rotation is a necessary step in factor analysis.
• Correction: Factor rotation is an optional step in factor analysis, and it should only be used if it improves the interpretability of the factors. (e.g., in a study on consumer attitudes, factor rotation might be used to simplify the factor loadings and improve the interpretability of the factors, but it is not necessary.)

Quick Application / Identification

Scenario: A marketing researcher is conducting a study on consumer preferences for eco-friendly products. The researcher uses factor analysis to identify the underlying factors that drive consumer preferences. The researcher finds that the factor "sustainability" has a high loading on the variable "environmental concerns." What is the researcher trying to identify?

Answer: The researcher is trying to identify the underlying factor that drives consumer preferences for eco-friendly products, which is the factor "sustainability."

Explanation: The researcher is using factor analysis to identify the underlying factors that drive consumer preferences, and the high loading on the variable "environmental concerns" indicates that the factor "sustainability" is strongly related to that variable.

Last-Minute Revision

• Factor loadings are the correlations between a variable and a factor.
• Communalities are the amount of variance in a variable that is explained by the factors.
• Total Variance Explained (TVE) is the percentage of variance in the data that is explained by the factors.
• Eigenvalues are the amount of variance explained by each factor.
• Scree plot is a graphical representation of the eigenvalues.
• Kaiser criterion is a rule of thumb for determining the number of factors to retain.
• Factor rotation is an optional step in factor analysis.
• Varimax rotation is a type of factor rotation that maximizes the variance of the factor loadings.
• Oblique rotation is a type of factor rotation that allows for correlations between factors.
• Factor score coefficients are the coefficients used to calculate the factor scores.
• Factor scores are the scores calculated for each case based on the factor loadings and factor score coefficients.
  ⚠️ Factor loadings are not the same as factor scores.
  ⚠️ The Kaiser criterion is a rule of thumb, not a hard and fast rule.
  ⚠️ Factor rotation is not necessary in all cases.