What are the advantages and disadvantages of different factor rotation methods for ordinal data?

2 Why is factor rotation important for ordinal data?

Ordinal data are data that can be ordered or ranked, but not measured on a continuous scale. For example, a survey question that asks respondents to rate their satisfaction with a service on a scale from 1 (very dissatisfied) to 5 (very satisfied) is an ordinal variable. Ordinal data pose some challenges for factor analysis, because they do not meet the assumptions of normality and linearity that are required for most parametric methods. Therefore, special techniques, such as polychoric correlation and ordinal factor analysis, are needed to account for the ordinal nature of the data. Factor rotation is important for ordinal data, because it can help to reveal the underlying structure and meaning of the factors, and to avoid misleading or ambiguous results.

3 What are the advantages and disadvantages of orthogonal rotation for ordinal data?

Orthogonal rotation is the simplest and most common type of factor rotation. It preserves the independence and simplicity of the factors, and makes it easier to compare the results across different studies and samples. Some examples of orthogonal rotation methods are varimax, quartimax, and equamax. However, orthogonal rotation also has some disadvantages for ordinal data. It may not capture the true relationships among the factors, especially if they are correlated in reality. It may also produce factors that are difficult to interpret or label, because they are not related to any meaningful concept or construct.

4 What are the advantages and disadvantages of oblique rotation for ordinal data?

Oblique rotation is a more complex and flexible type of factor rotation. It allows for some correlation among the factors, and can produce more realistic and nuanced results. Some examples of oblique rotation methods are promax, oblimin, and direct oblimin. However, oblique rotation also has some disadvantages for ordinal data. It may introduce some redundancy and complexity in the factors, and make it harder to compare the results across different studies and samples. It may also require more subjective judgment and expertise to select the appropriate rotation method and criterion, and to interpret the results.

5 How to choose the best factor rotation method for ordinal data?

There is no definitive answer to the question of which factor rotation method is the best for ordinal data. The choice depends on several factors, such as the research question, the theoretical framework, the data characteristics, and the analytical goals. However, some general guidelines can be followed to make an informed decision. First, consider whether your factors are expected to be independent or correlated, and choose an orthogonal or oblique rotation accordingly. Second, consider whether you want to maximize the simplicity or the realism of your factors, and choose a rotation method that reflects your preference. Third, consider whether you want to emphasize the loadings of a few variables or many variables on each factor, and choose a rotation method that achieves your balance. Fourth, consider whether you want to use a fixed or a variable number of factors, and choose a rotation method that supports your decision. Fifth, consider whether you want to use a standard or a customized rotation criterion, and choose a rotation method that allows for your adjustment.

6 Here’s what else to consider

This is a space to share examples, stories, or insights that don’t fit into any of the previous sections. What else would you like to add?