Simple and flexible SAS and SPSS programs for analyzing lag-sequential categorical data

Area of Science:

• Behavioral Science
• Data Analysis
• Statistical Software

Background:

• Lag-sequential analysis is crucial for understanding temporal dependencies in categorical data.
• Existing methods may lack flexibility or comprehensive statistical output.
• The need for accessible tools in statistical software like SAS and SPSS is evident.

Purpose of the Study:

• To present simple and flexible SAS and SPSS programs for lag-sequential categorical data analysis.
• To provide a wide range of statistical measures for detailed behavioral sequence analysis.
• To facilitate the examination of temporal patterns and dependencies in data.

Main Methods:

• Development of SAS and SPSS programs to process streams of categorical codes.
• Implementation of various lag-sequential statistics: transitional frequencies, probabilities, adjusted residuals, and Yule's Q.
• Inclusion of likelihood ratio tests for stationarity and homogeneity.
• Calculation of transformed kappas for different types of dependence (unidirectional, bidirectional, dominance).
• Application of parametric and randomization tests for significance levels.

Main Results:

• The programs successfully generate a comprehensive suite of lag-sequential statistics.
• They enable detailed analysis of transitional frequencies, probabilities, and dependence measures.
• Statistical tests for stationarity and homogeneity are readily available.
• Significance levels are provided using both parametric and randomization approaches.

Conclusions:

• The developed programs offer a powerful yet accessible method for lag-sequential analysis.
• Researchers can efficiently analyze complex categorical data sequences and dependencies.
• These tools enhance the study of temporal dynamics in various scientific fields.