7. ANALISIS FACTORIAL EXPLORATORIO SPSS FÁCIL Y RÁPIDO 🤓 ANÁLISIS E INTERPRETACIÓN PASO A PASO

New Section

The introduction to the concept of Factorial Analysis and its importance in research and test validation is discussed.

Understanding Validity Analysis in Factorial Analysis

• Validity analysis assesses the accuracy with which a test measures the intended psychological construct, crucial for test quality.

• Construct validity evaluates if the operational definition of a variable truly reflects the theoretical meaning of a concept by identifying common dimensions among items.

• Factorial analysis methods, exploratory and confirmatory, are essential for grouping items into expected dimensions to reflect theoretical meanings accurately.

• Exploratory factor analysis identifies common dimensions among variables, forming latent variables that cannot be directly measured.

• Factorial analysis aids in understanding variable structures, questionnaire construction, and reducing data complexity for better measurement accuracy.

Exploring Dimension Reduction in Factorial Analysis

Delving deeper into how factorial analysis aids in exploring variable structures and reducing data complexity effectively.

Utilizing Exploratory Factor Analysis

• Different age groups may require distinct latent variables due to developmental processes; exploratory factor analysis helps identify these variations.

• For adult studies where established latent variables exist, prior specification might be beneficial; however, exploratory analysis remains valuable for population-specific insights.

• When analyzing tests with dichotomous items like summing scores rather than individual item entry is necessary for factorial exploration efficiency.

Conducting Factorial Analysis Procedures

Detailed steps on conducting factorial analysis procedures efficiently and accurately.

Implementing Factorial Analysis Techniques

Analysis of Factorial Validity

In this section, the speaker discusses the criteria for assessing the adequacy of factor analysis and explains the Bartlett's sphericity test as a tool to determine if factors exist in a model.

Criteria for Factor Analysis Validity

• The range from 0 to .5 indicates unsuitability for factorial analysis with sample data.

• Adequacy ranges: .51-.7 (mediocre), .71-.8 (acceptable), .81-.9 (good), >.91 (excellent).

• Bartlett's sphericity test assesses if correlation matrix equality holds; p<.05 rejects null hypothesis, supporting researcher's hypothesis.

Extraction and Rotation in Factor Analysis

This part covers extraction methods like principal components and rotation techniques such as varimax or oblimin, emphasizing theoretical basis for selecting orthogonal or oblique rotations.

Extraction and Rotation Techniques

• Choose principal components extraction for identifying factors within a specific population.

• Opt for orthogonal rotation like varimax when underlying latent variables are theoretically related.

• Select oblique rotation such as oblimin or promax when no evidence supports relationships among latent variables.

Interpreting Factor Analysis Results

The discussion focuses on interpreting factor analysis results, including creating new latent variable columns and setting thresholds for factor loadings.

Interpreting Results

• Use scoring option to generate new columns with latent variables from factorial extraction.

• Consider KMO score (.87) indicating good model fit and significance in Bartlett's test to proceed with factor analysis.

• Analyze variance explained by each factor; accept only eigenvalues >1 during extraction process.

Understanding Factor Loadings and Rotations

Exploring communalities, final factor loadings, and naming latent variables based on task groupings post-extraction and rotation.

Factor Loadings Interpretation

• Communalities show common variance within variables pre/post-factor extraction.

• Identify primary components explaining variance; assign names based on task clusters like language or motor functions.

Validation through Rotational Matrix Examination

Evaluating rotational matrix coherence through diagonal values to validate construct validity of the test.

Validation Process

New Section

In this section, the speaker discusses the requirements for reporting results in APA format and mentions the components that need to be included. Additionally, a future topic of construct validity using confirmatory factor analysis is introduced.

Reporting Results in APA Format

• The following must be reported in APA format:

• Value of KMO

• Chi-square from Bartlett's test with p-value

• Table of communalities

• Table of rotated components

Construct Validity Discussion

• The next type of construct validity will involve confirmatory factor analysis.

• Further details on this topic will be covered in the upcoming video.

Conclusion and Call to Action

• Viewers are encouraged to leave any questions they may have in the comments section.