Notes on Carter et al. (2022) and Confirmatory Factor Analysis of the K6 Scale
Tutorial 4: Confirmatory Factor Analysis of the Kessler-6 Psychological Distress Scale
Overview
This tutorial focuses on the paper by Carter et al. (2022), specifically analyzing Tables 1, 2, and 3, as well as Figures 1 and 2, to understand the Confirmatory Factor Analysis (CFA) of the Kessler-6 (K6) Psychological Distress Scale. The K6 scale is used to measure psychological distress in a community sample of people living with severe and persistent mental illness.
The Kessler-6 (K6) Psychological Distress Scale
- The K6 scale aims to measure psychological distress.
- It consists of six questions about how often a person felt:
- Nervous
- Hopeless
- Restless or fidgety
- So depressed that nothing could cheer you up
- That everything was an effort
- Worthless
- Respondents answer each question based on the past 30 days.
- Response options range from "None of the time" to "All of the time."
Scoring Procedure
- Each question is coded from 0 to 4.
- The scores are then summed to obtain a total score.
- The score ranges from 0 to 24.
- A score range of 13-24 indicates a serious mental illness clinical range.
- Reference: Kessler, R. C., Andrews, G., Colpe, L. J., Hiripi, E., Mroczek, D. K., Normand, S. L., … & Zaslavsky, A. M. (2002). Short screening scales to monitor population prevalences and trends in non-specific psychological distress. Psychological Medicine, 32(6), 959-976.
Key Issues with the K6 Scale
- A central question is whether the K6 scale measures a general factor of psychological distress or if it captures two dimensions: depression and anxiety.
Analysis of Tables and Figures
Table 1: Sample Description
- Table 1 provides information about the sample used in the study.
Table 2: Distributions of Responses
- Table 2 shows the distributions of responses for each item on the K6 scale.
- The goal is to determine if the responses are skewed or roughly evenly distributed across the Likert scale.
Figure 1 vs. Figure 2: Model Testing
- Figure 1: Represents one model being tested.
- Figure 2: Represents another model being tested.
- The task is to explain what each model is testing.
Confirmatory Factor Analysis (CFA) vs. Exploratory Factor Analysis (EFA)
- The study used CFA instead of EFA.
- CFA is used to test pre-existing hypotheses (i.e., test a theory) unlike EFA which is exploratory in nature.
Correlation Between Factors (F1 and F2) in Figure 2
- F1 and F2 are correlated in Figure 2.
- The size and valence (positive or negative) of the correlation are important.
Table 3: Goodness-of-Fit Statistics
- Table 3 presents goodness-of-fit statistics for the single-factor and two-factor models of the K6 scale.
- The goal is to determine which model has a better fit based on these statistics.