Reliability Coefficients

Similar documents
Subsetting, dplyr, magrittr Author: Lloyd Low; add:

Study Guide. Module 1. Key Terms

050 0 N 03 BECABCDDDBDBCDBDBCDADDBACACBCCBAACEDEDBACBECCDDCEA

050 0 N 03 BECABCDDDBDBCDBDBCDADDBACACBCCBAACEDEDBACBECCDDCEA

Principal Leadership Reliability Measures

Chapitre 2 : modèle linéaire généralisé

SEM 1: Confirmatory Factor Analysis

SEM 1: Confirmatory Factor Analysis

How to use the SimpleCOMPASS Interface

Reproducible Research Week4

The lavaan tutorial. Yves Rosseel Department of Data Analysis Ghent University (Belgium) December 18, 2017

Day 1: Working with Data

Using R and the psych package to find ω

Data Analysis and Solver Plugins for KSpread USER S MANUAL. Tomasz Maliszewski

Introduction to Factor Analysis for Marketing

Path Analysis using lm and lavaan

Data visualization with ggplot2

Instructions for uploading a file into YouTube, setting up YouTube s Creator Studio and creating closed captions.

Package autovarcore. June 4, 2018

Contents 1 Admin 2 Testing hypotheses tests 4 Simulation 5 Parallelization Admin

Advanced Plotting Di Cook, Eric Hare May 14, 2015

Main, Paxton, & Dale Data Analysis

Cut Out The Cut And Paste: SAS Macros For Presenting Statistical Output ABSTRACT INTRODUCTION

Contents 1 Admin 2 Testing hypotheses tests 4 Simulation 5 Parallelization Admin

Using R to score personality scales

Using R to score personality scales

2006 Annual Report for the. National Athletic Trainers Association Board of Certification. Inc. CASTLE Worldwide, Inc. March, 2007

Simulating Multivariate Normal Data

A Worked Example of Goldberg s Bass Ackwards Method

Introduction to R. Nishant Gopalakrishnan, Martin Morgan January, Fred Hutchinson Cancer Research Center

New Features in qgraph Sacha Epskamp

ehealth literacy and Cancer Screening: A Structural Equation Modeling

CDAA No. 4 - Part Two - Multiple Regression - Initial Data Screening

Package MultiVarMI. April 9, 2018

1, if item A was preferred to item B 0, if item B was preferred to item A

Using EdSurvey to Analyze NAEP Data: An Illustration of Analyzing NAEP Primer

Package nsprcomp. August 29, 2016

The Mplus modelling framework

Package starank. May 11, 2018

Package starank. August 3, 2013

Using R for statistics and data analysis

Introduction to Mixed Models: Multivariate Regression

Unified Visualizations of Structural Equation Models

An Introductory Guide to R

Chong Ho Yu, Ph.D., MCSE, CNE. Paper presented at the annual meeting of the American Educational Research Association, 2001, Seattle, WA

Standard Errors in OLS Luke Sonnet

Package misclassglm. September 3, 2016

STATISTICS (STAT) Statistics (STAT) 1

NAME: BEST FIT LINES USING THE NSPIRE

Cluster Analysis. Angela Montanari and Laura Anderlucci

Supplementary Material

Package fsrm. January 23, 2016

Functions and data structures. Programming in R for Data Science Anders Stockmarr, Kasper Kristensen, Anders Nielsen

Performance of Latent Growth Curve Models with Binary Variables

Math CST 2003 Accuracy

Online Supplementary Appendix for. Dziak, Nahum-Shani and Collins (2012), Multilevel Factorial Experiments for Developing Behavioral Interventions:

Sub-setting Data. Tzu L. Phang

WELCOME! Lecture 3 Thommy Perlinger

DU PhD in Physical Education

Distributed Itembased Collaborative Filtering with Apache Mahout. Sebastian Schelter twitter.com/sscdotopen. 7.

WEEK 8: FUNCTIONS AND LOOPS. 1. Functions

3. CENTRAL TENDENCY MEASURES AND OTHER CLASSICAL ITEM ANALYSES OF THE 2011 MOD-MSA: MATHEMATICS

Comparison of Linear Regression with K-Nearest Neighbors

Independent Variables

Description/History Objects/Language Description Commonly Used Basic Functions. More Specific Functionality Further Resources

An Introduction to Growth Curve Analysis using Structural Equation Modeling

Package filematrix. R topics documented: February 27, Type Package

Why use R? Getting started. Why not use R? Introduction to R: Log into tak. Start R R or. It s hard to use at first

Differential Expression Analysis at PATRIC

Kingsley Associates Portal User Guide

Using the Manage Institution Users Tool

Product Catalog. AcaStat. Software

lavaan: an R package for structural equation modeling and more Version (BETA)

Package apricom. R topics documented: November 24, 2015

Count outlier detection using Cook s distance

Mathematics Masters Examination

Package SMAT. January 29, 2013

The SQLiteDF Package

Python for Data Analysis. Prof.Sushila Aghav-Palwe Assistant Professor MIT

Programming with R. Educational Materials 2006 S. Falcon, R. Ihaka, and R. Gentleman

Introduction to R. Educational Materials 2007 S. Falcon, R. Ihaka, and R. Gentleman

Package doppelgangr. March 25, 2019

Multivariate Capability Analysis

Programming with R. Educational Materials 2006 S. Falcon, R. Ihaka, and R. Gentleman

Simple_h2o Rob McCulloch April 15, 2019

Package texteffect. November 27, 2017

Application of Hierarchical Clustering to Find Expression Modules in Cancer

Package tidystats. May 6, 2018

Introduction to R: Using R for Statistics and Data Analysis. BaRC Hot Topics

Florida State University Libraries

Stat 579: Objects in R Vectors

Package linkspotter. July 22, Type Package

Intermediate Programming in R Session 2: Loops. Olivia Lau, PhD

Bivariate (Simple) Regression Analysis

Section 4: FWL and model fit

Special review: Evaluation of the exploratory factor analysis programs provided in SPSSX and SPSS/PC+

Introduction: EViews. Dr. Peerapat Wongchaiwat

- 1 - Fig. A5.1 Missing value analysis dialog box

Package endogenous. October 29, 2016

Estimation of Item Response Models

Transcription:

Reliability Coefficients Introductory notes That data used for these computations is the pre-treatment scores of all subjects. There are three items in the SIAS (5, 9 and 11) that require reverse-scoring. Those items are stored in the data frame in reverse-coded format. In other words, if a subject responded to one of those items with a score of 0, it has been coded as a 4; if the client responded with a score of 1 it has been coded as a 3, and so on. The following code computes categorical McDonald s omega using the MBESS::ci.reliability() function, classical Cronbach s alpha and the mean inter-item correlation using psych::alpha(). In general, we would expect McDonald s omega values to be higher than the alpha values, as (i) they are based on polychoric correlations, rather than Pearson correlations, and (ii) because it is well known in the psychometric literature that alpha tends to under-estimate reliability for congeneric measurement models. # Load libraries library(dplyr) Attaching package: 'dplyr' The following objects are masked from 'package:stats': filter, lag The following objects are masked from 'package:base': intersect, setdiff, setequal, union library(psych) library(mbess) Attaching package: 'MBESS' The following object is masked from 'package:psych': cor2cov library(lavaan) This is lavaan 0.5-23.1097 lavaan is BETA software! Please report any bugs. Attaching package: 'lavaan' 1

The following object is masked from 'package:mbess': cor2cov The following object is masked from 'package:psych': cor2cov # Load datafile load("dat.rda") # Setup dummy matrix for storing store reliabilty coefficients x <- matrix(na, nrow = 12, ncol = 3) rownames(x) <- c("sias20", "SIAS19", "SIAS17", "SIAS6", "SIPS-INT", "SPS", "SPS6", "SIPS-FOE", "SIPS-FAA", "SIPS-AGG", "Peters", "SIPS-Tot") colnames(x) <- c("omega", "alpha", "mean_r") # Function to compute omega, alpha, and mean inter-item correlation rel_stats <- function(data, items){ df <- data[, items] x <- c(mbess::ci.reliability(data = df, type = "categorical", interval.type = "none")$est, psych::alpha(df)[[1]][c(1,4)] %>% unlist() ) names(x) <- c("omega", "alpha", "mean_r") x } # Compute Omega, alpha and mean-inter-item correlation # Extract SIAS items at pre-treatment df <- dat %>% select(sias_0_1:sias_0_20) # Compute stats for SIAS measures x[1, ] <- rel_stats(df, 1:20) # sias20 x[2, ] <- rel_stats(df, c(1:4, 6:20)) # sias19 x[3, ] <- rel_stats(df, c(1:4, 6:8, 10, 12:20)) # sias17 x[4, ] <- rel_stats(df, c(2, 4, 6, 8, 10, 13)) # sias6 x[5, ] <- rel_stats(df, c(7, 10, 15, 16, 19)) # sips-int # Extract SPS items at pre-treatment df2 <- dat %>% select(sps_0_1:sps_0_20) 2

# Compute stats for SPS measures x[6, ] <- rel_stats(df2, 1:20) # sps20 x[7, ] <- rel_stats(df2, c(4, 7, 8, 15:17)) # sps6 x[8, ] <- rel_stats(df2, c(4, 6, 8, 13, 16, 17)) # sips-foe x[9, ] <- rel_stats(df2, c(12, 14, 15)) # sips-faa x[10, ] <- rel_stats(df2, c(4, 6, 8, 12:17)) # sips-agg # Peters total (SIAS6 + SPS6) df3 <- cbind(df[,c(2, 4, 6, 8, 10, 13)], df2[, c(4, 7, 8, 15:17)]) x[11, ] <- rel_stats(df3, 1:12) # SIPS total df4 <- cbind(df[,c(7, 10, 15, 16, 19)], df2[, c(4, 6, 8, 12:17)]) x[12, ] <- rel_stats(df4, 1:14) # sips total # Display coefficients round(x, 2) omega alpha mean_r SIAS20 0.91 0.83 0.21 SIAS19 0.90 0.83 0.23 SIAS17 0.89 0.85 0.26 SIAS6 0.71 0.69 0.27 SIPS-INT 0.81 0.79 0.45 SPS 0.96 0.91 0.34 SPS6 0.86 0.84 0.47 SIPS-FOE 0.88 0.86 0.51 SIPS-FAA 0.75 0.74 0.48 SIPS-AGG 0.91 0.87 0.44 Peters 0.87 0.83 0.29 SIPS-Tot 0.97 0.87 0.33 Note that in the above code, separate factor models were specified for each SIPS subscale. An alternate way of computing omega for the SIPS is to define the full two- or three-factor model, and use those solutions to compute omega. library(semtools) # This is semtools 0.4-14 All users of R (or SEM) are invited to submit functions or ideas for functions. # 3

Attaching package: 'semtools' The following object is masked from 'package:psych': skew # Rearrange SIPS items so that items 1-5 = interaction scale, 6-11 are foe and 12-14 faa df5 <- cbind(df[,c(7, 10, 15, 16, 19)], df2[, c(4, 6, 8, 13, 16, 17, 12, 14, 15)]) names(df5) <- paste0("v", 1:14) # SIPS 3 factor model m1 <- ' INT =~ v1 + v2 + v3 + v4 + v5 FOE =~ v6 + v7 + v8 + v9 +v10 + v11 FAA =~ v12 + v13 + v14 ' # Omega = Raykov's omega, omega2 = bentler's omega, omega3 = McDonald's omega # Note that alpha is model based on polychoric correlations and thus will differ # from traditional alpha (in the above table) computed using Pearson correlations fit_m1 <- cfa(m1, data = df5, estimator = "WLSMV", std.lv = TRUE, orthogonal = FALSE, missing = "pairwise") round(semtools::reliability(fit_m1),2) INT FOE FAA total alpha 0.79 0.86 0.74 0.87 omega 0.78 0.87 0.74 0.90 omega2 0.78 0.87 0.74 0.90 omega3 0.76 0.87 0.73 0.90 avevar 0.41 0.53 0.49 0.49 The code below compute omega for the SIPS two-factor model, where the two performance anxiety subscales (FOE and FAA) are aggregated. m2 <- ' INT =~ v1 + v2 + v3 + v4 + v5 AGG =~ v6 + v7 + v8 + v9 +v10 + v11 + v12 + v13 + v14 ' fit_m2 <- cfa(m2, data = df5, estimator = "WLSMV", std.lv = TRUE, orthogonal = FALSE, missing = "pairwise") round(semtools::reliability(fit_m2),2) INT AGG total 4

alpha 0.79 0.87 0.87 omega 0.78 0.88 0.89 omega2 0.78 0.88 0.89 omega3 0.76 0.88 0.89 avevar 0.41 0.45 0.45 SIAS-6 correlation matrix The reliability coefficients for the SIAS-6 is notably lower than for the other scales. We can probe the reason for this by examining the inter-item correlations which should be quite low (the mean inter-item correlation is.27). cor(df[, c(2, 4, 6, 8, 10, 13)]) %>% round(2) sias_0_2 sias_0_4 sias_0_6 sias_0_8 sias_0_10 sias_0_13 sias_0_2 1.00 NA 0.28 0.21 0.33 NA sias_0_4 NA 1 NA NA NA NA sias_0_6 0.28 NA 1.00 0.42 0.28 NA sias_0_8 0.21 NA 0.42 1.00 0.27 NA sias_0_10 0.33 NA 0.28 0.27 1.00 NA sias_0_13 NA NA NA NA NA 1 5