Ordinal Confirmatory Factor Analysis Research Articles

Cross-sectional evaluation of an asynchronous multiple mini-interview (MMI) in selection to health professions training programmes with 10 principles for fairness built-in

ObjectivesWe aimed to explore the psychometric properties of the first known online asynchronous multiple mini-interview (MMI) designed for fairness with subgroup analyses by key characteristics, usability and acceptability.DesignCross-discipline multimethod evaluation.SettingOne...

The Penn State Worry Questionnaire Is Invariant Across Age Groups

Prevalence of GAD is between 3 to 5% with onset in the early to mid-twenties (Kessler et al. 2009). The Penn State Worry Questionnaire (PSWQ) is a 16-item self-report instrument assessing generalized anxiety symptoms (Meyers, et al., 1990; Molina & Borkovec, 1994). Brown (2003) and Olatunji et al. (2007) conducted Confirmatory Factor Analyses identifying a two factor model of Worry Engagement and Absence of Worry. No published studies have examined the factor structure of the PSWQ across age groups. The current study presents data from 612 people across three groups: 221 young adults (Mage = 19.31, SD = 1.21), 283 middle-age adults (Mage = 48.27, SD = 5.13), and 108 older adults (Mage = 72.95, SD = 7.19). An ordinal confirmatory factor analysis (CFA) using robust weighted least squares (WLSMV) tested for invariance across groups. Results showed CFI/TLI values of .983/.981, 984/.983, and .981/.984 for Configural (CI), Metric (MI), and Scalar (SI) models. The RMSEA for CI, MI, and SI models was .064, .061, and .059. Based upon Cheung and Rensvold (2002), Sass (2011), and Chen (2007), a cutoff of ΔCFI ≥ 0.01 was established as evidence of invariance. The ΔCFI between CI and MI models was < .01 so analysis continued with the SI test. The ΔCFI between MI and SI models was < 0.01 and did not justify rejection of the null hypothesis. These analyses suggest PSWQ scores are valid across age groups and provide additional support for the multidimensional nature of the PSWQ.

Confirmatory Factor Analysis of the Perceived Stress Scale-10 Across Age Groups

The Perceived Stress Scale-10 (PSS; Cohen et al, 1983) was developed to measure subjective elements of stress. Most measures focus on objective characteristics of stress (e. g., frequency of occurrence) and specific situations that produce stress (e. g., divorce) but ignore the cognitive appraisal associated with stressful stimuli. The PSS-10 assesses the interplay between stressor and appraisal-mediated coping ability. Factor analytic studies provide support for two factors: perceived helplessness and perceived self-efficacy (see Roberti et al. 2006). The current study presents data from 591 people across three groups: 221 young adults (mean age 19.31, SD = 1.21), 283 middle-age adults (mean age 48.27, SD = 5.14), and 109 older adults (mean age 72.95, SD = 7.22). An ordinal confirmatory factor analysis (CFA) using robust weighted least squares (WLSMV) evaluated invariance across age groups. Results showed CFI/TLI values of .964/.953, 965/.960, and .964/.969 for Configural (CI), Metric (MI), and Scalar (SI) models. The RMSEA for CI, MI, and SI models was .086, .081, and .071. Based upon recommendations of Cheung and Rensvold (2002), Sass (2011), and Chen (2007), a cutoff of ΔCFI ≥ 0.01 was established as evidence of invariance. The ΔCFI between CI and MI models was < .01 so analysis continued with the SI test. Once again, ΔCFI between MI and SI models was < 0.01 which did not justify rejection of the null hypothesis. Based on these analyses, PSS-10 scores are valid across multiple age groups. Further, results support the multidimensional nature of the PSS-10.

Psychometric Evaluation of the Czech Version of Group Cohesiveness Scale (GCS) in a Clinical Sample: A Two-Dimensional Model.

The Group Cohesiveness Scale (GCS, 7 items) measures patient-rated group cohesiveness. The English version of the scale has demonstrated good psychometric properties. This study describes the validation of the Czech version of the GCS. A total of 369 patients participated in the study. Unlike the original study, the ordinal confirmatory factor analysis (CFA) supported a two-dimensional solution (RMSEA = 0.075; TLI = 0.986). The analysis demonstrated the existence of two moderately to highly associated (r = 0.79) domains of group cohesiveness—affective and behavioral. The two-dimensional model was invariant across genders, age, education, and time (retest after 6 weeks) up to factor means level. Internal consistency reached satisfactory values for both domains (affective, ω = 0.86; behavioral, ω = 0.81). In terms of convergent validity, only weak association was found between the GCS domains and the group working alliance measured by the Group Outcome Rating Scale (GSRS). This is the first revision of the factor structure of the GCS in the European context. The scale showed that the Czech version of the GCS is a valid and reliable brief tool for measuring both aspects of group cohesiveness.

Inventář tendence k vině a hanbě (ITVAH): pilotní studie

This paper constitutes a pilot study of a novel measure for assessing proneness to guilt and proneness to shame, Inventář tendence k vině a hanbě (ITVAH, Inventory of Proneness to Guilt and Shame). ITVAH is a scenario-based method distinguishing affective and behavioural components of guilt and shame and also considering the private or public experience of guilt and shame, respectively. Two items (affective and behavioural reaction) were created for each of the 16 scenarios, resulting in total of 32 items. A four-factor measurement model was assumed – Negative behaviour evaluation (NBE), Reparation, Negative self-evaluation (NSE), Withdrawal. Next, we assumed negative correlations of NSE and Withdrawal with self-esteem and self-forgiveness and positive correlations of NBE and Reparation with self-forgiveness. The pilot study was carried out on 249 Czech-speaking adults (67 % women, aged 18–75, Md = 22). An ordinal confirmatory factor analysis with residual covariances between items connected to the same scenarios was conducted. The hypothesized four-factor model was found acceptable. However, we found limited evidence for the hypothesized relationships with self-esteem and self-forgiveness – in contrast to theoretical expectations we found a medium-strong negative relationship between self-forgiveness and affective component of guilt. For further use of the measure in the Czech and Slovak environment it is necessary to assess additional evidence of construct validity and concurrent validity with The Test of Self-Conscious Affect-3 (TOSCA-3).

Validity and reliability of the Arabic version of the "Personal Wellbeing Index-Adults" on adults with hearing impairment.

Background: The Personal Wellbeing Index-Adults (PWI-A) is the most widely used instrumentfor measuring subjective-quality of life (QoL). The current study seeks to investigate the constructvalidity and reliability of the Arabic version of the PWI-A on adults with bilateral hearingimpairment by comparing the single-factor solution with the two-factor solution. Methods: A cross-sectional study was conducted at the Audio-Vestibular Medicine Unit of Alexandria University from July-2017 to January-2018. A total of 205 adults were interviewed tomeasure the subjective-QoL using the PWI-A instrument. Internal consistency was determinedusing both Cronbach’s alpha and composite reliability (CR). Validity was assessed by constructvalidity, including ordinal regression, ordinal exploratory factor analysis (OEFA), and ordinalconfirmatory factor analysis (OCFA). Results: The first four items of the PWI-A which are: satisfaction with living standard, health,achievements, and relationships were the most important indicators of subjective-wellbeing(Part r2 0.0547, 0.0324, 0.0361, and 0.0225, respectively). OEFA suggested that the two-factormodel contributes better than the single-factor model. OCFA validated this suggested solution;(two-factor: RMSEA=0.084 (90% CI=0.01-0.14); CFI=0.964; AIC=52.64; single-factor: RMSEA=0.119 (90% CI=0.07-0.17); CFI=0.922; AIC=62.77). Good internal consistency wasalso presented (two-factor: Cronbach’s alpha=0.719, 0.693; single-factor: Cronbach’s alpha =0.750). Conclusion: The Arabic version of the PWI-A is a multidimensional scale that consists of twodimensions: the first is related to subjective-QoL, and the second is related to satisfaction withthe community. Thus, it is recommended to use the short version of the PWI-A with only fouritems to measure subjective-QoL, as it achieved sufficient reliability and construct validity.

EXPLORING METHOD EFFECTS IN THE SIX-FACTOR STRUCTURE OF THE SURVEY OF ATTITUDES TOWARD STATISTICS (SATS-36)

We use ordinal confirmatory factor analysis techniques to investigate the six-factor structure of the Survey of Attitudes Toward Statistics (SATS-36) and to estimate method effects associated with its items and factors. We extend previous confirmatory research to include posttest, as well as pretest, item-level data. We also investigate method effects by adding a common method factor to the original six-factor model. Interestingly, results reveal noticeable proportions of common variance associated with the Difficulty construct only at pretest and with the Value and Interest constructs only at posttest. We examine the characteristics of the SATS items as a possible source for these method effects.&#x0D; First published November 2019 at Statistics Education Research Journal Archives

Development and validation of the Kernis-Goldman authenticity inventory-short form (KGAI-SF)

Psychological authenticity has long been of interest to social thinkers, and empirical research on this topic has steadily increased since the 1970s. The present work contributes to this body of knowledge by describing the development of a short form for the Kernis-Goldman Authenticity Inventory (KGAI), a widely used self-report instrument. Thirteen of the 45 items from the original KGAI were removed on the basis of semantic ambiguity or redundancy, and the remaining 32-items were subjected to cross-validated ordinal confirmatory factor analysis (N = 1252) in an effort to identify psychometrically redundant or unreliable items. The resulting 20-item short-form (KGAI-SF) exhibited good reliability across the ranges of the four underlying dimensions of authenticity (awareness, unbiased processing, behavior, relational orientation), as well as good discriminant and convergent validity.

A novel method for expediting the development of patient-reported outcome measures and an evaluation across several populations.

Item response theory (IRT) models provide an appropriate alternative to the classical ordinal confirmatory factor analysis (CFA) during the development of patient-reported outcome measures (PROMs). Current literature has identified the assessment of IRT model fit as both challenging and underdeveloped (Sinharay & Johnson, 2003; Sinharay, Johnson, & Stern, 2006). This study evaluates the performance of Ordinal Bayesian Instrument Development (OBID), a Bayesian IRT model with a probit link function approach, through applications in two breast cancer-related instrument development studies. The primary focus is to investigate an appropriate method for comparing Bayesian IRT models in PROMs development. An exact Bayesian leave-one-out cross-validation (LOO-CV) approach (Vehtari & Lampinen, 2002) is implemented to assess prior selection for the item discrimination parameter in the IRT model and subject content experts' bias (in a statistical sense and not to be confused with psychometric bias as in differential item functioning) toward the estimation of item-to-domain correlations. Results support the utilization of content subject experts' information in establishing evidence for construct validity when sample size is small. However, the incorporation of subject experts' content information in the OBID approach can be sensitive to the level of expertise of the recruited experts. More stringent efforts need to be invested in the appropriate selection of subject experts to efficiently use the OBID approach and reduce potential bias during PROMs development.

Shared Vision Among Acute Care Magnet® Hospital Nurses: Ordinal Confirmatory Factor Analysis.

Psychometric testing of the Shared Vision (SV) scale that measures team efforts toward common patient-centered goals was initially estimated among rural hospital nurse executives. The purpose of this study was to estimate the scale's reliability (internal consistency), convergent validity (Pearson correlation with Practice Environment Scale), and structural validity (ordinal confirmatory factor analysis) among acute care Magnet® hospital nurses. The study sample included 289 nurses from 27 acute care Magnet® hospitals. The scale demonstrated acceptable estimates for internal consistency (Cronbach's α = .902, 95% confidence interval [CI] = [0.883, 0.919]), convergent validity ( r = .720, p < .001), and structural validity with a one-factor structure. The findings of this study supported the reliability and validity of the SV scale as a unidimensional construct in measuring SV among nurses in acute care Magnet® hospitals. Further testing among different nursing providers and health care settings is needed to accumulate evidence and expand use of the instrument.

Assessing health-related quality of life with the SCOPA-PS in French individuals with Parkinson's disease having undergone DBS-STN: A validation study

IntroductionThe aims of this study were to validate the French version of the SCales for Outcomes in Parkinson's Disease-PsychoSocial (SCOPA-PS) in individuals with Parkinson's disease (PD) who underwent deep brain stimulation of the subthalamic nucleus (DBS-STN), to confirm the unifactorial structure of this questionnaire, and to establish its psychometric properties. MethodsRoutinely used psychological questionnaires (BDI-II, STAI-Y, PDQ-39, UPDRS III) and the SCOPA-PS were used for a cross-sectional observational study of 154 PD patients. SCOPA-PS acceptability, scaling assumption, reliability, ordinal confirmatory factor analysis and validity were assessed. ResultsThe ICC for two-week test-retest reliability was 0.88. SEM was 8.42. In confirmatory factor analysis, the one-factor model showed an acceptable fit to the data (Chi2/df=2.130; CFI=0.976; RMSEA=0.086). No floor or ceiling effects were observed. Skewness was 0.33. Item-total correlation coefficients ranged from 0.47 to 0.71. Cronbach's alpha was 0.86. SCOPA-PS SI correlated with PDQ-39 SI (rs=0.83) and with state-anxiety and depression (rs=0.56 and 0.69 respectively). The SCOPA-PS SI was higher in more depressed patients and in those with the most severe PD motor symptoms. Conclusion and discussionSCOPA-PS French version is a one-factor scale with satisfactory psychometric properties consistent with other language versions. This short scale can be used to evaluate the psychosocial component of QoL in PD patients treated with DBS-STN.

Factorial structure of the Health of the Nation Outcome Scales: An ordinal confirmatory factor analysis using a national sample of clinician ratings in England.

In the present study, the superior factorial structure of the Health of the Nation Outcome Scales (HoNOS) was investigated. The aims of the study were to test the fit of our new four-factor/subscale structure within and across a wide range of clinical populations compared to competing structures, and to identify any clinical populations where the HoNOS might require modification in order to improve its psychometric properties. Ordinal confirmatory factor analysis was performed for four competing HoNOS subscale structures using a national sample of 80,161 ratings taken at the point of referral to mental health services in England. A new four-factor structure with good/acceptable internal consistency for each of the factors/subscales demonstrated a significantly better fit to the data for 21 of the 24 data groups compared to three competing subscale models. A new four-factor/subscale structure was shown to be the most generalizable subscale structure currently available for HoNOS. However, although superior to other factorial structures, the statistical fit was less than optimal for some clinical populations. Therefore, we would recommend that additional items relevant to all mental health patients are identified and added to the HoNOS in order to allow the development of a single, more generalizable subscale structure, and a much improved version of the HoNOS.

Comparing interval estimates for small sample ordinal CFA models.

Robust maximum likelihood (RML) and asymptotically generalized least squares (AGLS) methods have been recommended for fitting ordinal structural equation models. Studies show that some of these methods underestimate standard errors. However, these studies have not investigated the coverage and bias of interval estimates. An estimate with a reasonable standard error could still be severely biased. This can only be known by systematically investigating the interval estimates. The present study compares Bayesian, RML, and AGLS interval estimates of factor correlations in ordinal confirmatory factor analysis models (CFA) for small sample data. Six sample sizes, 3 factor correlations, and 2 factor score distributions (multivariate normal and multivariate mildly skewed) were studied. Two Bayesian prior specifications, informative and relatively less informative were studied. Undercoverage of confidence intervals and underestimation of standard errors was common in non-Bayesian methods. Underestimated standard errors may lead to inflated Type-I error rates. Non-Bayesian intervals were more positive biased than negatively biased, that is, most intervals that did not contain the true value were greater than the true value. Some non-Bayesian methods had non-converging and inadmissible solutions for small samples and non-normal data. Bayesian empirical standard error estimates for informative and relatively less informative priors were closer to the average standard errors of the estimates. The coverage of Bayesian credibility intervals was closer to what was expected with overcoverage in a few cases. Although some Bayesian credibility intervals were wider, they reflected the nature of statistical uncertainty that comes with the data (e.g., small sample). Bayesian point estimates were also more accurate than non-Bayesian estimates. The results illustrate the importance of analyzing coverage and bias of interval estimates, and how ignoring interval estimates can be misleading. Therefore, editors and policymakers should continue to emphasize the inclusion of interval estimates in research.

Psychometric analysis of the Ten-Item Perceived Stress Scale.

Although the 10-item Perceived Stress Scale (PSS-10) is a popular measure, a review of the literature reveals 3 significant gaps: (a) There is some debate as to whether a 1- or a 2-factor model best describes the relationships among the PSS-10 items, (b) little information is available on the performance of the items on the scale, and (c) it is unclear whether PSS-10 scores are subject to gender bias. These gaps were addressed in this study using a sample of 1,236 adults from the National Survey of Midlife Development in the United States II. Based on self-identification, participants were 56.31% female, 77% White, 17.31% Black and/or African American, and the average age was 54.48 years (SD = 11.69). Findings from an ordinal confirmatory factor analysis suggested the relationships among the items are best described by an oblique 2-factor model. Item analysis using the graded response model provided no evidence of item misfit and indicated both subscales have a wide estimation range. Although t tests revealed a significant difference between the means of males and females on the Perceived Helplessness Subscale (t = 4.001, df = 1234, p < .001), measurement invariance tests suggest that PSS-10 scores may not be substantially affected by gender bias. Overall, the findings suggest that inferences made using PSS-10 scores are valid. However, this study calls into question inferences where the multidimensionality of the PSS-10 is ignored.

Measurement Effects of Survey Mode on the Equivalence of Attitudinal Rating Scale Questions

This study applies ordinal confirmatory factor analysis for multiple groups to assess equivalence of scale, random errors and systematic (nonrandom) errors of attitudinal questions surveyed on rating scales under different survey modes (Face-to-Face [F2F], Telephone, Paper, and Web). Empirical findings from a large-scale experiment are presented. Consistent with theoretical expectations, interviewer- and self-administered surveys measured all assessed questions on systematically different scales, with different systematic bias, and with differing extents of random error. These measurement effects were absent when comparing Paper with Web or F2F with Telephone. It is concluded that modes impact primarily systematic measurement effects affecting multiple items equally. Interviewer- and self-administered modes should only be combined with great care in mixed-mode surveys that focus on attitudinal constructs. Combining Paper and Web or Telephone and F2F are the viable options. Thereby choosing the self-administered modes appears more efficient, because these modes exhibited higher indicator reliabilities (smaller random error) than the interviewer modes.

The index of tobacco treatment quality: development of a tool to assess evidence-based treatment in a national sample of drug treatment facilities

BackgroundQuitting smoking improves health and drug use outcomes among people in treatment for substance abuse. The twofold purpose of this study is to describe tobacco treatment provision across a representative sample of U.S. facilities and to use these data to develop the brief Index of Tobacco Treatment Quality (ITTQ).MethodsWe constructed survey items based on current tobacco treatment guidelines, existing surveys, expert input, and qualitative research. We administered the survey to a stratified sample of 405 facility administrators selected from all 3,800 U.S. adult outpatient facilities listed in the SAMHSA Inventory of Substance Abuse Treatment Services. We constructed the ITTQ with a subset of 7 items that have the strongest clinical evidence for smoking cessation.ResultsMost facilities (87.7%) reported that a majority of their clients were asked if they smoke cigarettes. Nearly half of facilities (48.6%) reported that a majority of their smoking clients were advised to quit. Fewer (23.3%) reported that a majority of their smoking clients received tobacco treatment counseling and even fewer facilities (18.3%) reported a majority of their smoking clients were advised to use quit smoking medications. The median facility ITTQ score was 2.57 (on a scale of 1–5) and the ITTQ displayed good internal consistency (Cronbach’s alpha = .844). Moreover, the ITTQ had substantial test-retest reliability (.856), and ordinal confirmatory factor analysis found that our one-factor model for ITTQ fit the data very well with a CFI of 0.997 and an RMSEA of 0.042.ConclusionsThe ITTQ is a brief and reliable tool for measuring tobacco treatment quality in substance abuse treatment facilities. Given the clear-cut room for improvement in tobacco treatment, the ITTQ could be an important tool for quality improvement by identifying service levels, facilitating goal setting, and measuring change.