Robustness of factor solutions in exploratory factor analysis

Replicability has become a highly discussed topic in psychological research. The debates focus mainly on significance testing and confirmatory analyses, whereas exploratory analyses such as exploratory factor analysis are more or less ignored, although hardly any analysis has a comparable impact on entire research areas. Determining the correct number of factors for this analysis is probably the most crucial, yet ambiguous decision—especially since factor structures have often been not replicable. Hence, an approach based on bootstrapping the factor retention process is proposed to evaluate the robustness of factor retention criteria against sampling error and to predict whether a particular factor solution may be replicable. We used three samples of the “Big Five Structure Inventory” and four samples of the “10 Item Big Five Inventory” to illustrate the relationship between stable factor solutions across bootstrap samples and their replicability. In addition, we compared four factor retention criteria and an information criterion in terms of their stability on the one hand and their replicability on the other. Based on this study, we want to encourage researchers to make use of bootstrapping to assess the stability of the factor retention criteria they use and to compare these criteria with regard to this stability as a proxy for possible replicability.

Latent Factor Structure of Outcome Measures Used in the HABIT® Mild Cognitive Impairment Intervention Programs

Background: In Alzheimer’s disease and related disorders (ADRD) research, common outcome measures include cognitive and functional impairment, as well as persons with mild cognitive impairment (pwMCI) and care partner self-reported mood and quality of life. Studies commonly analyze these measures separately, which potentially leads to issues of multiple comparisons and/or multicollinearity among measures while ignoring the latent constructs they may be measuring. Objective: This study sought to examine the latent factor structure of a battery of 12-13 measures of domains mentioned above, used in a multicomponent behavioral intervention (The HABIT® program) for pwMCI and their partners. Methods: Exploratory factor analysis (EFA) involved 214 pwMCI-partner pairs. Subsequent Confirmatory factor analyses (CFA) used 730 pairs in both pre- and post-intervention conditions. Results: EFA generated a three-factor model. Factors could be characterized as partner adjustment (29.9%), pwMCI adjustment (18.1%), and pwMCI impairment (12.8%). The subsequent CFA confirmed our findings, and the goodness-of-fit for this model was adequate in both the pre- (CFI = 0.937; RMSEA = 0.057, p = 0.089) and post-intervention (CFI = 0.942; RMSEA = 0.051, p = 0.430) groups. Conclusion: Results demonstrated a stable factor structure across cohorts and intervention conditions suggesting that three broad factors may provide a straightforward and meaningful model to assess intervention outcome, at least during the MCI phase of ADRD.

Empirics vs. art theory: Exploring a factor structure of pictorial expression based on contemporary artworks

As a transdisciplinary questionnaire combining psychometrics and art theory, RizbA enables a quantitative measurement of pictorial expression in terms of a formal image analysis. This study explores its factor structure and a potential gap between theory and empirics.The sample consists of 275 contemporary pictorial works by artists and nonprofessionals, rated by 179 experts in a randomized online survey. Three path models were specified and computed as CFA: models A and B based on the results of the two previous studies, model C on the theoretical framework of the initial study. Model C was additionally tested on a combined dataset of all three studies.While models A and B did not converge, model C was associated with fit indices as follows: χ2 = 1299.752, df = 278, p = .000, RSMEA = .122 (90% CI = .116, .129), CFI = .712, TLI = .679, SRMR = .135 and for the combined dataset: χ2 = 6860.824, df = 278, p = .000, RSMEA = .086 (90% CI = .084, .088), CFI = .740, TLI = .696, SRMR = .084.Studies implicate reliability, but there might not be a globally stable factor structure across artworks. Only model C partly suggests an acceptable fit for the combined data. The results speak to a methodological gap between empirics and theory due to art being highly ambiguous with various analysis options. Further postdisciplinary approaches are needed to develop a theory-based measurement model for pictorial expression, which might to do justice to the arts.

Resonance frequency is not always stable over time and could be related to the inter-beat interval

Heart Rate Variability Biofeedback (HRVB) is based on breathing at an optimal rate (or resonance frequency, RF) corresponding to the respiratory sinus arrhythmia (RSA). Our aim is to check whether the RF is a stable factor and analyse the HRV parameters individually per each breathing rate, comparing it with free slow breathing. A sample of 21 participants were trained in a test–retest HRVB protocol. The results indicated that RF changed between Test and Retest sessions in 66.7% of participants. This instability could be related to the average of interbeat interval (IBI). HRV time domain parameters (SDNN and RMSSD) were significantly higher for RF than for other breathing rates, including 6 breath/min and free slow breathing. Free slow breathing showed a lower heart rate averages than RF and other slow breathing rates. Overall, our study suggests the relevance of assessing RF individually and before each HRVB session, because the maximum cardiovascular benefits in terms of increasing HRV were found only at RF. Thus, breathing at the individualized and momentary frequency of resonance increases cardiac variability.

Innovative m-car rental service quality in India

Purpose The purpose of the study is to explore and analyze m-car rental service quality dimensions. Design/methodology/approach Exploratory factor analysis method is applied to explore the m-car rental service quality dimensions. Further, confirmatory factor analysis is done to prove the reliability and validity of the factors using AMOS 22.0. Findings The results reveal the m-car rental service quality dimensions: ambient quality, technical quality, comfort, safety and employee service, mobile convenience, mobile responsiveness, mobile efficiency and reliability and mobile safety and billing. Research limitations/implications The explored dimensions of car rental services are in Indian environment. So, these dimensions can be further validated in other similar cultural context. Practical implications The proposed measurements can also be applied to measure and compare the service quality performance of car rental firms. Originality/value Current literature does not confirm the stable factor structure of m-car rental service quality. This study confirms the reliable and valid dimensions of care rental service through mobile app.

Validation and Psychometric Assessment of the Practice Environment Scale of the Nursing Work Index in the Portuguese Hospital Context

BackgroundThe nursing work environment has an impact on patient safety outcomes and its measurement should be a regular practice.PurposeTo assess the reliability and construct validity of the Practice Environment Scale of the Nursing Work Index in the Portuguese context.MethodsAn exploratory factor analysis followed by a confirmatory factor analysis to assess model adjustment quality was performed with a sample of 3,686 nurses.ResultsThe final solution for the exploratory factor analysis comprised 26 items and five factors (56.6% of the total variance). The confirmatory factor analysis, after refinement, showed a stable factor structure.ConclusionsThe final model showed good construct validity and high reliability, which supports the decision to exclude the items that are not essential to the construct being measured.

A Szakmai Elkötelezettség és A Munkahelyről Való Kilépési Szándék Összefüggései Az Életpálya Előrehaladtával

HáttérA mai karrierutakra a töredezettség, szekvenciák sora jellemző, ami együtt jár a szakmai célok, attitűdök és az identitás újrafogalmazásával (Wehmeyer és mtsai, 2019; Ashforth és Saks, 1995). Másrészt a szakmai identitás, a szakmával való azonosulás hosszú távú, stabilabb tényező (Colarelli és Bishop, 1990), kevésbé hat rá negatívan a munkahelyváltás, ha az egyén nem pályaelhagyó.CélKutatásunkban az idői tényezők tükrében vizsgáljuk a szakmához és a munkatevékenységhez való viszonyulást. Arra keressük a választ, hogy az életkor, a szakmában és a munkahelyen eltöltött idő mennyire határozza meg a pálya alakulása szempontjából fontos tényezőket, a szakmai elkötelezettséget, a munkába való bevonódást, a szubjektív pályasikerességet, és ezek együtt milyen kapcsolatban vannak a munkahelyről való kilépési szándékkal.MódszerAz online kutatásban (N = 1172) állandó munkaviszonnyal rendelkező munkavállalók vettek részt. A változók közötti összefüggéseket egy útmodell segítségével, valamint a korcsoportok összehasonlításával ellenőriztük.EredményekAz eredmények alapján az mondható, hogy a kilépési szándékot az idői tényezők nem befolyásolják közvetlenül, csak más tényezőkön (a munkába való bevonódás, szakmai elkötelezettség, szubjektív pályasikeresség) keresztül, míg a szakmai elkötelezettséget és a munkába való bevonódást befolyásolja a szakmában eltöltött idő és az életkor. A korcsoportok összehasonlítása arra mutat rá, hogy az életkorral növekszik a bevonódás és elkötelezettség mértéke, ugyanakkor a kilépési szándék csak a negyvenes korosztályig mutat csökkenő tendenciát, a legidősebb korosztálynál a harmincasok szintjére emelkedik ez az érték.BackgroundRecent career paths are typically fragmented, consisting of sequences, so redefi nition of work values, work attitudes and professional identity is necessary (Wehmeyer et al., 2019; Ashforth & Saks, 1995). On the other hand, professional identity and commitment is of long perspective, and more stable factor (Colarelli & Bishop, 1990), so change of workplace will more moderately affect it, if the individual stays I his profession.PurposeThe goal is to explore the impact of time factors on attitudes to profession and work. We would like to explore that in what extent the career relevant factors such as professional commitment, work engagement, and subjective career are infl uenced by age, years spent in profession and at recent workplace, and how these constructs relate to turnover cognitions.MethodsEmployees with active permanent employment were examined in an online survey (N = 1172). Hypothesis were tested by path analysis and comparing age groups.ResultsAccording to our results, time factors do not infl uence turnover cognitions in a direct way, just through variables of work engagement, professional commitment and subjective career success, while they affect the professional commitment and work engagement directly. The results of generational analysis reveal that the degree of work engagement and professional commitment increases with age, but the degree of the willingness of turnover decreases until the age of 40s, while it increases in the oldest group to the level of age of 30s.

CancerSupportSource®-15+: development and evaluation of a short form of a distress screening program for cancer patients and survivors

Purpose CancerSupportSource® (CSS) is a distress screening program implemented at community-based organizations and hospitals nationwide. The 25-item CSS assesses distress across five domains, with capacity to screen for clinically significant depression and anxiety. This study examined psychometric properties of a shortened form to enhance screening opportunities when staff or patient burden considerations are significant. Methods Development and validation were completed in multiple phases. Item reduction decisions were made with 1436 cancer patients by assessing external/internal item quality and judging theoretical and practical implications of items. Pearson correlations and confirmatory factor analysis were conducted on a separate sample of 957 patients to corroborate psychometric properties and dimensionality of the shortened scale. Nonparametric receiver operating characteristic (ROC) curve analyses determined scoring thresholds for depression and anxiety risk scales. Results Scale refinement resulted in a 15-item short form plus one screening item assessing tobacco and substance use (CSS-15+). At least two items from each CSS domain were retained to preserve multidimensionality. In confirmatory analysis, the model explained 59% of the variance and demonstrated good fit. Correlation between CSS-15+ and 25-item CSS was 0.99, p < 0.001. Sensitivity of 2-item depression and 2-item anxiety risk scales in the confirmatory sample were 0.82 and 0.83, respectively. Conclusions CSS-15+ is a brief, reliable, and valid multidimensional measure of distress. The measure retained excellent internal consistency (α = 0.94) and a stable factor structure. CSS-15+ is a practical and efficient screening tool for distress and risk for depression and anxiety among cancer patients and survivors, particularly in community-based settings.

Trust shifting as a reaction to pandemic: Cross country analysis

Governance systems all over the world are coming under huge stress. Nowadays, two factors become crucial in the fight against the pandemic and its negative consequences: the state's ability to withstand stress in the economy and society; and civil support and approval of the governments' anti-crisis actions. This study aims to recognize the difference in stress response in different countries of the world. As a criterion for such a response, we consider the level of trust to governments in the context of the Covid-19 pandemic. The Fragile States Index was used as a measure of the state fragility, and the indices of trust were used as indicators of the people's reaction. The study showed that there is no correlation between 1) the trust in the government and the fragility of the state; 2) the support for government actions during the pandemic and the trust to the government before the pandemic. The clustering of countries by the set of indices of trust and fragility showed that the clusters' means support both the assumptions of a direct and an inverse relationship between trust and state resistance to stress. In response to quarantine measures, we see multidirectional trends in both stable and fragile states: trust can grow, fall, or remain unchanged. However, in stronger states, the tendency to an increase in trust is stronger, while in weaker states - to its fall, which confirms the thesis that in crisis the weak weakens and the strong strengthens. The results of the analyses provide arguments in favor of the following. The stability of the state does not guarantee the high trust of citizens and support for its actions during a crisis, and high trust is not a stable factor of high support for government actions. At the same time, it can be expected that the positive reaction of citizens to government measures (increased trust) will be more significant than disappointment (loss of trust).