New computations for RMSEA and CFI following FIML and TS estimation with missing data.

PurposeTo evaluate the internal structure (structural validity and internal consistency) and propose a classification for the distress caused by the presence of pelvic floor dysfunction (PFD) symptoms based on the total score of the Pelvic Floor Distress Inventory (PFDI-20).MethodsCross-sectional study conducted with Brazilian women over 18 years of age. Exploratory and confirmatory factor analysis were performed with Parallel Analysis and to test three models to compare them with the Root Mean Square Error of Approximation (RMSEA) and Comparative Fit Index (CFI). Internal consistency was calculated using Cronbach's alpha. Partial credit model (PCM) was performed to classify the total score of the PFDI-20.ResultsData from 237 women (49.62 ± 16.95 years) were analyzed. The one-dimensional structure had 43.74% of the explained variance with α = 0.929. The one-dimensional model was the most appropriate (CFI = 0.987 and RMSEA = 0.022). The total PFDI-20 score was classified as the absence of symptoms (score zero), symptoms with mild distress (1 to 15 points), symptoms with moderate distress (16 to 34 points), and symptoms with severe distress (35 to 40 points).ConclusionThe PFDI-20 has an one-dimensional structure and the distress caused by the presence of PFD symptoms can be classified as mild, moderate and severe. Health professionals and future studies can use our classification to facilitate the understanding of the patient's health status and to obtain other analyses on the severity of the distress of the symptoms of PFD.

Missing data are nearly always a problem in research, and missing values represent a serious threat to the validity of inferences drawn from findings. Increasingly, social science researchers are turning to multiple imputation to handle missing data. Multiple imputation, in which missing values are replaced by values repeatedly drawn from conditional probability distributions, is an appropriate method for handling missing data when values are not missing completely at random. However, use of this method requires developing an imputation model from the observed data. This is typically a rigorous and time-consuming process. To encourage wider adoption of multiple imputation in social work research, a simple framework for designing imputation models is presented. The framework and its ability to generate unbiased estimates are demonstrated in a simulation study. KEY WORDS: missing data; multiple imputation; nonresponse ********** Missing data are ubiquitous in social research, and missingness or nonresponse can represent a threat to the validity of inferences because of undue effects on efficiency, power, and parameter bias (Shadish, Cook, & Campbell, 2002). Social work researchers are now addressing missing data in a more rigorous manner. Recently, Saunders et al. (2006) and Choi, Golder, Gillmore, and Morrison (2005) described important data imputation methods and dispelled misunderstandings regarding popular imputation methods, such as mean substitution. Recent advances in analytic methods, such as multiple imputation (MI), are taking hold in social work research. With MI, missing values are replaced with values repeatedly drawn from simulated conditional probability distributions (Schafer, 1997), thus creating multiple versions of the data set. Each version of the data set is analyzed according to the data analysis model, and the multiple results are combined into point estimates (Rubin, 1996). A critical task in MI is to devise an imputation model (Allison, 2002) or missing data model (Graham, Olchowski, & Gilreath, 2007), which involves specifying the measures that are putatively associated with the missing values. Although this process adds additional steps, the specification of an imputation model and the creation of multiple data sets can produce less-biased estimates in the presence of missing data across a wide variety of data analysis techniques (Schafer, 1997). Besides MI, there are many other methods for addressing missing data (Schafer, 1999; Schafer & Graham, 2002). An equally rigorous method known as direct or full information maximum likelihood (FIML) estimation can produce unbiased estimates and correct standard errors in the presence of missing data. When the number of imputations is sufficiently large, identical missing data models will produce the same estimates under MI and FIML (Graham et al., 2007). Unlike MI, FIML is limited to maximum likelihood analytic techniques and the missing data model must be included in the analysis model. Although we focus on MI, the steps we describe for developing an imputation model are equally appropriate for use in the missing data model for FIML. On the basis of the MI literature, this article describes a framework for developing an imputation model for use with any free or commercial software package that performs MI. BRIEF REVIEW OF MISSING DATA CONCEPTS Generally, both MI and a broad range of missing data issues have received ample attention in the applied literature (Graham et al., 2007). We briefly discuss the three types of distributions that describe the randomness of nonresponse given that this property has consequences for the development of an imputation model. For a discussion of general missing data concepts that are not critical to understanding our discussion of imputation model development, we refer readers to Schafer and Graham (2002). Distribution of Nonresponse The probability distribution of nonresponse--more frequently referred to as the missing data or nonresponse mechanism (Rubin, 1976)--is both an important factor in the decision to impute with MI and a context for the development of an imputation model. …

Strategic Self‐Regulation for Speaking English as a Foreign Language: Scale Development and Validation

Background and Aim: Community-based surveillance (CBS) is a critical mechanism for early detection of infectious diseases. Understanding the behavioral drivers of CBS participation is essential to strengthening community engagement. This study employed structural equation modeling (SEM) based on the theory of reasoned action (TRA) to investigate the impact of knowledge, subjective norms (SN), and attitudes on the intention and behavioral likelihood (BL) of participating in CBS activities. Materials and Methods: A cross-sectional survey was conducted among 470 schoolteachers selected through a multi-stage mixed sampling strategy across Kelantan, Malaysia. A structured questionnaire assessing sociodemographic factors, knowledge, attitudes, and perceptions toward CBS was developed and validated. Confirmatory factor analysis and SEM were employed with model parameters estimated using the robust maximum likelihood (MLR) approach. Model fit was assessed using comparative fit index (CFI), Tucker-Lewis index (TLI), root mean square error of approximation (RMSEA), and standardized root mean square residual (SRMR) indices. Results: The final model demonstrated good fit (CFI = 0.923; TLI = 0.913; RMSEA = 0.045; SRMR = 0.070). Knowledge (β = 0.335) and SN (β = 0.296) positively influenced intention to participate in CBS, whereas negative attitudes (β = −0.313) showed a significant negative association. Intention significantly predicted BL (β = 0.633). The model explained 40% of the variance in intention and 43% in BL. Intention mediated the effects of knowledge, norms, and attitudes on behavioral engagement. Conclusion: Knowledge, positive SN, and reduced negative attitudes are pivotal in fostering community participation in CBS initiatives. Intention emerged as a critical mediator linking cognitive and normative beliefs to actual behavioral engagement. These findings provide actionable insights for designing targeted interventions that enhance CBS participation and strengthen infectious disease surveillance at the community level. Keywords: community participation, community-based surveillance, infectious diseases, structural equation modeling, theory of reasoned action.

In structural equation modeling (SEM) studies with categorical data, researchers often use the root mean square error of approximation (RMSEA), comparative fit index (CFI), or standardized root mean squared residual (SRMR) to compare rival models. Model selection based on , , or is meaningful because (a) finding a better model is more scientific and easier than establishing a “good” model, (b) it avoids the problems with cutoffs for fit indices, (c) one is less likely to overlook other equally substantively plausible models, and (d) information criteria (e.g., AIC, BIC) are not applicable to categorical data SEM. In this paper, we propose point estimators and confidence intervals (CIs) for , , and under categorical data. Our methods are applicable to nonnested models and do not need a true model. Simulation results show our point estimators and CIs are all trustworthy, whereas the bias is large when estimating (, ) based on the common estimators in the current literature for RMSEA (CFI, SRMR).

The present study examined the effects of organizational climate, feedback seeking environment and innovation characteristics on the implementation of a 360-degree feedback system within a large financial holdings company. The effect of user attitudes toward the 360-degree feedback system was also assessed. A sample of responses from 255 360-degree feedback system users was collected over a 14-month time period. LISREL was used to test the hypothesized structural model. Overall goodness of fit for the model was poor (using generalized least squares estimates: the Non-Normed Fit Index (NNFI) = .95, Comparative Fit Index (CFI) = .96; Root Mean Square Error of Approximation (RMSEA) = .16; using maximum likelihood estimates: NNFI = .37, CFI = .48; RMSEA = .18) thus failing to provide support for the hypothesized model. Limitations of the present study and suggestions for future research are discussed.

ObjectivesDevelop and validate short and rapid forms of the 36-item Menstrual Practice Needs Scale (MPNS-36).DesignItem reduction prioritised content validity and was informed by cognitive interviews with schoolgirls in Bangladesh, performance...

To examine the factorial validity and measurement invariance of scales that measure smoking-related self-efficacy, beliefs, and intention between daily and intermittent adolescent smokers. Cross-sectional survey design. Selected high schools in eastern Texas. Subjects were 2888 high school current smokers; mean age 16.2 years; 53.5% male; 67.4% White, 17.8% Hispanic, 6.2% Black, 8.6% "other." Daily smokers comprised 37.9% of the sample (i.e., smoked at least one cigarette every day 30 days prior to the survey). Smoking-related self-efficacy and intention were measured by three items; beliefs were measured by four items. Confirmatory factor analyses were conducted to test the invariance of the measures between daily and intermittent smokers. Fit indices included comparative fit index (CFI), nonnormed fix index (NNFI), and root mean square error of approximation (RMSEA). The three-factor measurement model had good fit for daily (CFI = .97, NNFI = .95, RMSEA = .09) and intermittent (CFI = .96, NNFI = .95, RMSEA = .09) smokers. Evidence of strong factorial invariance was found for the factors between the smoking subgroups (CFI = .96, NNFI = .96, RMSEA = .08). These measures may be used to capture and compare scores on self-efficacy to resist smoking, beliefs about benefits of smoking, and intention to smoke between daily and intermittent adolescent smokers.

AimThis study aims to clarify the construct and criterion‐related validity of the Questionnaire for Older Senior Citizens (QO) during the COVID‐19 pandemic.MethodsThis cross‐sectional study was conducted in Japan between November 11, 2020 and January 10, 2021. Of the 1645 (63.5%) older adults who responded, data from 900 participants were analyzed. First, we conducted an exploratory factor analysis (EFA) among older‐older adults (aged ≥75 years) and extracted the factors. Next, we conducted a confirmatory factor analysis (CFA) using structural equation modeling. We also conducted this analysis among younger‐older adults (aged ≥65 and <75 years) using the same model. Moreover, we compared each item of the QO with frailty status.ResultsResults of the EFA revealed six factors: social conditions and lifestyle, subjective conditions, cognitive functions, physical activity, oral functions, and physical functions. The results of the CFA were as follows: comparative fit index (CFI) = 0.971, adjusted goodness of fit index (AGFI) = 0.978, root mean square error of approximation (RMSEA) = 0.018, and standardized root mean square residual (SRMR) = 0.030. The results among the younger‐older adults were as follows: CFI = 0.880, AGFI = 0.940, RMSEA = 0.037, and SRMR = 0.048. Many QO items were significantly associated with frailty (P < 0.05).ConclusionsAmong the older‐older adults group, the model used for the QO has sufficient suitability and construct validity; among the younger‐older adults group, there also is sufficient questionnaire suitability. Moreover, the QO has criterion related validity with frailty. Geriatr Gerontol Int 2021; 21: 1018–1025.

Adolescents living with HIV experience persistent challenges in achieving optimal adherence to antiretroviral therapy (ART), threatening viral suppression and long-term treatment success. In sub-Saharan Africa, including Ethiopia, developmental transitions, limited resources, and socio-cultural factors further complicate adherence. This study aims to develop and validate a model of antiretroviral therapy (ART) adherence determinants among Ethiopian adolescents living with HIV by extending the information-motivation-behavioral skills (IMB) model through the integration of HIV-related stigma as an additional construct, given its significant influence on adherence in this population. A cross-sectional study was conducted from March to October 2024 among 219 adolescents living with HIV in ALERT hospital, Addis Ababa. The data were analyzed via SPSS and the lavaan 0.6-18 package in R. The pathways through which the study constructs influenced ART adherence were estimated via the diagonally weighted least squares (DWLS) method. Model fit was evaluated via the chi-square to degrees of freedom ratio (χ²/df), comparative fit index (CFI), Tucker-Lewis index (TLI), and root mean square error of approximation (RMSEA). A total of 210 adolescents were included in the final analysis, of whom 68.6% were adherent to ART. The proposed model demonstrated excellent fit to the data (χ²/df = 1.23, CFI = 0.99, TLI = 0.98, RMSEA = 0.033). Treatment self-efficacy mediated the effects of ART knowledge and ART attitude on ART adherence. While social support exerted both significant direct (β = 0.20, p = .029) and indirect effects via treatment self-efficacy, resulting in a significant total effect on ART adherence (β = 0.26, p = .004), HIV-related stigma did not demonstrate significant direct or indirect effects on ART adherence. Interventions that strengthen ART-related knowledge, foster positive treatment attitudes, enhance social support, and build treatment self-efficacy may substantially improve ART adherence among adolescents living with HIV.

Cancer fatalism is the belief that cancer is uncontrollable and lethal. Individuals with less education are more likely to hold fatalistic beliefs about cancer, but the mechanism accounting for the relationship is unknown. We tested whether negative health information seeking experiences explain this relationship. Structural equation modeling was used to test this relationship across three datasets from the Health Information National Trends Survey. Across all datasets, the model showed good fit: Cycle 1 (coefficient of determination = .11, comparative fit index = .96, root mean square error of approximation = .047), Cycle 2 (coefficient of determination = .06, comparative fit index = .96, root mean square error of approximation = .046), and Cycle 3 (coefficient of determination = .08, comparative fit index = .95, root mean square error of approximation = .052). The link between lower education level and higher cancer fatalism was partially mediated by negative health information seeking experiences.

Background:Early detection of breast cancer is a crucial factor in surviving the disease. This study aimed to investigate the mammography screening based on the theory of planned behavior (TPB) among rural women in Fasa and Shiraz cities, Iran.Methods:This study is a cross-sectional study performed on 800 female clients referring to rural health centers in Fasa and Shiraz cities in southern Iran in early 2021. The authors decided to send and distribute the electronic questionnaire form through the WhatsApp application in collaboration with the health staff of rural health centers for the people covered by these centers. Data gathering tools were a questionnaire on demographic characteristics, a questionnaire based on constructs of TPB, and behavior of mammography screening. Using the structural equation model (SEM), the TPB constructs and demographic variables were entered into the model. Data analysis was executed employing SPSS software version 26 and Amos version 24 (IBM Co., Ann Arbor, MI, USA). Analyzing the data was carried out using the 1-way analysis of variance (ANOVA), logistic regression, and structural equation analysis. During data analysis, various model indicators such as the goodness of fit, including comparative fit index (CFI), goodness-of-fit index (GFI), root mean square error of approximation (RMSEA), and chi-square index/df were evaluated. The significance level in all tests was considered 0.05.Results:The knowledge, attitude, and perceived behavioral control were the predictors of intention and behavior of mammography screening among the women. Among demographic variables, age, literacy, being menopausal, cancer in family, city, and ethnicity contribute more to the variance variation in TPB constructs. In this study, 7.2% of Persians, 8% of Qashqai Turks, and 4.5% of Arabs are contemplating going to mammography screening. In total, 6.8% (54 people) of all individuals intended to go mammography screening, and 5.4% (43 people) had a history of mammography screening. Goodness-of-fit indices (χ2 = 18.45, df = 10, n = 800, χ2/df = 1.845, RMSEA = 0.032, GFI = 0.90, non-normed fit index (NNFI) = 0.91) of conceptual model of this study indicate the suitability of the model.Conclusions:The results of the study indicated that the constructs of the TPB can predict mammography screening behaviors in rural women. It has also demonstrated that mammographic behavior can be improved in rural women using education based on the TPB model, emphasizing critical psychological factors of creating or changing behavior.

Measurement equivalence of the short-form Quality of Life in Childhood Epilepsy Questionnaire (QOLCE-16)

PurposeTo identify the best internal structure of the Brazilian version of the Anterior Knee Pain Scale (AKPS), comparing different instrument structures (structural validity) and correlating the scores of the versions (criterion validity).MethodsWe included Brazilian volunteers, aged ≥ 18 years, with patellofemoral pain (PFP) for at least 3 months. We used the confirmatory factor analysis and considered the following fit indices: chi-square/degrees of freedom (DF), comparative fit index (CFI), Tucker-Lewis index (TLI), root mean square error of approximation (RMSEA). We considered the structure with the lowest values of the Akaike information criterion (AIC), sample size adjusted Bayesian information criterion (SABIC), and assessed criterion validity using Pearson correlation coefficient (r) to correlate the long and short versions.ResultsThe study included 101 participants, mostly women (65.3%), young adults (~ 31 years old), overweight (BMI > 25 kg/m2), incomplete higher education (37.6%), and physically active (64.4%). The original 1-domain, 13-item structure showed adequate fit indices (chi-square/GL < 3.00, TLI and CFI > 0.90, and RMSEA < 0, 08). However, items 11 and 12 had a factorial load of less than 0.23. Therefore, we excluded items 11 and 12 and found adequate fit indices (chi-square/GL < 3.00, TLI and CFI > 0.90, and RMSEA < 0, 08) and lower AIC and SABIC values. We observed a correlation coefficient above the acceptable cutoff of 0.70 (r = 0.966, p-value < 0.001) between the versions.ConclusionThe 11-item AKPS (without items 11 and 12) is the version with the most adequate internal structure and correlates satisfactorily with the long version of the instrument.

ObjectiveDepression and anxiety often coexist in patients with end-stage-kidney disease. Recently, studies showed that a composite ‘general distress score’ which combines depression and anxiety symptoms provides a good fit in dialysis and oncology patients. We aim to investigate if the three most frequently used self-report questionnaires to measure depression and anxiety in dialysis patients are sufficiently unidimensional to warrant the use of such a general distress score in two cohorts of dialysis patients. MethodsThis study includes two prospective observational cohorts of dialysis patients (total n = 749) which measured depression and anxiety using Beck Depression Inventory (BDI), Beck Anxiety Inventory (BAI) and Hospital Anxiety and Depression Scale (HADS). Confirmatory factor analyses was used to investigate both a strictly unidimensional model and a multidimensional bifactor model that includes a general distress, depression and anxiety factor. The comparative fit index (CFI) and The Root Mean Square Error of Approximation (RMSEA) were used as model fit indices. ResultsFactor analysis did not show a good fit for a strictly unidimensional general distress factor for both the BDI/BAI and HADS (CFI 0.690 and 0.699, RMSEA 0.079 and 0.125 respectively). The multidimensional model performed better with a moderate fit for the BDI/BAI and HADS (CFI 0.873 and 0.839, RMSEA 0.052 and 0.102). ConclusionsThis data shows that the BDI/BAI and HADS are insufficiently unidimensional to warrant the use of a general distress score in dialysis patients without also investigating anxiety and depression separately. Future research is needed whether the use of a general distress score might be beneficial to identify patients in need of additional (psychological) support.