Measurement Properties Research Articles

Validation of Patient-Reported Outcomes in Patients With Nonmetastatic Breast Cancer Receiving Comprehensive Nodal Irradiation in the RadComp Trial.

Our purpose was to evaluate the measurement properties of patient-reported outcome (PRO) measures used in the ongoing RadComp pragmatic randomized clinical trial (PRCT). The deidentified and blinded data set included 774 English-speaking female participants who completed their 6-month posttreatment assessment. Eleven PRO measures were evaluated, including the Trial Outcome Index from the Functional Assessment of Cancer Therapy-Breast (FACT-B), Satisfaction with Breast Cosmetic Outcomes, the BREAST-Q, and selected Patient-Reported Outcomes Measurement Information System (PROMIS) measures. PROs were measured at 3 timepoints: baseline, completion of radiation therapy (RT), and 6 months post-RT. Ten variables were used as validity anchors. Pearson or Spearman correlations were calculated between PROs and convergent validity indicators. Mean PRO differences between clinically distinct categories were compared with analysis of variance methods (known-groups validity). PRO change scores were mapped to change in other variables (sensitivity to change). Most correlations between PROs and validity indicators were large (≥0.5). Mean score for Satisfaction with Breast Cosmetic Outcomes was higher (better) for those with a lumpectomy compared with those with a mastectomy (P < .001). Mean scores for the FACT-B Trial Outcome Index and for PROMIS Fatigue and Ability to Participate in Social Roles and Activities were better for those with good baseline performance status compared with those with poorer baseline performance status (P < .05). At completion of RT and post-RT, mean scores for Satisfaction with Breast Cosmetic Outcomes and BREAST-Q Radiation were significantly different (P < .001) across categories for all Functional Assessment of Chronic Illness Therapy -Treatment Satisfaction - General items. There were medium-sized correlations between change scores for FACT-B Trial Outcome Index, Fatigue, Anxiety, and Ability to Participate in Social Roles and change scores in the Visual Analog Scale. For patients with nonmetastatic breast cancer receiving radiation in the RadComp PRCT, our findings demonstrate high reliability and validity for important PRO measures, supporting their psychometric strength and usefulness to reflect the effect of RT on health-related quality of life.

Development of an International Standard Set of Outcomes and Measurement Methods for Routine Practice for Adults with Epilepsy: The International Consortium for Health Outcomes Measurement Consensus Recommendations.

At present, there is no internationally accepted set of core outcomes or measurement methods for epilepsy clinical practice. Therefore, the International Consortium for Health Outcomes Measurement (ICHOM) convened an international working group of experts in epilepsy, people with epilepsy and their representatives to develop minimum sets of standardized outcomes and outcomes measurement methods for clinical practice that support patient-clinician decision-making and quality improvement. Consensus methods identified 20 core outcomes. Measurement tools were recommended based on their evidence of strong clinical measurement properties, feasibility, and cross-cultural applicability. The essential outcomes included many non-seizure outcomes: anxiety, depression, suicidality, memory and attention, sleep quality, functional status, and the social impact of epilepsy. The proposed set will facilitate the implementation of the use of patient-centered outcomes in daily practice, ensuring holistic care. They also encourage harmonization of outcome measurement, and if widely implemented should reduce the heterogeneity of outcome measurement, accelerate comparative research, and facilitate quality improvement efforts.

Gaming disorder test: Assessing psychometric properties, prevalence, temporal stability, and invariance using a Czech two-time-point longitudinal sample

In 2019, Gaming Disorder (GD) was acknowledged as an official diagnosis by the World Health Organization. The Gaming Disorder Test (GDT) is the most widely used tool to measure GD; however, due to its novelty, various measurement properties are still unexplored, and the number of validated language variants is still limited. The present study is the first to assess the psychometric properties of the Czech version of the GDT. Further, it focuses on its temporal prevalence and stability, gaming genre invariance, and criterion validity. A large-scale sample of adult Czech gamers collected at two points within nine months was analysed – T1 N = 5356; T2 N = 6077; longitudinal sample N = 1430. Confirmatory factor analysis (CFA), structural equation modelling (SEM), and multigroup CFA were employed to assess the measurement invariance. The study confirmed the one-factor structure of the GDT and showed that it is invariant across preferred gaming genres and the time of data collection. It showed a negative relationship with life satisfaction and a positive relationship with anxiety, even when controlling for their mutual relationships. The prevalence in the longitudinal sample was equal to or below 1.9% in each wave, but only 0.5% in the longitudinal sample (hence n = 7 participants fulfilled in both waves the criteria for GD). The study suggests that the Czech version of the GDT has good psychometric properties, including temporal stability and invariance across gaming genres, so it is suitable for the survey type and epidemiological investigation of the ICD-11's Gaming Disorder.

Psychometric validation of the short version of the Information Needs in Cardiac Rehabilitation scale through a first global assessment.

Tailored education is recommended for cardiac patients, yet little is known about information needs in areas of the world where it is most needed. This study aims to assess (i) the measurement properties of the Information Needs in Cardiac Rehabilitation short version (INCR-S) scale and (ii) patient's information needs globally. In this cross-sectional study, English, simplified Chinese, Portuguese, or Korean versions of the INCR-S were administered to in- or out-patients via Qualtrics (January 2022-November 2023). Members of the International Council of Cardiovascular Prevention and Rehabilitation community facilitated recruitment. Importance and knowledge sufficiency of 36 items were rated. Links to evidence-based lay education were provided where warranted. A total of 1601 patients from 19 middle- and high-income countries across the world participated. Structural validity was supported upon factor analysis, with five subscales extracted: symptom response/medication, heart diseases/diagnostic tests/treatments, exercise and return-to-life roles/programmes to support, risk factors, and healthy eating/psychosocial management. Cronbach's alpha was 0.97. Construct validity was supported through significantly higher knowledge sufficiency ratings for all items and information importance ratings for all subscales in cardiac rehabilitation (CR) enrolees vs. non-enrolees (all P < 0.001). All items were rated as very important-particularly regarding cardiac events, nutrition, exercise benefits, medications, symptom response, risk factor control, and CR-but more so in high-income countries in the Americas and Western Pacific. Knowledge sufficiency ranged from 30.0 to 67.4%, varying by region and income class. Ratings were highest for medications and lowest for support groups, resistance training, and alternative medicine. Identification of information needs using the valid and reliable INCR-S can inform educational approaches to optimize patients' health outcomes across the globe.

The Bergen–Social Media Addiction Scale (BSMAS): longitudinal measurement invariance across a two-year interval

ABSTRACT Objective This study aims to examine the longitudinal measurement invariance of the Bergen Social Media Addiction Scale (BSMAS) over a two-year interval, addressing a gap in research on its consistency over time. Method Confirmatory Factor Analysis with a robust maximum likelihood chi-square estimator was utilised to assess the BSMAS among 276 adults (mean age = 31.86 years; SD = 9.94 years; 71% male) at three time points across two years. This method evaluates the scale’s structural consistency and reliability longitudinally. Results The analysis supported full measurement invariance (configural, metric, scalar, and error variance) of the BSMAS, indicating stable measurement properties over time. Furthermore, temporal stability and equivalency of the BSMAS total mean scores were confirmed across the three-time points, suggesting consistent measurement of social media addiction. Conclusions The findings validate the BSMAS as a reliable instrument for measuring social media addiction over extended periods. Demonstrating its psychometric stability enhances its utility for longitudinal studies, making it a valuable tool for tracking changes in social media addiction behaviours. These results have significant implications for future research and clinical practice, highlighting the BSMAS’s applicability in long-term studies.

The telephone-based application of the modified fatigue impact scale in individuals with multiple sclerosis: An investigation of measurement properties

BackgroundFatigue is one of the most common symptoms reported by individuals with multiple sclerosis and it contributes to the reduction of the functionality. Therefore, it is important to assess the impact of fatigue on daily living of individuals with multiple sclerosis. The Modified Fatigue Impact Scale (MFIS) is an instrument to assess the perception of the impact of fatigue. However, there have been no investigations about the measurement properties of this instrument administered over the telephone for individuals with multiple sclerosis. ObjectivesTo verify the concurrent validity, the test-retest reliability, the standard error of measurement (SEM) and the minimal detectable change (MDC) of the MFIS applied over the telephone to assess the perception of the impact of fatigue of individuals with multiple sclerosis. MethodsThe MFIS, composed of 21 items covering three domains: physical, cognitive and psychosocial, was applied at three different moments with an interval of 5–7 days. To establish the concurrent criterion validity, the face-to-face application and the first evaluation by telephone were used. The two telephone assessments were used to assess the test-retest reliability The intraclass correlation coefficient (ICC) with 95 % confidence interval (CI) and the Bland-Altman method were used. The standard error of measurement (SEM) and the MDC was calculated according to reliability results. ResultsThirty individuals (40.83 ± 10.61, 60 % female) were included. The median score on the Expanded Disability Status Scale was 2.00 (±4.00). The majority of participants experienced fatigue (n = 17; 56.67 %). A significant and high magnitude correlation (0.70 ≤ ICC ≤ 0.87, p < 0.001) was found in the investigation of concurrent criterion validity. The Bland-Altman method showed a mean difference between 0.70 to 2.17 points between face-to-face and telephone-based application of the MFIS. For test-retest reliability, a significant and very high magnitude correlation (0.91 ≤ ICC ≤ 0.97, p < 0.001) was found. The Bland-Altman method showed a mean difference between -0.03 and -0.77 points between two telephone-based applications. The SEM and the MDC were 0.71 and 1.97 points. ConclusionThe telephone-based application of the MFIS to assess the perceived impact of fatigue in individuals with multiple sclerosis demonstrated adequate measurement properties, and may be a valuable tool to assess patients in clinical practice.

Measuring Quality of Life in Deprescribing Trials: A Scoping Review.

Quality of life (QoL) is an important outcome to capture in clinical trials evaluating deprescribing interventions. We aimed to conduct a scoping review to examine how QoL has been measured in deprescribing trials among older people and identify potentially relevant QoL scales, to better inform QoL measurement in future deprescribing trials. We searched MEDLINE, Embase, PsycINFO, the Cochrane Central Register of Controlled Trials, Google Scholar, Epistemonikos, ClinicalTrials.gov, and reference lists of eligible studies (from inception to October 2023). We included randomized and non-randomized comparative studies with a control group that evaluated deprescribing and polypharmacy reduction interventions in people ≥ 65 years of age and measured QoL as an outcome. We also included studies describing the development and validation of QoL scales related to deprescribing, polypharmacy, or medication burden in adults ≥ 18 years of age. Two independent reviewers screened titles and abstracts, then full texts. Two independent reviewers extracted data from 25% of eligible studies in order to verify agreement, then a single reviewer extracted data from the remaining studies, which a second reviewer cross-checked. We critically appraised scales based on the COSMIN checklist. We retrieved 7290 articles, of which 52 were eligible for inclusion, including 44 deprescribing trials and eight scale development studies. From these studies, we found 21 scales that have been used in the context of deprescribing/polypharmacy (12 generic scales used in clinical trials and nine medication-specific scales). Variations of the generic EQ-5D were the most used scales. The measurement properties of scales for capturing changes in QoL from deprescribing were uncertain. Medication-specific QoL scales have not been employed in deprescribing clinical trials and thus, their performance in this context is also not clear. Several existing QoL scales have been applied to the context of deprescribing/polypharmacy clinical trials, and new scales specific to the problem have been proposed. If deprescribing does impact QoL, our findings suggest it is uncertain whether existing QoL scales can practically and reliably capture such a change or whether any scale is best. However, this review compares various aspects of the scales that researchers and clinicians can consider in decisions about measuring QoL in deprescribing trials, and in planning future research. Open Science Framework: osf.io/aez6w.

The Knowledge of Autism Questionnaire-UK: Development and Initial Psychometric Evaluation.

Tools to measure autism knowledge are needed to assess levels of understanding within particular groups of people and to evaluate whether awareness-raising campaigns or interventions lead to improvements in understanding. Several such measures are in circulation, but, to our knowledge, there are no psychometrically-validated questionnaires that assess contemporary autism knowledge suitable to the UK context. We aimed to produce a brief measure to assess between-respondent variability and within-respondent change over time. A pool of questionnaire items was developed and refined through a multi-stage iterative process involving autism experts and a lay sample. Attention was paid to face validity, clarity, consensus on correct responses, and appropriate difficulty levels. Initial validation data was obtained from a lay sample of 201 people. Difficulty and discrimination ability were assessed using item response theory and low-performing items were removed. Dimensionality was evaluated with exploratory factor analysis, which revealed a one-factor structure of the questionnaire. Further items were removed where they did not load strongly on their main factor. This process resulted in a final 14-item questionnaire called the Knowledge of Autism Questionnaire-UK. Internal consistency was satisfactory, and the final questionnaire was able to distinguish between parents of autistic people and those without an affiliation to autism. The KAQ-UK is a new, freely-available measure of autism knowledge that could be used to assess between-respondent variability and within-respondent change over time. Further evaluation and validation of its measurement properties are required.

Cancer-Specific Health Utilities: Evaluation of Core Measurement Properties of the EORTC QLU-C10D in Lung Cancer Patients-Data from Four Multicentre LUX-Lung Trials, Applying Six Country Tariffs.

Cost-utility analysis generally requires valid preference-based measures (PBMs) to assess the utility of patient health. While generic PBMs are widely used, disease-specific PBMs may capture additional aspects of health relevant for certain patient populations. This study investigates the construct and concurrent criterion validity of the cancer-specific European Organisation for Research and Treatment of Cancer (EORTC) Quality of Life Utility-Core 10 dimensions (QLU-C10D) in non-small-cell lung cancer patients. We retrospectively analysed data from four multicentre LUX-Lung trials, all of which had administered the EORTC Quality of Life Questionnaire (QLQ-C30) and the EQ-5D-3L. We applied six country-specific value sets (Australia, Canada, Italy, the Netherlands, Poland, and the United Kingdom) to both instruments. Criterion validity was assessed via correlations between the instruments' utility scores. Correlations of divergent and convergent domains and Bland-Altman plots investigated construct validity. Floor and ceiling effects were assessed. The comparison of the EORTC QLU-C10D and EQ-5D-3L produced homogenous results for five of the six country tariffs. High correlations of utilities (r > 0.7) were found for all country tariffs except for the Netherlands. Moderate to high correlations of converging domain pairs (r from 0.472 to 0.718) were found with few exceptions, such as the Social Functioning-Usual Activities domain pair (max. r = 0.376). For all but the Dutch tariff, the EORTC QLU-C10D produced consistently lower utility values compared to the EQ-5D-3L (x̄ difference from -0.082 to 0.033). Floor and ceiling effects were consistently lower for the EORTC QLU-C10D (max. 4.67% for utilities). The six country tariffs showed good psychometric properties for the EORTC QLU-C10D in lung cancer patients. Criterion and construct validity was established. The QLU-C10D showed superior measurement precision towards the upper and lower end of the scale compared to the EQ-5D-3L, which is important when cost-utility analysis seeks to measure health change across the severity spectrum.

Development and evaluation of a food literacy questionnaire for schoolchildren in France

Food literacy is gaining importance in nutrition education programs for children. To date, food literacy assessment tools have been developed in many countries, however, none exist in France. The objectives of this study were to develop a questionnaire and to evaluate its measurement properties among French schoolchildren aged 8 to 11 years. The questionnaire was developed in three phases: i) item selection (literature review and adaptation or creation of items) and content validity (submission to an expert panel), ii) questionnaire development including a pre-test in a small sample of children (n=41) and item reduction and dimensionality based on the responses of children who completed the questionnaire in 31 schools between December 2022 and March 2023, and iii) questionnaire evaluation in terms of reliability, validity and acceptability. In total, 1,187 responses were included in the analysis. The mean age of the children was 9.6±0.7 years (girls: 51.2%, boys: 48.8%). The development process resulted in a 25-item questionnaire with good acceptability and satisfactory estimated reliability (McDonald omega coefficient=0.73). Factor evaluation revealed a three-dimensional structure encompassing food and nutrition knowledge, participation in food preparation activities and food habits. To our knowledge, this study was the first to assess food literacy for schoolchildren in France. Our questionnaire can contribute to assess the factors that make food literacy vary, especially regarding socioeconomic variables to target priority populations for nutrition education actions and to describe changes in food literacy scores from a longitudinal perspective.

Translation, cross-cultural adaptation, and measurement properties of the dysmenorrhea symptom interference (DSI) scale–Brazilian version

BackgroundThe Dysmenorrhea Symptom Interference (DSI) scale is a reliable, valid, and responsive tool to assess the interference of menstrual pain in the physical, mental, and social activities of women. ObjectiveTo translate and cross-culturally adapt the DSI scale into Brazilian-Portuguese (DSI-BrPt) and investigate the measurement properties of this version in on- and off-menses versions. MethodsThe original (United States) scale was translated and culturally adapted following existing guidelines. Measurement properties of the DSI-BrPt were investigated in 1387 women with dysmenorrhea. Reliability was analyzed via intraclass correlation coefficients (ICC) and test-retest reliability. Furthermore, structural validity, internal consistency (Cronbach's alpha), cross-cultural validity, construct validity (correlation with WHODAS 2.0 and SPS-6 scores questionnaires), and floor and ceiling effects were determined. ResultsNo significant adaptations were needed during the translation process of the DSI-BrPt. The values of Cronbach's α were adequate (α ≥0.87) for the unidimensional scale. The test-retest reliability was considered adequate (ICC >0.78) and there was no systematic error for both on-menses and off-menses versions. Moreover, the DSI had a positive and strong correlation with WHODAS 2.0. There were no floor and ceiling effects neither for the total sample, nor among off-menses, or on-menses women. ConclusionThe DSI-BrPt scale has been translated and cross-culturally adapted successfully. The DSI-BrPt scale presented adequate measurement properties. The scale is valid and reliable, and, therefore, an adequate tool for monitoring dysmenorrhea symptoms in Brazilian women during and between menses.

Stability Properties of Feature Selection Measures

Stability Properties of Feature Selection Measures

Connor-Davidson Resilience Scale: a systematic review psychometrics properties using the COSMIN.

Psychometrical evaluation of persons of diverse contexts and different populations, including general or clinical. This review study aimed to evaluate the psychometrics quality of resilience scales. International and Iranian databases were searched with MESH terms, including "psychometric", "validity", "reliability", "Connor-Davidson resilience scale", "Resilience scale", for published articles up to 1 February 2023. For each of the selected studies, the risk of bias was evaluated using the COSMIN Risk of Bias Checklist. Then the COSMIN checklist was used to evaluate the entire text of the article for methodological quality. Considering the inclusion criteria, 80 documents were evaluated. According to the COSMIN's criteria for evaluating the risk of bias, the current study findings revealed the included studies' limitations in assessing the three versions of CD-RISC cross-cultural and content validity as well as their stability (e.g. conducting test re-test), whereas the majority of psychometric studies of CD-RISC-25, and CD-RISC-2 rated as very good or adequate in terms of structural validity. In terms of quality assessment of the included studies, the current study indicated that investigating the structural validity of the CD-RISC was mainly done based on exploratory factor analysis (EFA), and confirmatory factor analysis was absent. The general result indicates the acceptability of the quality of the studies. However, concerns for measurement properties such as responsiveness and criterion validity as well as the standard error of measurement have been neglected.

Measurement properties of the Activities-specific Balance Confidence Scale in adults from the general population with concussion: a report from the Toronto Concussion Study

ObjectiveThe aim of this study was to establish the internal consistency and construct validity of the Activities-specific Balance Confidence (ABC) Scale and ABC-6 in adults from the general population with concussion. DesignProspective analysis. SettingOutpatient concussion care clinic. ParticipantsAdults from the general population with concussion referred to a concussion care clinic within 7 days of injury (n=511). InterventionsNot applicable. Main outcome measuresBalance confidence was assessed with the Activities-specific Balance Confidence Scale (ABC) and the ABC-6. Concussion symptoms were characterized using the Sport Concussion Assessment Tool (SCAT5) symptom checklist. Instrumented measures of balance and gait included centre of pressure (COP) velocity and double support time, respectively. Balance was also assessed using the mBESS (modified Balance Error Scoring System). ResultsThe ABC and ABC-6 were strongly correlated (ρ=0.980, p<0.001). Cronbach's α for ABC and ABC-6 were 0.966 and 0.940, respectively. Factor analysis verified the existence of 2 components of the ABC, one including all items of the ABC-6 as well as 3 additional items. ABC and ABC-6 were moderately significantly correlated with SCAT5 symptom number, severity, and symptom domain (ρ=-0.350 to -0.604). However, correlations between ABC and ABC-6 with instrumented measures of balance and gait were not statistically significant, except for of double support time during dual task gait with ABC-6 (ρ=-0.218). ConclusionsIn community dwelling adults with concussion, the ABC and ABC-6 have good internal consistency. Convergent validity is stronger for symptom endorsement measures within SCAT5 domains, which has a similar construct (subjectivity) to balance confidence. Both the ABC and ABC-6 are valid measures of balance self-efficacy in adults from the general population with concussion. The ABC-6 may be a useful tool for characterizing the impact of concussion on perceptions of the ability to perform functional tasks that challenge balance and mobility.

Concentration Fluctuation/Microheterogeneity Duality Illustrated with Aqueous 1,4-Dioxane Mixtures.

The structural properties of aqueous 1-4 dioxane mixtures are studied by computer simulations of different water and dioxane force field models, from the perspective of illustrating the link between structural properties at the molecular level and measurable properties such as radiation scattering intensities and Kirkwood-Buff integrals (KBIs). A strategy to consistently correct the KBI obtained from simulations is proposed, which allows us to obtain the genuine KBI corresponding to a given pair of molecular species, in the entire concentration range, and without necessitating excessively large system sizes. The application of this method to the aqueous dioxane mixtures, with an all-atom CHARMM dioxane model and 2 water models, namely, SPC/E and TIP3P, allows one to understand the differences in the structure of the corresponding mixtures at the molecular level, particularly concerning the role of the water aggregates and its model dependence. This study allows us to characterize the dual role played by the concentration fluctuations and the domain segregation, particularly in what concerns the calculated X-ray spectra.

Recommended measurement instruments for menopausal vasomotor symptoms: the COMMA (Core Outcomes in Menopause) consortium.

The aim of the study is to identify suitable definitions and patient-reported outcome measures (PROMs) to assess each of the six core outcomes previously identified through the COMMA (Core Outcomes in Menopause) global consensus process relating to vasomotor symptoms: frequency, severity, distress/bother/interference, impact on sleep, satisfaction with treatment, and side effects. A systematic review was conducted to identify relevant definitions for the outcome of side-effects and PROMs with acceptable measurement properties for the remaining five core outcomes. The consensus process, involving 36 participants from 16 countries, was conducted to review definitions and PROMs and make final recommendations for the measurement of each core outcome. A total of 21,207 publications were screened from which 119 reporting on 40 PROMs were identified. Of these 40 PROMs, 36 either did not adequately map onto the core outcomes or lacked sufficient measurement properties. Therefore, only four PROMs corresponding to two of the six core outcomes were considered for recommendation. We recommend the Hot Flash Related Daily Interference Scale to measure the domain of distress, bother, or interference of vasomotor symptoms and to capture impact on sleep (one item in the Hot Flash Related Daily Interference Scale captures interference with sleep). Six definitions of "side effects" were identified and considered. We recommend that all trials report adverse events, which is a requirement of Good Clinical Practice. We identified suitable definitions and PROMs for only three of the six core outcomes. No suitable PROMs were found for the remaining three outcomes (frequency and severity of vasomotor symptoms and satisfaction with treatment). Future studies should develop and validate PROMs for these outcomes.

Reliability and Validity of Maximal Respiratory Pressures.

Maximal respiratory pressure is used to assess the inspiratory and expiratory muscles strength by using maximal inspiratory pressure (PImax ) and maximal expiratory pressure (PEmax). This study aimed to summarize and evaluate the reliability and validity of maximal respiratory pressure measurements. This systematic review followed the Consensus-based Standards for the Selection of Health Measurement Instruments (COSMIN) recommendations and was reported by using the PRISMA checklist. Studies published before March 2023 were searched in PubMed and EMBASE databases. A total of 642 studies were identified by using the online search strategy and manual search (602 and 40, respectively). Twenty-three studies were included. The level of evidence for test-retest reliability was moderate for PImax and PEmax (intraclass correlation coefficient > 0.70 for both), inter-rater reliability was low for PImax and very low for PEmax (intraclass correlation coefficient > 0.70 for both), and the measurement error was very low for PImax and PEmax. In addition, concurrent validity presented a high level of evidence for PImax and PEmax (r > 0.80). Only concurrent validity of maximal respiratory pressure measured with the manometers evaluated in this review presented a high level of evidence. The quality of clinical studies by using maximal respiratory pressure would be improved if more high-quality studies on measurement properties, by following well established guidelines and the COSMIN initiative, were available.

Three Approaches to Improve Inferences Based on Survey Data Collected with Mixed-mode Designs

Abstract Mixed-mode designs have become increasingly common in survey data collection. Although different modes often have different measurement properties, the standard practice is to treat mixed-mode data as if they had been collected with a single mode, neglecting the potential impact of mode effects. To account for potential mode effects when making inferences for mixed-mode samples, we propose (i) a Testimator approach, (ii) a Bayesian approach, and (iii) a model averaging approach. In the Testimator approach, we test whether the means and the variances of mixed-mode samples are the same. If the means are the same, we take the average of mode-specific estimates. If the means are different, we take the average when we have no prior information about preferred modes and take the smaller (or larger) estimate when we have prior information about preferred modes (e.g., a smaller estimate is better). In the Bayesian approach, we assume some prior information. We use a data-driven method to determine whether there are mode effects. If there are no mode effects, we draw inferences using a common mean model. If there are mode effects, we draw inferences using the data collected with the mode that produces smaller estimates. In the model averaging approach, we combine estimates of different models (characterized by whether assume same means and variances across modes) using marginal posteriors as weights. We evaluate the approaches in simulation studies and find that they achieve robust inferences compared to the standard approach. We apply the methods to the Arab Barometer study, which employs a randomized mixed-mode design.

Some properties of new general fractal measures

Some properties of new general fractal measures

The Pre-Exposure Prophylaxis Stigma Scale: Measurement Properties of an Adaptation in German and French

Objectives:This study aimed to adapt and validate the HIV PrEP Stigma Scale (HPSS) in French and German languages (HPSS-FR/DE) and assess its applicability across diverse linguistic and cultural contexts.Methods:The original scale was adapted to French and German and administered through an online survey in multiple European nations. A four-factor structure was extracted from the data, including negative social consequences, social pressure, self-support, and external support. The scale’s construct validity, reliability, and cross-linguistic consistency were evaluated.Results:The adapted HPSS-FR/DE demonstrated robust psychometric properties, good construct validity, acceptable reliability, and consistent measurement across different languages. This adaptation enhances its utility in multicultural settings, offering a comprehensive tool to assess PrEP-related stigma.Conclusion:This study provides a suitable tool to address PrEP stigma in multicultural environments to enhance PrEP uptake and adherence among men who have sex with men. Moreover, it lays the groundwork for further investigations into PrEP stigma across diverse populations and cultural settings, enabling the development of targeted public health interventions and policies to combat this issue effectively.