Strict Unidimensionality Research Articles

Rasch analysis of the forgotten joint score in patients with total hip arthroplasty.

To assess the internal construct validity, including local independence, unidimensionality, monotonicity, and invariance, reliability, and targeting of the Forgotten Joint Score within the Rasch Measurement Theory framework. Cross-sectional study. A total of 111 patients with total hip arthroplasty at least 3 months after surgery. The Forgotten Joint Score was submitted to each subject during their rehabilitative treatment in an Italian centre and then to Rasch analysis. The base Rasch analysis showed a satisfactory fit to the model with strict unidimensionality and no differential item functioning. However, monotonicity (11 out of 12 items showed disordered thresholds) and local independence were violated. After rescoring 10 items and creating 5 subtests to account for local dependence, the scale satisfied all the other Rasch model requirements (i.e. invariance, local independence, monotonicity, unidimensionality, and multi-group invariance), with reliability indexes (> 0.850) for measurement at the individual level and proper targeting. A raw-score-to-measure conversion table was provided. After structural (i.e. collapsing items categories) and non-structural (i.e. creating subtests) strategies, the Forgotten Joint Score satisfied the measurement requirements of the Rasch model, and it can be used in patients with total hip arthroplasty in clinical and research settings.

Enhancing precision of the Telephone Interview for Cognitive Status–Modified (TICS-M) using the Rasch model.

The Telephone Interview for Cognitive Status-modified (TICS-M) is a well-established and widely used screening instrument for dementia and assessment of global cognitive function in older people. This study aimed to evaluate the psychometric properties of the TICS-M and to enhance the accuracy of the instrument using Rasch methodology. Partial Credit Rasch model was applied to the TICS-M scores. The sample selected for Rasch analysis consisted of 432 participants aged 70-90 years (M = 78.85, SD = 4.73) including 195 males (237 females), and 132 (30.56%) of whom were diagnosed with dementia after the baseline assessment. Initial analysis indicated good reliability of the TICS-M assessment scores, but there were three misfitting items and local dependency issues. Combining locally dependent and misfitting items into super-items achieved the best Rasch model fit for the TICS-M. This modification improved reliability of the assessment scores and resulted in no misfitting items, no local dependency, strict unidimensionality, and invariance across individual factors such as participants age, sex, diagnosis, and in-person neuropsychological assessment scores. Satisfying Rasch model expectations allowed for creation of a transformation table to convert raw TICS-M scores into interval-level data, which improves precision of the instrument. In summary, the TICS-M assessment scores demonstrated excellent reliability as reflected by Person Separation Index (PSI = 0.86) and met expectations of the unidimensional Rasch model after minor adjustments. The ordinal-to-interval transformation table can be used to increase accuracy of the TICS-M without altering its current format. These findings contribute to more accurate assessments of cognitive decline in older people and screening for conditions such as dementia. (PsycInfo Database Record (c) 2023 APA, all rights reserved).

Measuring distress in older population: Rasch analysis of the Kessler Psychological Distress Scale

ObjectivesThe 10-item Kessler Psychological Distress Scale (K−10) is a widely applied distress measure; however, its psychometric properties were not established with older populations using advanced methodology. The aim of this study was to examine psychometric properties of the K-10 through application of Rasch methodology and if possible, develop an ordinal-to-interval conversion to improve its reliability in older populations. MethodThe Partial Credit Rasch Model was applied to analyse K-10 scores of the sample including 490 participants (56.3 % females) aged 70 to 90 years and without dementia from the Sydney Memory and Ageing Study (MAS). ResultsThe initial analysis of the K-10 showed poor reliability and significant deviation from the expectations of the Rasch model. The best model fit was evident after correcting disordered thresholds and creating two testlet models to address local dependency between items (χ2(35) = 29.87, p = 0.71). The modified K-10 demonstrated strict unidimensionality, enhanced reliability and scale invariance across personal factors, such as sex, age, and education and permitted development of ordinal-to-interval transformation algorithms. LimitationsOrdinal-to-interval conversion can only be applied for older adults with complete data. ConclusionsThe K-10 satisfied principles of fundamental measurement defined by Rasch model after minor modifications. Clinicians and researchers can transform K-10 raw scores into interval-level data using converging algorithms published here without altering the original scale response format, which increases reliability of the K-10.

Enhancing the Assessment of Gratitude in Mindfulness Research: a Rasch Analysis of the 6-Item Gratitude Questionnaire

ObjectivesThe 6-item gratitude questionnaire (GQ-6) is a widely used measure reported to be reliable and valid under traditional and Rasch investigations. However, recent investigations were inconsistent regarding the item structure of the GQ-6, with three investigations concluding that item 6 be removed. Previous Rasch analyses also did not produce interval conversion tables, a key benefit of this method which provides a means of improving scale accuracy and aligning the scale with the fundamental principles of measurement.MethodsA Partial Credit Rasch Model was used to evaluate psychometric properties of the GQ-6 using a combined sample of 663 respondents from the USA (n = 345) and India (n = 318).ResultsInitial analysis showed significant scale dysfunction, with overall and individual item misfit, local dependency, disordered response thresholds, instances of differential item functioning by age and nationality, along with poor reliability. Through the use of recent advances in the Rasch methodology, locally dependent items were combined into two super-items and the best fit to the Rasch model was obtained with evidence of strict unidimensionality, invariance, and strong reliability. Results indicate the GQ-6 is suitable for individual and group assessment, while also permitting the creation of an ordinal-to-interval conversion algorithm which has been provided here.ConclusionsThis study confirmed the robust psychometric properties of the GQ-6 after minor modifications and provides a means for clinicians and researchers to improve the accuracy of this widely used measure in mindfulness research and other relevant studies without modification of its original response format.

Validation of the Social Media Disorder Scale using network analysis in a large representative sample of Czech adolescents.

BackgroundThe importance of studying the excessive use of social media in adolescents is increasing and so is the need for in-depth evaluations of the psychometric properties of the measurement tools. This study investigated the properties of the Social Media Disorder Scale (SMDS) in a large representative sample of Czech adolescents.MethodsWe analyzed the representative sample of 13,377 Czech adolescents (50.9% boys), 11–16 years old, who participated in the Health Behavior in School-aged Children (HBSC) survey (2017–18), using confirmatory factor analysis (CFA) and network models. Furthermore, we evaluated the measurement invariance and constructed the validity of the SMDS.ResultsWe found support for a single dominant factor but not for strict unidimensionality. Several residual correlations were identified. The strongest were for: problems–conflicts–deceptions; persistence–escape; and preoccupation–tolerance–withdrawal. Girls, particularly 13- and 15-year-olds, scored higher than boys in the same age group, and 13- and 15-year-olds achieved higher scores than 11-year-olds, although some items were not invariant between the groups. The SMDS was positively related to other online activities, screen time, and falling asleep late, but negatively related to well-being and mental health.Discussion and conclusionsThe SMDS showed solid psychometric properties and construct validity. However, small violations of measurement invariance were detected. Furthermore, the network analysis showed important residual relationships between the items.

Enhancing precision of the 16-item Informant Questionnaire on Cognitive Decline in the Elderly (IQCODE-16) using Rasch methodology.

This study aimed to investigate psychometric properties and enhance precision of the 16-item Informant Questionnaire on Cognitive Decline in the Elderly (IQCODE-16) up to interval-level scale using Rasch methodology. Partial Credit Rasch model was applied to the IQCODE-16 scores using longitudinal data spanning 10 years of biennial follow-up. Community-dwelling older adults aged 70-90 years and their informants, living in Sydney, Australia, participated in the longitudinal Sydney Memory and Ageing Study (MAS). The sample included 400 participants of the MAS aged 70 years and older, 109 out of those were diagnosed with dementia 10 years after the baseline assessment. The IQCODE-16. Initial analysis indicated excellent reliability of the IQCODE-16, Person Separation Index (PSI) = 0.92, but there were four misfitting items and local dependency issues. Combining locally dependent items into four super-items resulted in the best Rasch model fit with no misfitting or locally dependent items, strict unidimensionality, strong reliability, and invariance across person factors such as participants' diagnosis and relationship to their informants, as well as informants' age and sex. This permitted the generation of conversion algorithms to transform ordinal scores into interval data to enhance precision of measurement. The IQCODE-16 demonstrated strong reliability and satisfied expectations of the unidimensional Rasch model after minor modifications. Ordinal-to-interval transformation tables published here can be used to increase accuracy of the IQCODE-16 without altering its current format. These findings could contribute to enhancement of precision in assessing clinical conditions such as cognitive decline in older people.

Enhancing the precision of the Positive and Negative Affect Schedule (PANAS) using Rasch analysis

The Positive and Negative Affect Schedule (PANAS; Watson, Clark, & Tellegan, Journal of Personality and Social Psychology, 54(6), 1063–1070, 1988) is a widely used affect measure but it has the limitations of an ordinal scale such as low precision and unsuitability for use with parametric statistics. Rasch analysis is an appropriate methodology used to investigate and enhance the precision of ordinal scales through transformation from ordinal to interval scores given the data fits the model. This study applied the Partial Credit Rasch model to the adequate sample of 396 participants. Positive and Negative Affect subscales were analysed independently and local dependence was identified between several items in each subscale that affected both model fit and reliability. The best Rasch model fit was achieved for each subscale after applying recent methodological advances that involved combining locally dependent items into super-items resulting in strong reliability (PSI > 0.80) and strict unidimensionality. This permitted the development of ordinal-to-interval conversion tables. The results support robust psychometric properties of the PANAS. Using conversion tables published here researchers can improve the precision of PANAS subscales to satisfy assumptions of parametric statistics and fundamental measurement principles.

Sexual Morbidity Assessment in Gyne-Oncology Follow-Up: Development of the Sexual Well-Being After Cervical or Endometrial Cancer (SWELL-CE) Patient-Reported Outcome Measure

BackgroundClinical assessment and management of sexual difficulties after gynecological cancer remain a neglected aspect of women’s rehabilitation. AimTo develop and validate a patient-reported outcome measure of sexual well-being for women experiencing sexual consequences of cervical and endometrial cancer treatment for use in routine follow-up. MethodsThis is a sequential mixed method study comprising (i) in-depth qualitative interviews (n=21 of 118) to generate items regarding sexual consequences of cervical or endometrial cancer and treatment; (ii) questionnaire construction with 51 core items (all respondents) and 4 subsections (18–58 items), depending on the relationship status and whether or not participants were sexually active (SA/NSA); (iii) item refinement following cognitive debriefing (n=13 of 21); (iv) validation of resultant items via postal survey (n=788 women) and Rasch analysis; and (v) creation of brief (14-item) clinical screener. Women attending routine follow-up (3 months to 5 years) at 6 English cancer centers and members of 3 UK cancer patient websites, who met the study inclusion criteria, were invited to participate. OutcomesThe primary outcome of this study was the construction and initial psychometric testing of SWELL-CE short and long form versions. Results21 women participated in interviews and 250 of 788 (32%) returned the postal survey (T1). 110 draft items were evaluated using cognitive testing (n=13) to refine instrument design and test face validity, comprehension, and acceptability. Exploratory factor analysis of survey data (n=250) produced an initial 6 domain structure as a guidance for the Rasch analysis. Subsequent Rasch analysis yielded a 3 domain structure: physical sexual function, sexual and relationship concerns, and sexual desire and sexual self-esteem, each satisfying Rasch model requirements within their respective SA (item pool =59) and NSA (item pool =53) categories, including the absence of local response dependency and all showing strict unidimensionality. The 3 subscales demonstrated good psychometric properties, external validity, and test-retest reliability. A valid Rasch short form of 14 items was created from the larger item pool. Clinical ImplicationsThis PROM may assist clinicians to improve identification, discussion, and management of women who could benefit from sexual rehabilitation. Strengths & LimitationsInitial evaluation supports psychometric validity and reliability in the assessment of physical sexual function, sexual interest and sexual self-esteem, and sexual and relationship concerns in this study sample. However, given this study’s modest response rate (32%, n=250), findings should be interpreted with caution. This PROM identifies sexual concerns in women who are sexually active or sexually non-active due to illness or treatment-associated sexual difficulties. ConclusionSexual Well-being after Cervical or Endometrial Cancer is a novel and psychometrically valid sexual well-being measure for clinical assessment of female sexual difficulties after cervical or endometrial cancer treatment.White ID, Tennant A, Taylor C, Sexual Morbidity Assessment in Gyne-Oncology Follow-Up: Development of the Sexual Well-Being After Cervical or Endometrial Cancer (SWELL-CE) Patient-Reported Outcome Measure. J Sex Med 2020;17:2005–2015.

New avenue for the geriatric depression scale: Rasch transformation enhances reliability of assessment

BackgroundDepression is a common problem in older adults. The 15-item Geriatric Depression Scale (GDS-15) is a widely used psychometric tool for measuring depression in the elderly, but its psychometric properties have not been yet rigorously investigated. The aim was to evaluate psychometric properties of the GDS-15 and improve precision of the instrument by applying Rasch analysis and deriving conversion tables for transformation of raw scores into interval level data. MethodsThe data was extracted from the prospective cohort Sydney Memory and Ageing Study of initially not demented individuals aged 70 years and older. The GDS-15 items scores of 212 participants (47.2% males) were analysed using the dichotomous Rasch model. ResultsInitially poor reliability of the GDS-15, Person Separation Index (PSI) = 0.68, was improved by combining locally dependent items into seven super-items. These modifications improved reliability of the GDS-15 (PSI = 0.78) and resulted in the best Rasch model fit (χ2(28)=37.72, p = =0.104), strict unidimensionality and scale invariance across personal factors such as gender, diagnostic and language background. LimitationsPresence of participants with cognitive impairment may be a potential limitation. ConclusionsReliability and psychometric characteristics of the GDS-15 were improved by minor modifications and now satisfy expectations of the unidimensional Rasch model. By using Rasch transformation tables published here psychiatrists, psychologists and researchers can transform GDS raw scores into interval-level data, which improves reliability of the GDS-15 without the need to modify its original response format. These findings increase accuracy of clinical psychometric assessments, leading to more precise diagnosis of depression in the elderly.

Development and Validation of an Item Bank for Depression Screening in the Chinese Population Using Computer Adaptive Testing: A Simulation Study.

With the increasing prevalence of depression, creating a simple and precise tool for measuring depression is becoming more important. This study developed a computer adaptive testing for depression (CAT-Depression) from a Chinese sample. The depression item bank was constructed from a sample of 1,135 participants with or without depression using the Graded Response Model (GRM; Samejima, 1969). The final depression item bank with strict unidimensionality comprised 68 items, which had local independence, good item-fit, high discrimination, no differential item functioning (DIF), and each item measured at least one symptom of diagnostic criteria for depression in ICD-10. In addition, the mean IRT discrimination of the item bank reached 1.784, which clearly showed that the item bank of CAT-Depression was high-quality. Moreover, a simulation CAT study with real response data was conducted to investigate the characteristics, marginal reliability, criterion-related validity, and predictive utility (sensitivity and specificity) of CAT-Depression. The results revealed that the proposed CAT-Depression had acceptable and reasonable marginal reliability, criterion-related validity, and sensitivity and specificity.

The educational needs of patients with undifferentiated spondyloarthritis: Validation of the ENAT questionnaire and needs assessment.

The aim of the present study was to validate the Swedish version of the educational needs assessment tool (SwENAT) in undifferentiated spondyloarthritis (USpA) and use it to study the educational needs of patients with USpA. This was a cross-sectional study, recruiting a random sample of patients with USpA from a hospital register in Sweden. Educational needs data were collected, together with disease activity and function indices (Bath Ankylosing Spondylitis Disease Activity Index [BASDAI] and Bath Ankylosing Spondylitis Functional Index [BASFI]). Rasch analysis was utilized to evaluate the construct validity, internal consistency and unidimensionality of the SwENAT before studying differences in educational needs between patient subgroups (gender, age and disease severity). Complete responses were obtained from 77 patients (48 women), with a mean age (standard deviation [SD]) of 50 (12) years, a disease duration of 16 (11) years, a BASDAI score of 4.9 (1.9) and a BASFI score of 3.1 (2.3). The SwENAT satisfied the requirements of the Rasch model (χ2 = 11.488; p = 0.119), including strict unidimensionality. Overall, the mean (SD) SwENAT score was 86 (32). Women reported higher needs than men in the domains of pain (mean [SD] 13.1 [6.8] versus 10.1 [6.0]; p = 0.05); movement (mean [SD] 13.0 [5.5] versus 9.9 [5.7]; p = 0.02) and self-help (mean [SD] 17.0 [5.8] versus 14.1 [5.0]; p = 0.03). Higher disease activity (BASDAI >4) was associated with higher educational needs (mean [SD] 92.6 [31.9] versus 73.7 [29.4]; p = 0.02). These data suggest that the SwENAT is valid in USpA. Women and patients with higher disease activity are more likely to have high levels of educational needs, so special attention and strategies to target patient education are warranted.

General, Health-Specific, and Housing-Specific Self-Efficacy Scales: Preliminary Reliability and Validity Evidence with Homeless or Vulnerably Housed Adults

Self-efficacy (SE) refers to one’s sense of personal competence and is a key element of human agency. Among individuals who are homeless or vulnerably housed, SE has the potential to provide important information about an individual’s ability to seek out and make use of resources and persevere in the face of multiple challenges. SE is understudied as a personal variable in research with homeless samples. Thus, it is important to identify appropriate measures of general and domain-specific SE for this population and evaluate their psychometric properties. Three relevant SE scales are the Generalized Self Efficacy Scale, Perceived Health Competence Scale, and Housing Self-Efficacy Scale. The purposes of this study were to examine the internal structure and score reliability for each of these SE scales, report performance on and intercorrelations among the SE scales, and examine the relationship of demographic variables to SE scale scores, with a sample of adults who were homeless or vulnerably housed. Strict unidimensionality was evaluated using confirmatory factor analysis. Model fit was examined using fit indices and residual polychoric correlation matrices. Essential unidimensionality was evaluated using exploratory factor analysis and the ratio of first to second eigenvalues. Internal consistency estimates of reliability were obtained using ordinal alpha. None of the SE measures were found to be strictly unidimensional but all three measures were found to be essentially unidimensional. This finding supported the use of total scores for each measure. Ordinal alphas ranged from .87 to .93 for the three SE measures and thus were satisfactory. Correlations among the three measures ranged from .36 to .45. Demographic variables showed little relationship to the three SE measures. The study findings provided initial psychometric evidence to support the use of these three SE measures with a homeless or vulnerably housed adult sample.

Measuring Physical and Cognitive Fatigue in People With Post-Polio Syndrome: Development of the Neurological Fatigue Index for Post-Polio Syndrome (NFI-PP)

BackgroundFatigue in post-polio syndrome (PPS) has been shown to affect quality of life adversely. There is currently no disease-specific measure of fatigue for PPS. ObjectiveTo develop a scale to measure fatigue in PPS that meets rigorous psychometric standards. DesignQualitative followed by validation and test-retest studies. SettingPolio clinic followed by national questionnaire studies. ParticipantsA total of 45 participants from polio clinic for qualitative; 319 participants from clinic or self-referral for validation study, of whom 87 completed the retest questionnaire. MethodsDraft questionnaire items on PPS fatigue were derived from transcripts of qualitative interviews. After cognitive debriefing, the draft measure was administered by mail along with comparator questionnaires to a new sample. Main Outcome MeasurementsDraft PPS fatigue measure, Fatigue Severity Scale, and visual analog scale for fatigue. ResultsAnalysis of 271 of 319 (85%) questionnaires identified a 2-factor solution (RMSEA 0.058). For the physical subscale, a 20-item scale showed good fit (χ2P = .189), strict unidimensionality (t-test 5.17%), and reliability 0.91. For the cognitive subscale, a 7-item scale showed excellent fit (χ2P = .917), strict unidimensionality (t-test 5.2%), and reliability 0.89. Evidence of a “difficulty factor” emerged also supporting a total score that showed good fit (χ2P = .151), strict unidimensionality (t-test 0.4%), and reliability consistent with group use at 0.73. Test-retest correlations for all scales were greater than 0.85. Standard error of measurement on metric ranges was 5.4 for total, 2.9 for physical, and 1.69 for cognitive domains. With the latent estimate of the total score transformed to a 0-100 scale, the mean score was 49.5 (SD 6.9). Spearman correlations with the Fatigue Severity Scale and visual analog scale were 0.60 and 0.55, respectively. ConclusionsA patient-derived Neurological Fatigue Index for PPS, with physical and cognitive subscales and a total score, has demonstrated good reliability, appropriate concurrent validity, and satisfies the Rasch measurement model. A raw-score to interval scale transformation is available for parametric applications and the calculation of change scores. Level of EvidenceIII

The Manual Work Instability Scale: development and validation

Increasing awareness of the burden of absenteeism and reduced performance at work highlights the importance of early identification of individuals experiencing work instability (WI), a mismatch between functional and cognitive abilities and job demands. To develop and validate a screening questionnaire to measure WI in manual workers. Questionnaire items were generated via qualitative interviews with manual workers and a draft survey instrument was completed by workers in a variety of fields. Rasch analysis was used interactively to assess the psychometric aspects of the emerging scale, including unidimensionality and absence of item bias (differential item functioning). A total of 17 qualitative interviews generated 110 potential items for the questionnaire. The item set resolved to a 25-item scale, which satisfied model expectations (item residual mean = -0.13, SD = 1.04; person residual mean = -0.29, SD = 0.75), had good reliability (alpha = 0.86) and strict unidimensionality (t-test 7.5% confidence interval 3.8-11.2). The Manual Work Instability Scale is a short psychometrically robust questionnaire based on the concept of WI, which incorporates both musculoskeletal symptoms and relevant psychosocial factors. It may prove effective in screening and identifying WI in workers in predominantly physical occupations.

PFI-14©: A Rasch Analysis Refinement of the Psoriasis Family Index

Objective: We aimed to assess the psychometric properties of the Psoriasis Family Index (PFI), a disease-specific instrument, to measure the secondary impact of psoriasis on the quality of life (QoL) of family members and partners of psoriasis patients. Methods: PFI data were collected from 150 accompanying family members of psoriasis patients attending the dermatology department of a secondary referral hospital. Rasch analysis was used to examine various psychometric properties of the PFI, including dimensionality, response category functioning, fit statistics, scale reliability and validity, item targeting and differential item functioning (DIF). Results: The 15-item PFI with 4 response options showed good overall fit to the Rasch model. The response category almost entirely followed the Linacre criteria, demonstrating its optimal functionality. All items except one had good fit to the Rasch model. The mean fit residual for the items was -0.085 ± 1.184 (range -1.54 to 2.7). Strict unidimensionality was achieved following removal of the misfitting item 5. The revised 14-item PFI maintained strong reliability (person separation index = 0.87). The scale showed minor mistargeting (mean difference between person and item location = 0.95 logits). None of the items displayed DIF across gender or relationship type. Conclusions: The PFI is the first disease-specific measure to quantify the QoL of family members of psoriasis patients. Following the application of Rasch analysis we now recommend the use of a 14-item version of the PFI: question No. 5 from the 15-item version has been deleted. This study demonstrates the promising psychometric performance of the PFI and confirms its potential as a useful outcome measure in future clinical research.

Testing Rating Scale Unidimensionality Using the Principal Component Analysis (PCA)/<i>t</i>-Test Protocol with the Rasch Model: The Primacy of Theory over Statistics

Psychometric theory requires unidimensionality (i.e., scale items should represent a common latent variable). One advocated approach to test unidimensionality within the Rasch model is to identify two item sets from a Principal Component Analysis (PCA) of residuals, estimate separate person measures based on the two item sets, compare the two estimates on a person-by-person basis using t-tests and determine the number of cases that differ significantly at the 0.05-level; if ≤5% of tests are significant, or the lower bound of a binomial 95% confidence interval (CI) of the observed proportion overlaps 5%, then it is suggested that strict unidimensionality can be inferred; otherwise the scale is multidimensional. Given its proposed significance and potential implications, this procedure needs detailed scrutiny. This paper explores the impact of sample size and method of estimating the 95% binomial CI upon conclusions according to recommended conventions. Normal approximation, “exact”, Wilson, Agresti-Coull, and Jeffreys binomial CIs were calculated for observed proportions of 0.06, 0.08 and 0.10 and sample sizes from n= 100 to n= 2500. Lower 95%CI boundaries were inspected regarding coverage of the 5% threshold. Results showed that all binomial 95% CIs included as well as excluded 5% as an effect of sample size for all three investigated proportions, except for the Wilson, Agresti-Coull, and JeffreysCIs, which did not include 5% for any sample size with a 10% observed proportion. The normal approximation CI was most sensitive to sample size. These data illustrate that the PCA/t-test protocol should be used and interpreted as any hypothesis testing procedure and is dependent on sample size as well as binomial CI estimation procedure. The PCA/t-test protocol should not be viewed as a “definite” test of unidimensionality and does not replace an integrated quantitative/qualitative interpretation based on an explicit variable definition in view of the perspective, context and purpose of measurement.

Rasch analysis of the International Wellbeing Index in older adults

The International Wellbeing Index is a measure of general quality of life formed by two scales: the Personal Wellbeing Index (PWI) and the National Wellbeing Index (NWI). This paper studies the psychometric properties of the PWI and NWI, using Rasch analysis and classic psychometric methods. The PWI and NWI were applied to a representative sample of 1106 community-dwelling adults, aged 60 years and over, residing in Spain. Mean ± standard deviation age was 72.07 ± 7.83 years and 56.3% were women. Five PWI items (achieving in life, relationships, safety, community connectedness, and future security), and five NWI items (economic situation, state of environment, social conditions, business, and national security of the country) fitted the Rasch model. After adjusting the response scale format, satisfactory fit was obtained, with good reliability (person separation index of 0.91 for both the PSI and NWI), local independency of items, and strict unidimensionality. The measures showed adequate external construct validity with related measures. The PWI and the NWI, with fewer items and simpler response scale formats, provided valid and reliable linear measures in older adults, according to Rasch and classic psychometric analyses.

Cross-cultural validation of the Educational Needs Assessment Tool in RA in 7 European countries

BackgroundThe Educational Needs Assessment Tool (the ENAT) is a 39-item patient questionnaire originally developed in the UK to assess educational needs of patients with rheumatoid arthritis (RA). The objective of this study was to assess the cross-cultural validity of the ENAT in 7 European countries.MethodsThe ENAT was translated into Dutch, Finnish, Norwegian, Portuguese, Spanish and Swedish versions by using Beaton's cross-cultural adaptation process, and was completed by a convenience sample of patients with RA in each country. The generated country-specific data were assessed for construct validity and were then pooled and assessed for cross-cultural invariance using Rasch analysis.ResultsIndividual country-specific analysis showed adequate fit to the Rasch model after adjustment for local dependency within domains. When data from the different countries were pooled, the 39 items deviated significantly from Rasch model's expectations (X2 = 977.055, DF = 351, p = 0.000, PSI = 0.976). Again, most items within domains were found to be locally dependent, significantly affecting the fit. Consequently each domain was treated as a unit (i.e. testlet) and the ENAT was re-analysed as a seven-testlet scale resulting into a good fit to the Rasch model (X2 = 71.909; DF = 63; p = 0.207, PSI = 0.951). A test of strict unidimensionality confirmed that all domains contributed to measuring a single construct. Cross-cultural non-invariance was discounted by splitting domains for DIF maintaining an excellent fit to the Rasch model. This allowed calibration of the ENAT into an interval scale.ConclusionThe ENAT is a simple tool, which is a valid measure of educational needs of people with RA. Adjustment for cross-cultural non-invariance is available if data from the 7 European countries are to be pooled or compared.

Rasch measurement analysis of the Mayo-Portland Adaptability Inventory (MPAI-4) in a community-based rehabilitation sample.

The precise measurement of patient outcomes depends upon clearly articulated constructs and refined clinical assessment instruments that work equally well for all subgroups within a population. This is a challenging task in those with acquired brain injury (ABI) because of the marked heterogeneity of the disorder and subsequent outcomes. Although essential, the iterative process of instrument refinement is often neglected. This present study was undertaken to examine validity, reliability, dimensionality and item estimate invariance of the Mayo-Portland Adaptability Inventory - 4 (MPAI-4), an outcome measure for persons with ABI. The sampled population included 603 persons with traumatic ABI participating in a home- and community-based rehabilitation program. Results indicated that the MPAI-4 is a valid, reliable measure of outcome following traumatic ABI, which measures a broad but unitary core construct of outcome after ABI. Further, the MPAI-4 is composed of items that are unbiased toward selected subgroups except where differences could be expected [e.g., more chronic traumatic brain injury (TBI) patients are better able to negotiate demands of transportation than more acute TBI patients]. We address the trade-offs between strict unidimensionality and clinical applicability in measuring outcome, and illustrate the advantages and disadvantages of applying single-parameter measurement models to broad constructs.

Adaptation and cross-cultural validation of the rheumatoid arthritis work instability scale (RA–WIS)

Background:Despite recent advances, work disability in rheumatoid arthritis (RA) remains common. Work disability is frequently preceded by a period of work instability characterised by a mismatch between an individual’s functional...