Principal Component Analysis Of Residuals Research Articles

The Oswestry Disability Index is not a good measure for low back pain in adolescents: A Rasch analysis of data from 200 people.

To evaluate the psychometric properties of the modified Oswestry Disability Index in adolescents with low back pain through Rasch analysis. This is a psychometric study. Physical therapy clinics of a large pediatric hospital in Columbus, Ohio (United States). Two hundred adolescent patients with low back pain (112 female, 15.4 ± 1.9 years old). The modified Oswestry Disability Index. The mean modified Oswestry Disability Index score was 17.1 ± 12.1, with scores ranging from 0 to 56 on the 100-point scale. The Rasch person reliability for the Oswestry Disability Index was 0.70, representing minimal acceptable person reliability. The Oswestry Disability Index demonstrated poor item spacing, hierarchy, and targeting of the adolescents' disability level (>1 logit between person mean and item mean). A significant misfit was observed in three of 10 items. There were significant issues with the functioning of responses on all items of the Oswestry Disability Index. Principal component analysis of residuals revealed a ratio of 5.3:1, demonstrating acceptable unidimensionality. Due to multiple psychometric issues, the Oswestry Disability should not be used to assess disability in adolescents with low back pain. The Oswestry Disability Index significantly mistargets adolescent ability and only provides an acceptable assessment of the most disabled adolescents.

Rasch validation of the revised body awareness rating questionnaire (BARQ-R) in adults with and without musculoskeletal pain

BackgroundThe Revised Body Awareness Rating Questionnaire (BARQ-R) is a self-report measure of body awareness. First aim: evaluate the structural validity of BARQ-R with Rasch analysis in community-dwelling Americans with and without musculoskeletal pain. Subaim: validate a Rasch analysis of BARQ-R done in Norwegian adults with musculoskeletal pain, through a secondary analysis in our sample of Americans with musculoskeletal pain.MethodsBARQ-R has 12 items with scores ranging from 0 (completely disagree) to 3 (completely agree), with higher total scores reflecting lower degrees of body awareness. Through Rasch analysis, we evaluated unidimensionality, item hierarchy, and structural validity with item and person fit, targeting, person separation reliability (PSR), local item dependence (LID), differential item functioning (DIF), and principal components analysis of residuals (PCAR).ResultsWe recruited 623 adults with and without musculoskeletal pain (average age = 50.27 ± 17.25 years). After rescoring 1 item and deleting 3 items, the 9-item Rasch-based BARQ-R had no misfitting items, the hierarchical ordering of the items followed clinical expectations, 3 (0.48%) misfitting persons, person mean location: -0.62 ± 1.03 logits (max -0.53, min 0.72 logits), minimal floor effect (1.93%) and ceiling effect (0.48%), no DIF, and PSR = 0.72. LID was found in 5 item pairs. The PCAR’s eigenvalue was 2.18. The secondary Rasch analysis in 152 adults with musculoskeletal pain (average age = 52.26 ± 16.13 years), demonstrated that, after rescoring 2 items, BARQ-R had no misfitting items and only 2 (1.32%) misfitting persons, good targeting (person mean location: -0.36 ± 0.88 logits), minimal floor effect (0.01%), no ceiling effect (0.00%), and PSR = 0.75. LID was found in 6 item pairs. The PCAR’s eigenvalue was 2.47.ConclusionsBARQ-R had good item and person fit. PSR with items covering a limited logit range suggests that differing levels of body awareness are measured with only modest precision. Adding and revising items to cover a wider range of body awareness and to better address concepts of internal body awareness and body movements would improve BARQ-R’s utility. Further analyses are needed before BARQ-R’s use for research or in the clinic. In addition, future BARQ-R Rasch validation is needed in other populations with body awareness deficits, such as stroke or spinal cord injury.

Activities-specific balance confidence scale: a Rasch evaluation of the Arabic version in lower-limb prosthetic users

Purpose To evaluate the psychometric properties of the Arabic version of the Activities-Specific Balance Confidence Scale using the 5-option response categories for individuals with lower limb amputation (ABC-5/Ar). Materials and methods This was a methodological study on a convenience sample of individuals with unilateral lower-limb amputation attending outpatient rehabilitation centres in Saudi Arabia and Turkey (N = 155). Rasch analysis (WINSTEPS version 4.6.5) was used to examine the 5-categories rating scale structure, item fit, item difficulty hierarchy, person separation index, unidimensionality, local item dependency, and differential item functioning. Results The ABC-5/Ar 5-response option demonstrated an appropriate model fit. Most items fit the Rasch model, except for item #12 “walk in a crowded mall,” which showed an overfitting value of 0.63. The person separation indices 2.95 (Cronbach’s α = 0.96). Principal component analysis of residuals confirmed the unidimensionality of the scale; however, local dependency was detected between item #14 “Ride in escalator holding rail,” and item #15 “Ride in escalator not holding rail.” Conclusions The findings suggest that the ABC-5/Ar shows promise in assessing balance confidence in Arabic-speaking lower-limb prosthesis users. However, further studies with larger sample sizes and in diverse clinical contexts are needed to confirm its effectiveness in various clinical settings.

Computerized Adaptive Testing for the Berg Balance Scale Improves Measurement Efficiency Without Compromising Precision in People With Stroke.

Objectives of this study were to confirm the Berg Balance Scale's (BBS) measurement properties and unidimensionality with an item response theory analysis in persons with subacute and chronic stroke and examine the precision and efficiency of computerized adaptive testing (CAT). Data were obtained from 519 ambulatory persons with subacute and chronic stroke in 2 retrospective databases. A principal component analysis (PCA) of residuals was used to evaluate unidimensionality. BBS fit to a rating scale model versus a partial credit model was examined and item parameters were generated for CAT calibration. Person measures from all 14 items were defined as actual balance ability. BBS CAT simulations were used to examine changes in measurement precision with increasing number of items administered and a precision-based stopping rule (0.5 logit standard error (SE) threshold). A PCA of residuals supports the BBS's unidimensionality and Rasch analysis supports using the rating scale model for measurement. Maximum precision for BBS CAT was SE = 0.40 logits when administering all items. BBS CAT estimated balance ability was highly correlated with actual ability when 4 or more items were administered (r > 0.9). Precision was within 0.5 logits when 5 or more items were administered (SE < 0.48 logits). BBS CAT estimated balance ability was highly correlated with actual ability (r = 0.952) using a precision-based stopping rule. The average number of items administered with the precision-based stopping rule was 5.43. The BBS is sufficiently unidimensional and the rating scale model can be used for measurement. BBS CAT is efficient and replicates the full instrument's reliability when measuring balance ability in ambulatory persons with subacute and chronic stroke. Future work should aim to enhance the interpretability of measures to facilitate clinical decision making. BBS CAT provides an efficient way of measuring balance ability for individuals in stroke rehabilitation giving clinicians more time with patients.

English medium instruction learners’ self-efficacy, engagement, and satisfaction; developing a measurement instrument

ABSTRACT This paper describes the development of a survey instrument to measure learners’ self-efficacy, engagement, and satisfaction within the context of English Medium Instruction (EMI) classes, with reference to both language issues and the discipline being taught. The paper presents details of analysis of the instrument using the Rasch model. The survey was administered to 287 tertiary students enrolled in EMI classes in Japan. Results of a Rasch Principal Components Analysis of Residuals showed the three intended constructs to be largely unidimensional. Item-level analysis, as well as reliability and separation measures and an investigation of Wright maps, provided further evidence that the items are working adequately to measure each of the three constructs and to allow differentiation between levels of each construct among the respondents. In addition, the effectiveness of the 6-point Likert scale used was confirmed with an analysis of category structure functioning. Finally, a correlation analysis with a subset (n = 97) of the participants for which English proficiency test scores were available was conducted. Results showed a small, statistically significant and positive relationship between reading proficiency and each of the three constructs.

Rasch Analysis of the Mayo-Portland Adaptability Inventory for Posthospital Brain Injury Rehabilitation.

(1) Reexamine the item structure and reliability of the Mayo-Portland Adaptability Inventory (4th ed; MPAI-4) through Rasch analysis of admission and discharge scores for a large sample of adults with acquired brain injury (ABI) who participated in various types of posthospital brain injury rehabilitation (PHBIR) programs; (2) compare differential item functioning (DIF) for traumatic brain injury (TBI), stroke and other ABI; and (3) explore the viability of more specific subscales in addition to the established indices. Data from Residential Neurobehavioral, Residential Neurorehabilitation, Home and Community, Day Treatment, and Outpatient rehabilitation programs serving individuals with ABI. A total of 2154 individuals with TBI, stroke, or other ABI. Retrospective analysis of de-identified admission and discharge data from the Foundation to Advance Brain Rehabilitation (FABR) consortium database. MPAI-4. After adjusting 4 misfitting items and eliminating 20 misfitting persons, the MPAI-4 demonstrated real person reliability/separation = 0.93/3.52 and real item reliability/separation = 1.00/24.02. Independent Rasch analyses by diagnostic category found similar reliabilities and separations. Residual item correlations and principal component analysis of residuals (PCAR) indicated areas of local dependence arranged hierarchically reflecting the full-scale item hierarchy and providing the basis for 3 new subscales of Physical Abilities, Cognitive Abilities, and Autonomy. DIF across diagnostic categories revealed differences in item elevations characteristic of typical patients in each category. Measure means and SDs were very similar across categories. MPAI-4 items demonstrate very good person and item reliabilities for individuals with TBI, stroke, and other ABI at a level that supports individual evaluation. Variations in item calibrations across diagnostic categories reflect the differential characteristics of typical patients within categories. The entire measure provides an overall assessment of common sequalae of ABI, and standard indices used in combination with newly derived subscales provide more specific assessments of rehabilitation needs for treatment planning.

Validation of Hamdan intelligence scale in upper elementary grades using the Rasch model: exploratory study.

Hamdan Intelligence Scale (HIS) is the first intelligence scale that has been developed and normed in the United Arab Emirates (UAE). This study aimed to examine the refinement, validity, and reliability of HIS in upper elementary grades using the Rasch model. A total of 4,301 students (34.3% Male; 65.7% Female) from grade 4 to 6 (32.1% grade 4; 33.7% grade 5; 34.2% grade 6) were administered to the HIS. The confirmatory factor analysis was first conducted to verify the fitness of the one-factor model of the HIS. The results of validity showed strong correlation coefficients between the HIS and the Aurora-g battery (0.83) and the Raven's Standard Progressive Matrices (RSPM; 0.86). Moreover, the results of the developmental trends demonstrated that raw scores of the HIS increase with age and grade relatively constantly across composite scores. Unidimensionality was confirmed through the Confirmatory Factor Analysis and Principal Component Analysis of Residuals (PCAR). The low eigenvalues of the first contrast were below 2, and additionally, the infit and outfit mean squares ranged from 0.88 to 1.14 and 0.84 to 1.14. Rasch's person reliability result of 0.62 was acceptable reliability. The results provided strong support for the validity and reliability of using the Hamdan Intelligence Scale in the UAE environment.

Deep Residual Principal Component Analysis as Feature Engineering for Industrial Data Analytics

Deep Residual Principal Component Analysis as Feature Engineering for Industrial Data Analytics

Finding functionality: Rasch analysis of the Functionality Appreciation Scale in community-dwelling adults in the US.

The Functionality Appreciation Scale (FAS) measures an individual's appreciation for the functions their body can perform, regardless of the individual's physical limitations. Prior studies reported on internal consistency, test-retest reliability, convergent validity, and exploratory or confirmatory factor analyses, but Rasch analysis has not yet been performed to evaluate the structural validity of the FAS. We recruited community-dwelling adults at the Minnesota State Fair and through contact lists of participants identifying interest in research done in the Brain Body Mind Lab (University of Minnesota). Community-dwelling adults with spinal cord injury (SCI) completed the FAS over Zoom. We analyzed the FAS using Rasch Measurement Theory, which produced the following outputs: item, and person fit, targeting, unidimensionality, person separation reliability (PSR), local item dependence (LID), principal component analysis of residuals (PCAR), and differential item functioning (DIF). We recruited 567 participants (average age 52.15 ± 17.5 years, 63.84% women), among which 14 adults with SCI. After rescoring 3 items and deleting 1 item, the FAS had good person and item fit (except item 4). The PCAR and subsequent paired t-tests (3.53%) confirmed the unidimensionality of the scale. There was no DIF and only one item pair had LID (item 5-6). PSR was 0.75, reflecting a capacity to differentiate groups of people with high or low functionality appreciation levels. However, there was a significant ceiling effect (28.04%) and the person mean location was 3.06 ± 2.07 logits, indicating the FAS is too easy for community-dwelling adults in the US. The 6-item Rasch-based FAS demonstrated unidimensionality, good item fit (except item 4) and person fit, but the FAS will require more difficult items to be added to improve the targeting of the scale, and better reliability.

Assessing the Measurement Properties of the Self-Administered Amyotrophic Lateral Sclerosis Functional Rating Scale-Revised (ALSFRS-R): A Rasch Analysis.

The self-administered version of the Amyotrophic Lateral Sclerosis Functional Rating Scale-Revised (ALSFRS-R) is used to monitor function and disease progression in individuals with Amyotrophic Lateral Sclerosis (ALS). However, the performance of the self-administered ALSFRS-R has not been assessed using Rasch Measurement Theory. Therefore, the purpose of this study was to examine the psychometric properties of the self-administered ALSFRS-R using Rasch analysis. Rasch analysis was performed on self-administered ALSFRS-R data from individuals with ALS across Canada. The following 6 aspects of Rasch analysis were examined using RUMM2030: fit via residuals and chi-square statistics, targeting via person-item threshold maps, dependency via item residual correlations, unidimensionality through principal components analysis of residuals, reliability via person separation index, and stability through differential item functioning analyses for sex, age, and language. Analysis was performed on 122 participants (mean age: 52.9years; 62.8% men). The overall scale demonstrated good fit, reliability, and stability; however, multidimensionality was found. To address this issue, items were divided into 3 subscales (bulbar, motor, and respiratory function), and Rasch analysis was performed for each subscale. The subscales demonstrated good fit, reliability, stability, and unidimensionality. However, there were still issues with item dependency for all subscale and targeting for bulbar and respiratory subscales. The self-administered ALSFRS-R is reliable, internally valid, and stable across sex, age, and language subgroups; however, it is recommended that the ALSFRS-R be scored by subscale. Future studies can look at revising and/or adding items to tackle misfit, redundancy, and ceiling effects. Self-administered measures are simple to administer and inexpensive. The self-administered ALSFRS-R was found to be psychometrically sound and can be used as a tool to monitor disease progression and function in ALS.

Rasch validation of a new scale to measure dependency in arm use in daily life: the Upper Limb Lucerne ICF-based Multidisciplinary Observation Scale.

About 77% of adults with stroke have upper limb impairments. Many scales are available to measure the impairment and activity level of the affected limb. However, an observational scale to assess dependency on others in upper limb performance during daily life activities instead of laboratory settings is lacking. Therefore, we developed a new 5-item Upper Limb Lucerne ICF-based Multidisciplinary Observation Scale (UL-LIMOS). As next step in the psychometric analysis, we evaluated the unidimensionality and structural validity of the UL-LIMOS with Rasch Measurement Theory and we calculated a cut-off score for independent arm use in daily life activities at discharge. This is a single-center cross-sectional study in adults with (sub) acute stroke. We applied Rasch Measurement Theory (RMT) to analyze the structural validation and unidimensionality of the UL-LIMOS. The outputs provide evidence of unidimensionality, item and person fit, overall fit, differential item functioning (DIF), principal component analysis of residuals (PCAR), person separation reliability (PSR), and residual item correlations (to identify local item dependence). Person mean location, floor and ceiling effects identify proper targeting. We recruited 407 adults with (sub) acute stroke (median age 63 years, 157 women). All items and persons fit the Rasch model. The PSR of 0.90 indicates that clinicians and researchers can reliably use the scale for individual decision-making. There were small floor (2.70%) and ceiling (13.00%) effects. The average person mean location was 1.32 ± 2.99 logits. There was no DIF. PCAR eigenvalue was 2.46 with 49.23% explained variance. Paired t-tests revealed that 0.89% of person locations were significantly different, confirming unidimensionality. One pair of items (arm and hand use and fine hand use) showed residual item correlations. The ROC's AUC was 0.90, CI95% = [0.85-0.96] with cut-off score of ≥14/20, and high sensitivity (87%, CI95% = [81%-91%]), specificity (83%, CI95% = [77%-87%]) for independent arm use in daily living at discharge. The new Rasch-based UL-LIMOS is a valid ICF-based observation performance scale at the ICF-activity level, to evaluate dependency during upper limb use in daily life in adults with stroke. Additional psychometric analyses are warranted. The UL-LIMOS would be a valuable addition to the core assessments of adults with (sub) acute stroke.

Preliminary Rasch analysis of the multidimensional assessment of interoceptive awareness in adults with stroke.

The Multidimensional Assessment of Interoceptive Awareness (MAIA) measures interoceptive body awareness, which includes aspects such as attention regulation, self-regulation, and body listening. Our purpose was to perform a preliminary validation of the MAIA in adults with stroke using Rasch Measurement Theory. The original MAIA has 32 items that measure interoceptive sensibility, which is an aspect of body awareness. We performed a preliminary analysis with Rasch Measurement Theory to evaluate the unidimensionality and structural validity of the scale. We investigated overall fit to assess unidimensionality, person and item fit, person separation reliability, targeting, local item dependence, and principal components analysis of residuals. Forty-one adults with chronic stroke (average 3.8 years post-stroke, 13 women, average age 57±13 years) participated in the study. Overall fit (χ 2 = 62.26, p = 0.26) and item fit were obtained after deleting 3 items and rescoring 26 items. One participant did not fit the model (2.44%). There were no floor (0.00%) or ceiling effects (0.00%). Local item dependence was found in 42 pairs. The person separation reliability was 0.91, and the person mean location was 0.06±1.12 logits. The preliminary structural validity of the MAIA demonstrated good targeting and reliability, as well as unidimensionality, and good item and person fit in adults with chronic stroke. A study with a larger sample size is needed to validate our findings.

Validity and test-retest reliability of the Ugandan version of the Pediatric Evaluation of Disability Inventory (PEDI-UG) in children and youth with cerebral palsy.

Validity of the Ugandan version of the Pediatric Evaluation of Disability Inventory (PEDI-UG) was previously investigated on typically developing children. This study aimed to investigate the validity, test-retest reliability and minimal detectable change (MDC) of the PEDI-UG in children and youth (C&Y) with cerebral palsy (CP). A cross-sectional study design with 118 C&Y with CP (44.7% girls) aged 10months-22.5years were included in the study; 37 of them completed the PEDI-UG twice to investigate test-retest reliability, determined by calculating the intraclass correlation coefficient (ICC). Additionally, data from 249 typically developing children were used for differential item functioning (DIF) analysis. The validity of the PEDI-UG was investigated by Rasch analysis. The Kruskal-Wallis test and Spearman's correlation coefficient were calculated to investigate associations between PEDI-UG scores and external classification systems. The principal component analysis of residuals indicated unidimensionality in all domains. The ICC values were excellent (0.98-0.99), and the MDCs were less than 6 and 13 (on a 0-100 scale) for the functional skills and caregiver assistance parts, respectively. The four-category caregiver assistance rating scale fulfilled the criteria for the analysis of rating scale functioning. In total, 78 of 189 items in the functional skills domain and two items in the caregiver assistance domain demonstrated DIF between C&Y with CP and TD children. The Kruskal-Wallis test (p < 0.05) and Spearman's correlation (coefficients of -0.93 to -0.78) supported the validity of PEDI-UG. The current diagnose-specific version of PEDI-UG demonstrates evidence for validity as a measure of ability in C&Y with CP in Uganda and other similar settings, being a promising tool for use in clinical practice and research. Conversion tables and MDC values are provided to facilitate clinical adoption of the measure.

The effectiveness of weighted least squares means and variance adjusted based fit indices in assessing local dependence of the rasch model: Comparison with principal component analysis of residuals.

Local independence is a principal assumption of applying latent variable models. Violations of this assumption might be stemmed from dimensionality (trait dependence) and statistical independence of item responses (response dependence). The purpose of this study is to evaluate the sensitivity of weighted least squares means and variance adjusted (WLSMV) based global fit indices to violations of local independence in Rasch models, and compare those indices to principal component analysis of residuals (PCAR) that is widely used for Rasch models. Dichotomous Rasch model is considered in this simulation study. The results show that WLSMV-based fit indices could detect trait dependence, but are to be limited with regard to response dependence. Additionally, WLSMV-based fit indices have advantages over the use of PCAR since WLSMV-based global fit indices are consistent regardless of sample size and test length. Though it is not recommended to apply exact benchmarks for those indices, they would provide practitioners with a method for evaluating the degree to which assumption violation is problematic for their data diagnostic purpose.

Development and validation of the symptom burden questionnaire for long covid (SBQ-LC): Rasch analysis

ObjectiveTo describe the development and validation of a novel patient reported outcome measure for symptom burden from long covid, the symptom burden questionnaire for long covid (SBQ-LC).DesignMultiphase, prospective mixed methods...

Item-Level Analysis of a Newly Developed Interactive Nutrition Specific Physical Exam Competency Tool (INSPECT) Using the Rasch Measurement Model.

The Interactive Nutrition Specific Physical Exam Competency Tool (INSPECT) is a tool designed specifically to observe and measure registered dietitian nutritionists’ (RDNs) nutrition-focused physical exam (NFPE) competence in authentic acute care settings. The initial INSPECT items were generated and tested for content and face validity using expert RDNs’ input. The INSPECT was further examined for inter-rater, intra-rater, and internal consistency using clinical supervisor observations of RDNs performing NFPE on patients in real-life acute care settings. These previous studies showed the INSPECT to have excellent content validity, acceptable face validity, good inter-rater reliability, moderate to strong intra-rater reliability, and excellent internal consistency. In the current study, the Rasch measurement model was applied to examine the item-level properties of the INSPECT. Results confirm that the INSPECT measured a single construct. All items fit the established criteria for clinical observations of >0.5 and <1.7, had positive point measure correlations, met the Wright Unidimensionality Index criteria of ≥0.9, exhibited one latent construct with >40% variance explained by the Rasch dimension as well as a sub-dimension based on item difficulty from the principal component analysis of the first contrast Rasch residuals. Rasch rating scale analysis revealed that the rating scale and majority of the items (39/41) fit the Rasch model. Rasch item hierarchy analysis matched the a priori hypothesized hierarchy for the top-most and bottom-most items. Ceiling effects were seen for three items (hand hygiene, personal protective equipment, and patient position) and one item (handgrip using hand dynamometer) reached the floor effect. Rasch reliability assessment demonstrated high person reliability (0.86), high item reliability (0.96), and person separation of 3.56 ability levels. The principal component analysis of residuals revealed two factors based on item difficulty, one for micronutrient exam and another for macronutrient exam, initial steps, and bedside manner. The resulting two factors may likely be due to a sub-dimension of the latent NFPE trait. Overall, the INSPECT items were found to have good item-level psychometrics. Continued testing of the INSPECT with RDNs at different ability levels will help to determine cut-off scores ranging from novice to expert. Establishing cut-off scores for the INSPECT will further enhance the utility of the tool.

Psychometric Properties of the Patient Activation Measure in Community-Dwelling Adults in Singapore.

IntroductionMeasuring health activation in general population using valid instruments is needed to facilitate the evaluation of health education and behavioral programs in community. The 13-item Patient Activation Measure was well validated in patients with different chronic diseases but rarely validated in general population. The objective of this study was to assess the psychometric properties of the Patient Activation Measure among community-dwelling adults in Singapore.MethodsData of participants having valid responses to the English-version measure (N = 824) were analyzed. The psychometric properties were assessed by demonstrating evidence for uni-dimensionality using Rasch Principal Component Analysis of Residuals, known-group validity, convergent and divergent validity, and internal consistency reliability using Cronbach’s alpha.ResultsThe uni-dimensionality of the Patient Activation Measure was supported by the Rasch Principal Component Analysis of Residuals results. Participants having multimorbidity or polypharmacy and being inactive in physical activity had significantly lower activation scores. The activation score was positively and moderately correlated with health confidence measured by the Health Confidence Measure (r = .38, P < .001), and negatively and weakly correlated with depressive symptoms measured by the Patient Health Questionnaire (r = − .13, P < .001). The internal reliability was good with a Cronbach’s alpha of .82.ConclusionThe 13-item Patient Activation Measure has acceptable construct validity and good internal consistency among community-dwelling adults. It is a potential instrument to measure health activation in this population. Further research is required to investigate the expansion of response options, validate the cut-off scores for the activation levels and examine the test-retest reliability and responsiveness.

Examining the STEM-Science Achievement Test (SSAT) Using Rasch Dichotomous Measurement Model

The main purpose of this study is to develop a valid and reliable instrument to measure the STEM-Science achievement of primary school students in Malaysia. Six Year 4 Science topics (Scientific Skills, Life Processes of Human, Properties of Materials, Measurement, Solar System, Importance of Technologies in Life) and Six Year 5 Science topics (Rules and Regulation in Science Lab, Life Processes of Plants, Acids and alkali, Electricity, Earth and Space Science, Technology and Sustainable Life) have been included in the development of the STEM-Science Achievement Test (SSAT). 226 Year 4 and 226 Year 5 primary school students in Sabah responded to the developed instrument to test their STEM-Science knowledge. The Rasch dichotomous measurement model approach was used to evaluate the validity and reliability of the SSAT. The validity assessed the Point-Measure Correlation (PTMEA CORR), Principal Component Analysis of Residuals (PCAR), as well as Infit and Outfit Mean Squares (MNSQ). In terms of reliability, Cronbach’s alpha, item reliability, and item separation index were analysed. The analysis results revealed the presence of unidimensionality for both objective and subjective items. For objective items, the reported values for Cronbach’s Alpha are .81 and .83; item reliability are .95 and .95; item separation are 4.21 and 4.25 for Year 4 and Year 5 students, respectively. Standardised residual correlations for Year 4 and Year 5 subjective items also showed satisfactory values. The assessment using Rasch measurement model has proven that SSAT is a valid and reliable instrument to measure Malaysian primary students’ STEM-Science-related knowledge.

ASSESSING THE VALIDITY AND RELIABILITY OF SCIENCE MULTIPLE CHOICE TEST USING RASCH DICHOTOMOUS MEASUREMENT MODEL

Multiple choice tests are widely applied to assess students’ knowledge in science education. This study aimed at assessing the validity and reliability of Science Multiple-choice Test in Malaysia. The items for this test were formulated by the researcher together with a panel of science teachers and the head of the science department with close reference to Secondary School Standard Curriculum (KSSM) syllabus. The test consists of 50 multiple-choice items with four options. Rasch measurement model was adopted to evaluate the quality of the Science Multiple-choice Test in terms of reliability analysis, item polarity analysis (PTMEA-CORR), item fit analysis and Principal Component Analysis of Residuals (PCAR). The reliability analysis was performed using Cronbach’s Alpha, and the results of reliability and separation index respectively indicated good reliability level of the test items. In order to improve the validity of the test, two negatively worded items (Q39 and Q40) were removed. Lastly, the PCAR analysis showed the unexplained variance in the 1st contrast (5.4%) was found to be well controlled and was below the ceiling value of one-third of the variance explained by the item (18.7%). However, the positive value of the disattenuated correlations indicate no evidence of the presence of secondary dimension. Keywords: multiple-choice test, Rasch measurement model, reliability and validity, science subject, secondary school standard curriculum

Cross-Diagnostic Validity of the Congenital Glaucoma Caregiver's Quality of Life Questionnaire (CarCGQoL).

PurposeThe Caregiver's Congenital Glaucoma QoL (CarCGQoL) questionnaire was proposed as a measure of QoL of caregivers with children with primary congenital glaucoma (PCG). Support for its psychometric properties among other diagnostic groups is required for scores to be interpreted in the same manner across groups. Therefore we investigated the measurement properties and cross-diagnostic validity of the CarCGQoL questionnaire among caregivers of children with congenital cataract, retinopathy of prematurity (ROP), and blinding corneal disorders.MethodsEight hundred ninety-one caregivers (mean age, 28.3 years; 76% mothers) of children with congenital cataract (n = 407), ROP (n = 272), and blinding corneal disorders (n = 212) completed the CarCGQoL questionnaire. Rasch analysis was used to investigate the psychometric properties. Unidimensionality (by principal components analysis of residuals, PCA) was examined for each group and for pooled sample. Differential item functioning (DIF) was investigated to explore whether bias in responses to the questionnaire existed for certain subgroups as compared to the reference group (PCG).ResultsAcross groups, six items necessitated removal because of misfit (two common and four uncommon), after which three different versions of the questionnaire emerged. Measurement precision was adequate for each group and for the pooled sample (0.80). Unidimensionality was observed, albeit with some DIF. Regardless of the level of QoL, caregivers in the pooled sample were more likely to endorse two items reflecting ability to face child's disease, and interest to pursue leisure activity, as compared with caregivers of children with PCG.ConclusionsCare must be taken when data from the CarCGQoL questionnaire from different pediatric ocular conditions are pooled, given the presence of DIF between the reference group (PCG) and the pooled sample.Translational RelevanceWhen evaluating the impact of interventions on the caregiver's QoL using the CarCGQoL questionnaire in a pooled sample of pediatric ocular conditions, cross-diagnostic DIF must be taken into account.