Coefficient Alpha and Reliability of Scale Scores

Almehrizi, Rashid S.

doi:10.1177/0146621613484983

Cited by 22 publications

(21 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The Cronbach's alpha coefficient ranging from 0.0 to 1.0 was used to measure internal consistency of the tool. Higher values are associated with higher internal consistency, while the cut-off point equal to or exceeding 0.7 was used to consider the test reliable [17]. To evaluate reproducibility of the tool, we performed a test-retest assessment after 15 days and evaluated it using Pearson's correlation coefficient.…”

Section: Discussionmentioning

confidence: 99%

Cross-cultural Adaptation and Validation of Patient Safety Culture Assessment Tool for Clinical Staff in Kazakhstan

Kyrykbayeva

Semenova

Goremykina

et al. 2021

Open Access Maced J Med Sci

View full text Add to dashboard Cite

BACKGROUND: Modern healthcare systems lack empirical models and appropriate tools to evaluate the safety culture, which encourages the need for their development. The national healthcare system of Kazakhstan has no empirically tested robust tools to evaluate the safety culture. AIM: The present paper aims at performing translation, cross-cultural adaptation, and validation of the “Patient safety culture” tool (from here – the tool) as a means for evaluating safety culture in healthcare establishments of the Republic of Kazakhstan. METHODS: The study design was cross-sectional. The original “Patient safety culture” tool is an integrated evaluation framework consisting of 7 scales and 62 items (6 general questions and 56 targeted questions) and reflecting different aspects of risk-related components of safety culture. The Cronbach’s alpha coefficient ranging from 0.0 to 1.0 was used to measure the internal consistency of the tool. To evaluate the tool’s reproducibility, we performed a test-retest assessment after 15 days and evaluated it using Pearson’s correlation coefficient. RESULTS: The overall Cronbach’s alpha for different scales ranged from 0.27 to 0.75. Subscales “Safety procedures” and “Safety Training” had the lowest coefficients. At test-retest, the Pearson’s correlation coefficient ranged from 0.934 to 0.969 among the scales. CONCLUSION: The study is dedicated to cross-cultural adaptation and validation of the “Patient safety culture” tool and describes the development of a translation protocol with subsequent cognitive debriefing and field testing. The field testing helped to confirm good validity, reliability, and reproducibility. The tool might undergo additional modification after further testing.

show abstract

Section: Discussionmentioning

confidence: 99%

Cross-cultural Adaptation and Validation of Patient Safety Culture Assessment Tool for Clinical Staff in Kazakhstan

Kyrykbayeva

Semenova

Goremykina

et al. 2021

Open Access Maced J Med Sci

View full text Add to dashboard Cite

show abstract

“…Fleiss and Shrout (1978) and Shrout and Fleiss (1979) suggested confidence intervals for intraclass correlations (generalizability coefficients) for one‐way model (single‐facet nested design) and two‐way model (single‐facet crossed design). Also, several researchers using different methods have proposed the sampling distributions of coefficient alpha (Cronbach, 1951), which is equal to generalizability coefficient for single‐facet crossed design (Almehrizi, 2013). For example, Kristof (1963) and Feldt (1965) independently derived the exact distribution of the coefficient alpha and its standard error under the assumption that the covariance matrix of items has a compound symmetry form.…”

Section: Statistic Estimators Traditional ε̂2mentioning

confidence: 99%

Standard Errors of Variance Components, Measurement Errors and Generalizability Coefficients for Crossed Designs

Almehrizi

2020

J Educational Measurement

Self Cite

View full text Add to dashboard Cite

Estimates of various variance components, universe score variance, measurement error variances, and generalizability coefficients, like all statistics, are subject to sampling variability, particularly in small samples. Such variability is quantified traditionally through estimated standard errors and/or confidence intervals. The paper derived new standard errors for all estimated statistics for two crossed designs (single‐facet design and two‐facet design) in generalizability theory. The derivation was based on the assumption of multivariate normal distribution for observation scores using delta method. The derivation was differentiating between fixed and random facets. The adequacy of the derived standard errors was examined using Mont Carlo simulation for the two designs under several test conditions and compared to the traditional existing methods as well as implementation on real data. Results showed that the derived standard errors for all estimators are converging to the empirical standard errors for all simulation conditions with the two designs for both normal and non‐normal continuous data.

show abstract

“… is the PDF of the xj scores given examinee a's ability. This PDF can be obtained through the Lord and Wingersky (1984) recursion formula using which was employed by Almehrizi (2013;2016).…”

Section: The New Section-level Person Fit Statisticmentioning

confidence: 99%

Residual-Based Person Fit Statistics over Test Sections

Almehrizi¹

2019

J. educ. psychol. stud.

View full text Add to dashboard Cite

Most tests are composed of multiple sections (each section has group of items) such as different item formats, different content category, competencies, different difficulty levels, test dimensions, testlets, and interpretive exercise items. Students could show unexpected and unacceptable responses across these sections. Studying person fit over item level cannot detect aberrant response over test sections. The study proposes a residual-based person fit statistic over test sections with a dichotomous IRT model. The paper demonstrates the new section-level person fit statistic and investigates its distributional properties and power of detecting aberrance in person responses with comparison to Wright's between person fit statistic. The proposed section-level person fit statistic shows superior distributional properties with both true and real ability and item parameters. Moreover, the performance of the proposed person fit statistic is also examined with real data.

show abstract

Coefficient Alpha and Reliability of Scale Scores

Cited by 22 publications

References 32 publications

Cross-cultural Adaptation and Validation of Patient Safety Culture Assessment Tool for Clinical Staff in Kazakhstan

Cross-cultural Adaptation and Validation of Patient Safety Culture Assessment Tool for Clinical Staff in Kazakhstan

Standard Errors of Variance Components, Measurement Errors and Generalizability Coefficients for Crossed Designs

Residual-Based Person Fit Statistics over Test Sections

Contact Info

Product

Resources

About