Evidence Regarding the Internal Structure: Confirmatory Factor Analysis

Lewis, Todd F.

doi:10.1080/07481756.2017.1336929

Cited by 92 publications

(72 citation statements)

References 4 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Table 4 shows the number of items before and after eliminating items from each data set. Although with 242 EC and 170 EL participants, sample sizes for these two groups are smaller than typically recommended for these analyses, the models converged and results suggest relatively strong model fit, particularly in the improved model, as evident in the Technical Appendix (Lewis, 2017). Tables 5 and 6 with model fit statistics available in the Technical Appendix.…”

Section: Resultsmentioning

confidence: 94%

Developing Instruments to Measure Montessori Instructional Practices

Murray

Chen

Daoust

2019

JoMR

View full text Add to dashboard Cite

Researchers who study any intervention must rule out potential alternative explanations for their results by establishing that the program being investigated is implemented with fidelity. Various instructional practices are attributed to the Montessori Method because the term is not legally protected, meaning any school can say it is Montessori regardless of the degree to which it adheres to practices generally understood to represent Montessori education. Researchers have used a variety of tools to measure the fidelity of Montessori environments they study, but most of these tools lack an extensive psychometric foundation or are labor intensive, requiring in-person observation. The purpose of this study was to examine the psychometric properties of instruments that were developed to measure Montessori implementation through Early Childhood (EC) and Elementary (EL) teachers’ reported instructional practices. Findings supported three hypothesized dimensions of Montessori implementation (structure, curriculum, and freedom), which worked fairly well in describing practices. While the properties of these instruments are promising and provide preliminary supporting evidence, results of this analysis suggest further refinement of the items in these instruments is necessary with larger and more diverse samples. While we do not suggest that these are finalized tools, we believe they provide a valuable starting point that is a vast improvement over the requirement of investigators to develop their own instruments as part of each Montessori study they design. The authors hope other researchers will incorporate these instruments into their studies to help build a robust body of evidence supporting their use.

show abstract

Section: Resultsmentioning

confidence: 94%

Developing Instruments to Measure Montessori Instructional Practices

Murray

Chen

Daoust

2019

JoMR

View full text Add to dashboard Cite

show abstract

“…Although the chi-square statistics were significant, it is often advocated that this statistic can be over-sensitive and over-inflated with large sample sizes. 49 In addition, the Cronbach alpha (α) and McDonald's omega (ω) of 0.93 across the composite sample provided further support for the internal consistency of the DWiSS.…”

Section: Confirmatory Factor Analysismentioning

confidence: 81%

“…We first tested the factor structure separately for each gender, which confirmed an excellent model fit for men and women. As noted above the chi-square statistic can be sensitive to over-inflation in larger sample sizes, 49 and thereby the significant chi-square statistic in men may be due to the larger number of men in the sample. Then, we tested for configural invariance whereby no constraints were placed on parameters, which indicated this was supported via an excellent model fit.…”

Section: Gender Invariancementioning

confidence: 99%

Testing the validity and reliability of the doping willingness in sport scale

Stanger

Whitaker²,

Backhouse

2020

Scandinavian Med Sci Sports

View full text Add to dashboard Cite

Although research investigating doping in sport is burgeoning, there is still a lack of proxy measures of doping behavior that have undergone extensive psychometric testing. To address this issue, we modified a previously used measure of doping willingness in sport and tested aspects of validity and reliability across four studies. In Study 1, we provided support for the face and content validity of the items, and then found support for the factor structure of the scale in a sample of athletes (N = 205) using confirmatory factor analysis. In Study 2, we collected data from an independent sample of athletes (N = 236) to provide further evidence for the factor structure of the scale using confirmatory factor analysis as well as provided evidence for concurrent and discriminant validity. In Study 3, a further independent sample of athletes (N = 144) completed the scale and provided support for discriminant and predictive validity of the scale. In Study 4, we collected data from a further independent sample (N = 74) to provide support for the test‐retest reliability, and stability of items. Lastly, a confirmatory factor analysis was conducted on the samples across Studies 3 and 4, and the composite sample across all four studies which provided further support for the factor structure of the final 8‐item scale. Taken together, these findings provide psychometric support for the scale to be used to measure the willingness of athletes to use banned substances to help facilitate future research investigating doping in sport.

show abstract

“…Statistical power analysis . Lewis (2017) indicated 250 to 500 participants as necessary for data to stabilize within confirmatory factor analysis procedures. Our sample of 339 participants was situated within that acceptable range.…”

Section: Methodsmentioning

confidence: 99%