The Development and Assessment of an Instrument for Measuring Mental Model Styles in Korea Page: 10
20 p.View a full description of this article.
Extracted Text
The following text was automatically extracted from the image on this page using optical character recognition software:
Analysis Plan
Aligned with the primary purpose of this research, the following statistical procedures
were performed: (1) exploratory factor analysis (EFA) of the theoretically proposed scales, (2)
confirmatory factor analysis (CFA) of the revised factor structure resulting from the EFA, and
(3) reliability analysis of the final scale scores. All data analyses were performed with SPSS and
LISREL statistical software. A total of 701 completed responses were used for further data
analysis after deleting non-completed responses.
Because we were able to obtain an adequate sample size (n = 701), according to the
Central Limit Theorem (CLT), basic normal distribution could be assumed (Rice, 2006; Urdan,
2005). Empirically, the pattern and shape of the sample distribution supported normal
distribution as well in terms of the Skewness values (ranges from -.470 to -.179) and Kurtosis
values (ranges from -.467 to .515) (Hair, Black, Babin, Anderson, & Tatham, 2006; Leech,
Barrett & Morgan, 2005).
Construct Reliability and Validity
EFAs were conducted using responses from a randomly chosen half of the sample (n =
348) to examine the factor structure of the MMSS against its theoretical model. The remaining
responses were saved for a CFA (n2 = 353) to confirm the results of the EFA. The full sample
was then used to compute reliability coefficients for the final scale scores.
EFA was chosen as the first analysis strategy based on our expectation that the theoretical
model of the MMSS might not fit the data well and that the factor model would require
refinement. Our goal was to isolate factors that had a simple structure or as Thompson (2004)
indicated, were interpretable. For the EFAs, we used principal axis factoring and promax
rotation. These methods were chosen because an underlying theoretical structure was
hypothesized and it was expected that some of the factors would correlate (c.f., Henson &
Roberts, 2006).
To confirm the psychometric properties of the MMSS scales derived from the EFA, a
CFA was conducted. CFA is regarded as one of the most reliable tests to assess the construct
validity of the proposed measurement based on fitness between measurement and collected
responses (Hair et al., 2006; Thompson, 2004). To assess model fit, two error term detection
indices, Steiger's root mean square error of approximation (RMSEA; 1990) and J6reskog and
S6rbom's standardized root mean square residual (SRMR; 2001), and four goodness of model fit
indices, J6reskog & S6rbom's goodness-of-fit index (GFI; 2001), Bentler's comparative fit index
(CFI; 1990), Bentler and Bonett's nonnormed fit index (NNFI; 1980), and goodness-of-fit (GFI)
were considered along with basic chi-square estimates.
To assess reliability of MMSS scale scores, coefficient alpha was used. Coefficient alpha
is a common statistic used to measure the internal consistency of scale scores (Henson, 2001).
Results
This section presents the results of the EFA, CFA, and reliability analyses for the
subsample 1, subsample 2, and the full sample, respectively, along with descriptive statistics.
EFA (Subsample 1)
In the initial EFA, five factors were extracted in keeping with the theoretical model of the
items. The five factors explained 62.87% of the variance of the 25 items. An analysis of theLearning and Performance Quarterly, 1(1), 2012
10
Upcoming Pages
Here’s what’s next.
Search Inside
This article can be searched. Note: Results may vary based on the legibility of text within the document.
Tools / Downloads
Get a copy of this page or view the extracted text.
Citing and Sharing
Basic information for referencing this web page. We also provide extended guidance on usage rights, references, copying or embedding.
Reference the current page of this Article.
Chermack, Thomas J.; Song, Ji Hoon; Nimon, Kim F.; Choi, Myungweon & Korte, Russel F. The Development and Assessment of an Instrument for Measuring Mental Model Styles in Korea, article, 2012; [Denton, Texas]. (https://digital.library.unt.edu/ark:/67531/metadc115197/m1/10/: accessed April 19, 2024), University of North Texas Libraries, UNT Digital Library, https://digital.library.unt.edu; crediting UNT College of Information.