Assessing Between-Group Differences in Information Systems Research: A Comparison of Covariance- and component-Based SEM1
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Multigroup or between-group analyses are common in the information systems literature. The ability to detect the presence or absence of between-group differences and accurately estimate the strength of moderating effects is important in studies that attempt to show contingent effects. In the past, IS scholars have used a variety of approaches to examine these effects, with the partial least squares (PLS) pooled significance test for multigroup becoming the most common (e.g., Ahuja and Thatcher 2005; Enns et al. 2003; Zhu et al. 2006). In other areas of social sciences (Epitropaki and Martin 2005) and management (Mayer and Gavin 2005; Song et al. 2005) research, however, there is greater emphasis on the use of covariance-based structural equation modeling multigroup analysis. This paper compares these two methods through Monte Carlo simulation. Our findings demonstrate the conditions under which covariance-based multigroup analysis is more appropriate as well as those under which there either is no difference or the component-based approach is preferable. In particular, we find that when data are normally distributed, with a small sample size and correlated exogenous variables, the component-based approach is more likely to detect differences between-group than is the covariance-based approach. Both approaches will consistently detect differences under conditions of normality with large sample sizes. With non-normally distributed data, neither technique could consistently detect differences across the groups in two of the paths, suggesting that both techniques struggle with the prediction of a highly skewed and kurtotic dependent variable. Both techniques detected the differences in the other paths consistently under conditions of non-normality, with the component-based approach preferable at moderate effect sizes, particularly for smaller samples.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it