← Every exhibit

Figure Figure 1 From the research bench

The bilingual measurement-release gate

Match the intended English-German comparison to the factor evidence, sample, method, and limitation before release.

A five-column worksheet with five blank rows. The columns ask for release question, evidence or test, invariance level or method, permitted comparison, and comparison not licensed. It is a blank decision framework, not score data.RELEASE QUESTIONConstruct, groups,languages,comparison?EVIDENCE OR TESTTranslation, loadings,intercepts, thresholds,sample, mode?LEVEL OR METHODConfigural, metric,scalar, partial orapproximate?COMPARISONPERMITTEDWhat exact sentencecan be stated fromthe result?COMPARISON NOTLICENSEDWhich stronger claimremains outside theevidence, and why?Five rows are blank reader inputs. The worksheet is a release gate, not a statistical result or universal translationstandard.
Vector SVG Cite Embed

Reference & Evidence

Source: Author's release worksheet grounded in Cieciuch et al. (2014), Klopp and Klößner (2023), and Prem et al. (2021). Blank fields are reader inputs; no instrument, participant, or score data is supplied.

Each line is a claim from the register this journal publishes against, resolved from the register at build time.

  • A Measurement invariance concerns whether the same construct is measured in the same way across groups; metric and scalar invariance license different comparisons cieciuch-davidov-2014-exact-approximate-measurement-invariance · CDAS14-C1
  • A Configural invariance fixes the loading pattern, metric invariance constrains loadings, and scalar invariance additionally constrains indicator intercepts cieciuch-davidov-2014-exact-approximate-measurement-invariance · CDAS14-C2
  • A Establishing invariance in one study does not make an instrument permanently invariant; it should be tested again across groups cieciuch-davidov-2014-exact-approximate-measurement-invariance · CDAS14-C3
  • A In a metric model, scaling restrictions needed for identification must be distinguished from the equality conditions being tested, which is why the paper can show that "it is impossible to test the invariance of the estimated loading of exactly one indicator, because metric MI models aimed at doing so are actually equivalent to the configural model" klopp-klossner-2023-metric-measurement-invariance · KK23-C2
  • A When scaling restrictions are placed in one group only, key fit quantities do not depend on the scaling choice: "neither the choice of the scaling method in general nor the choice of a particular referent indicator affects the value of the discrepancy function", under the rule "Apply the scaling restriction in one group only!" klopp-klossner-2023-metric-measurement-invariance · KK23-C3
  • A The researchers translated the German CODE scale into English using native speakers in both languages and a back-translation method prem-et-al-2021-english-german-code-scale · PREM21-C2
  • A The German and English samples were tested sequentially for configural, metric, and scalar invariance, with subscale means treated as comparable under the reported scalar model prem-et-al-2021-english-german-code-scale · PREM21-C3

Grades: A, verified against the printed page of the primary source · B, primary source, text layer only · C, authoritative secondary · D, reported.

Related exhibits