Skip to main content

Table 1 Sequence of Rasch measurement models estimated for each data set

From: Is it harder to know or to reason? Analyzing two-tier science assessment items using the Rasch measurement model

Model Dimensions
Number Terms Data Set 1 (LPDI) Data Set 2 (CTSR)
1 None Unidimensional
(17 parameters)
Unidimensional
(25 parameters)
2 Tier Two dimensions:
Tier 1; Tier 2
(19 parameters)
Two dimensions:
Tier 1;Tier 2
(27 parameters)
3 Content Two dimensions:
Light Propagation; Visibility
(19 parameters)
Four dimensions:
Control of Variables; Combinatorial Reasoning; Proportional Reasoning; Probabilistic Reasoning
(34 parameters)
4 Tier × Content Four dimensions:
Tier 1; Tier 2
×
Light Propagation; Visibility
(26 parameters)
Eight dimension:
Tier 1; Tier 2
×
Control of Variables; Combinatorial Reasoning; Proportional Reasoning; Probabilistic Reasoning
(60 parameters)
  1. Note. All models were estimated as dichotomous Rasch models. The Model 4 for both data sets was estimated as between-item models, meaning that an item was assigned to a combination of tier and content. The parameters listed indicate the complexity of the statistical model by representing the combination of dimensions and items in that model