Development and Validation of the Minnesota Inference Assessment - Assessment for Effective Intervention 2021 - Vol. 47(1) 47-52

Oct 15, 2026

Uploaded by Aran

The study develops and validates the initial version of the Minnesota Inference Assessment (MIA), a web-based measure of inference processes for students in Grades K–2. MIA evaluates inferencing without reading by pairing age-appropriate fiction and nonfiction videos with structured questioning, and a proof-of-concept test assessed technical adequacy. Results support interpreting MIA as a promising, valid, and reliable tool for nonreading inferencing. Future work will expand multiple parallel forms for K–2 progress monitoring.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages6

Reading Time15 min

AccessFree to download

Abstract

Theories of Comprehension and Types of Inferences

Limitations of Current Assessment Approaches

Existing Measures of Inference Processes

MOCCA (Multiple-Choice Online Causal Comprehension Assessment)

Bridge-IT (Bridging Inferences Test)

LARRC Inference Task (Nonreading Context)

Know-IT (Gan Facts and Episode-Based Testing)

Development and Validation of the Minnesota Inference Assessment Assessment for Effective Intervention 2021, Vol. 47(1) 47–52 © Hammill Institute on Disabilities 2020 Article reuse guidelines: [sagepub.com/journals-permissions](sagepub.com/journals-permissions)[ ](sagepub.com/journals-permissions)[DOI: 10.1177/1534508420937781](DOI: 10.1177/1534508420937781)[ ](DOI: 10.1177/1534508420937781)[aei.sagepub.com](aei.sagepub.com) Panayiota Kendeou, PhD 1, Kristen L. McMaster, PhD 1, Reese Butterfuss, MA 1, Jasmine Kim, MA 1, Susan Slater, MA 1, and Okan Bulut, PhD2 Abstract The overall aim of the current investigation was to develop and validate the initial version of the Minnesota Inference Assessment (MIA). MIA is a web-based measure of inference processes in Grades K–2. MIA leverages the affordances of different media to evaluate inference processes in a nonreading context, using age-appropriate fiction and nonfiction videos coupled with questioning. We evaluated MIA’s technical adequacy in a proof-of-concept study. Taken together, the results support the interpretation that MIA shows promise as a valid and reliable measure of inferencing in a nonreading context for students in Grades K–2. Future directions involve further development of multiple, parallel forms that can be used for progress monitoring in K–2. Keywords inference, assessment, comprehension Theories of comprehension specify how comprehension emerges during moment-by-moment processing and assume that it is based on a mental representation that is constructed over the course of reading (e.g., Kintsch, 1998) . Most theories also emphasize inferences as core processes because they are necessary for the construction of a coherent situation model, which is the product of reading comprehension (McNamara & Magliano, 2009) . Two types of inferences are necessary in this context (Graesser, 2015; McNamara, 2004): Bridging inferences, which involve connections between ideas within the text, and elaborative inferences, which involve connections between information in the text and prior knowledge (Barth et al., 2015) . To understand how inference processes break down for readers who fail to create a coherent mental representation of a text, it is essential to evaluate the construction of this representation as it occurs moment-by-moment (Kendeou, 2015; McMaster et al., 2012). However, comprehension is typically assessed after reading a text, providing very little insight into the inferential processes during reading and why these processes may succeed or fail. As we elaborate next, there are readers who struggle with precisely these processes, but existing measures that examine individual differences in comprehension processes are limited in the information they provide. Existing Measures of Inference Processes There are very few measures of inference processes in the context of reading comprehension. One such measure is the Multiple-Choice Online Causal Comprehension Assessment (MOCCA; Biancarosa et al., 2019; Carlson et al., 2014) . MOCCA evaluates the processes by which students in Grades 3 to 5 generate causal inferences. For this measure, students read short texts in which one sentence is omitted. Students are presented three sentences and choose which one best fills in the omitted line from each text. The two incorrect responses represent processes on which poor comprehenders have been shown to rely; thus, patterns in response selections provide diagnostic information about the processes students engage in. Carlson et al. reported correlations between MOCCA scores and several reading-related measures (e.g., DIBELS Oral Reading Fluency [ORF]; Woodcock–Johnson [WJ] -III word ID; Curriculum-Based Measurement [CBM] Maze) between r = −.37 to .75, and generally acceptable internal consistency (i.e., α > .60) . A second measure is the Bridging Inferences Test (Bridge-IT; Barth et al., 2015) . Bridge-IT was designed to evaluate inferential processes for st

FAQs

What is the Minnesota Inference Assessment (MIA) designed to measure?

MIA measures inference processes in Grades K–2 in a nonreading context. It uses fiction and nonfiction videos paired with questioning to evaluate inferencing.

How does MIA evaluate inferences differently from traditional reading-based assessments?

MIA removes reliance on decoding by assessing inference processes through video-based materials. Students answer questions without reading the original text.

Which types of inferences are emphasized in the theoretical background of the study?

The study distinguishes bridging inferences that connect ideas within a text and elaborative inferences that connect text information to prior knowledge.

Oct 15, 2026

Uploaded by Aran