Purposes: This study uses think-alouds and cognitive interviews to provide validity evidence for an online formative assessment--i-Ready Standards Mastery (iSM) mini-assessments--which involves a heavy use of innovative items. iSM mini-assessments are intended to help teachers determine student understanding of each of the on-grade-level Common Core State Standards (CCSS) before, during, or after instruction. These mini-assessments focus on measuring how well students have mastered the material and identifying potential misconceptions that require targeted instruction. That said, it is clear that three salient features of iSM will shape the motif of this study: (1) a CCSS-aligned assessment; (2) a formative assessment that aims to provide teachers with actionable information to pinpoint misconceptions and guide remediation; and (3) the use of innovative items. To evaluate the degree to which these features and inferences can be supported by the items and tests, we seek to gather validity evidence to understand the following three questions: (1) In general, are students using the cognitive processes that are intended to be measured? (construct validity) (2) In general, for students who answer items wrong, does our analysis of why students might answer an item incorrectly match up with what was observed? (misconception analysis) and (3) Are students able to use the online assessment interface without difficulty? In other words, are the features that students will be asked to use, such as dropdown menus, coordinate graphs, and number line features intuitive or difficult to understand? (usability issues) Methodology: The methodology employed in this study is a hybrid of think-aloud and cognitive interview. The former is to record students' thought processes in real time, whereas the latter is a retrospective interview following the think-aloud protocol. Specifically, in this study, as a student says out loud his/her thought processes while responding to an item, administrators marks down his/her observation of the student's cognitive processes, reaction and engagement to the actual item. This process is followed by in-depth cognitive interviews, which are conducted by well-trained administrators following carefully-designed protocols. These protocols are designed to ask both item-specific and overall questions aimed to address the three research questions laid out above. We carefully selected 23 reading and 38 math forms for this study, with six items per form, ranging from grade 2 to grade 6. 47 school-age children of employees volunteered to participate, resulting in 82 sessions in total. Preliminary results Analysis results show that 91% of the students' cognitive processes matched the expectation of the CCSS, and 69% of the misconceptions matched or were related to the response analysis. Results also reveal some usability issues that might introduce construct-irrelevant variance into test scores. Specifically, students indicated they struggled with some features with some innovative item types, some online tools, and the navigation system. Educational significance: This study shows how think-alouds and cognitive interviews can be used to provide validity evidence regarding response process for a formative assessment. Findings can help inform decision-making regarding how to (1) better assess CCSS constructs, (2) provide more instructionally-relevant feedback, and (3) minimize usability issues associated with innovative items.