Los Primeros Acercamientos

According to the National Research Council’s report, it is possible to close and even

eliminate the achievement gap in US K12 science education (NRC, 2012). One way to improve all students learning documented by many is through formative assessment (Anderson & Palm, 2017; Black & Wiliam, 1998; Herman, 2013; Kingston & Nash, 2011). Despite this growing recognition for formative assessment, teachers are reluctant to practice it due to- 1) dominance of high-stake summative assessment (Furtak & Ruiz-Primo, 2008); 2) teachers lack the knowledge of creating assessment tasks and struggle with its implementation (Anderson & Palm, 2017; Darling-Hammond & Adamson, 2013; Heredia et al., 2016). Heritage (2008) argued that learning is a developmental process. Understanding how students may progress throughout a domain can help teachers develop formative assessment abilities (Bennett, 2011; Heritage, 2009; Herman, 2013). Therefore, teachers need knowledge of the pathway along which students evolve across the unit and where they are concerning the learning goals as they participate in practices to develop core ideas in science (NRC, 2012). However, there are no verifiable formative assessment strategies for improving teachers’ three-dimensional teaching and learning (Fick,

2017; Herman, 2013). Insight into how science teachers collect, analyze, and interpret data of students’ evolving sophistication of disciplinary core ideas is essential. The goal of this review

was to critically explore and synthesize current information on formative assessment and three- dimensional teaching and learning, to identify what is present in the literature and what needs further exploration. The analysis of literature was divided into two parts; the first section includes criteria for including and excluding articles, search engines, terminologies, and a table to provide a summary visual of the articles reviewed. The second section comprises the body of

the paper containing the literature discussed under the following headings- 1) Definition of formative assessment; 2) Embedded formative assessment task 3) Classroom conversations and response trajectory 4) Perception on and nature of feedback loop 5) Professional development and teacher change 6) Formative assessment in relation to teacher knowledge 7) Sequencing learning for formative assessment 8) Three-dimensional teaching and learning. This review concluded with significant findings, limitations of the studies and the gap, implications, and thoughts for the future, and suggested questions for further investigation.

Criteria for inclusion and exclusion

The articles included in this review met the following criteria

• Time frame was after Black and Wiliam’s (1998) seminal work (given that relevant sources were included in their review)

• Sources were primary studies reported by the original researchers

• Population of science teachers and their students and exception given to teachers and students of mathematics and other discipline if the study provides unique insight to FA. • The level spans from elementary to higher education, to expand the context and sources • Data type was either self-collected data or data as part of a team of a project

• Language of study was English

• Studies that reported findings on at least one feature of FA such as descriptive feedback, self-assessment, scaffolding, learning progression, goal and success criteria,

collaboration or classroom interaction and dialogue.

• Studies on professional development and others that enhance teachers’ practice and quality of FA, and studies that examine teachers’ thinking and practice of FA

• Studies relevant to FA, provided relevant insights, added to the academic discourse, and provided implication and recommendation for future research.

The Following Criteria Were Used to Exclude Articles

• Studies that discuss FA in general in other disciplines

• Studies that mainly compares FA to summative assessment with no input or output variables

• Literature reviews on FA although their reference lists served as sources for potential articles to be included

Search engine and terminology

Terminologies identified from these review articles included formative assessment, classroom assessment, assessment for learning, three-dimensional teaching/learning, and secondary science. These terms were used to search for articles in EBSCO Host-ERIC,

Academic Search Complete, Education Full Text, and Google Scholar. This review started with a Google scholar search for literature reviews on FA. The result yielded three original articles whose reference lists served as a starting point for relevant terminologies and article sources. A challenge with the article search using these key terms was that they are not used uniformly to address the interest of this paper. FA will be used as an umbrella term to include an assessment where the primary purpose in its design and enactment is to improve instruction and foster student learning (Black, Harrison, Lee, & Marshall, 2004).

[Table 14.- see Appendix]

Definitions of Formative Assessment

The use of formative assessment remains an mystery as many different definitions

be a persistent shortage of scientific evidence of its impact so long as vagueness in the constitutive and operational definition exist (Dunn & Mulvenon, 2009). Researchers use different terminologies to mean the same thing, and sometimes, the same vocabulary to say different things (Black & Wiliam, 1998; Meyer, 1992). A clear definition is essential to help researchers’ document effectiveness and compare its impact across studies and transfer the knowledge to other contexts. Early on, Bloom (1969) defined formative assessment as an

evaluative process “… to provide feedback and correctives at each stage in the teaching-learning process” (p. 48). Chappuis and Stiggins (2002) defined a formative assessment as an assessment designed to monitor students’ progress during the learning process. While Black and Wiliam (1998), described formative assessment as encompassing “all activities undertaken by teachers, and/or by their students, which provide information to be used as feedback to modify the teaching and learning activities in which they are engaged” (pp. 7-8). Assessment in their definitions was interpreted as instruments for collecting information about students’ progress during the learning process (Dunn & Mulvenon, 2009) and adjust instruction during instruction. A more inclusive definition by Chappuis et al. (2012) goes beyond looking at what happens during instruction to include teachers and students assessing students’ strengths and weaknesses, knowledge, and skills before, during, and after instruction. The Framework for K12 education’s goal provides a similar definition for assessment as, “an ongoing activity, one that relies on multiple strategies and sources for collecting information that bears on the quality of student work and that then can be used to help both the students and the teacher think more pointedly about how the quality might be improved” (NRC, 2001, p. 30). The teacher and students recognize and respond to the student’s learning, and the teacher adjusts instruction to enhance learning and teaching during learning and teaching (Bell & Cowie, 2001). The trend that keeps

repeating in the literature is that “it is not the instrument that is formative; it is the use of the information gathered, by whatever means, to adjust teaching and learning, that merits the “formative” label” (Chappuis, 2009, p. 4). Black and Wiliam (2009) argue that, a

practice in a classroom is formative to the extent that evidence about student

achievement is elicited, interpreted, and used by teachers, learners, or their peers, to make decisions about the next steps in instruction that are likely to be better, or better founded than the decisions they would have taken in the absence of the evidence that was elicited (p. 7).

Embedded Formative Assessment Tasks

Teachers have become experts in implementing standardized tests over the years, but the same does not apply to formative tasks. Model for teachers’ reasoning about assessment requires

a combination of cognitive, observational, and interpretive skills (NRC, 2001). The essence of creating assessment tasks is for teachers to explore different factors or reconstruct formative assessment strategies to accommodate their classroom style, to inform them about students’

evolving understanding continuously. The teacher assesses the gap and provides timely feedback to students on how to close the gap in their knowledge (Black & Wiliam, 2004a). These mini- assessments are coordinated with end-of-unit assessments to signal a unit’s organizational goal and give direction to the teacher (Shavelson et al., 2008). Thus, in creating new assessment tasks, the skills and knowledge incorporated should be that which is assessed (Black & Wiliam, 1998; Tomanek et al., 2007). Formative assessments tasks as fragments of formal assessments are embedded at strategic junctures in the curriculum or unit to create goals or sub-goals directed towards teachable moments before the student progresses to the next lesson (Yin et al., 2008).

The studies reviewed in this section highlight the complexity of constructing and challenges in implementing embedded formative assessments.

Wiliam et al. (2004) conducted a study to examine the achievement of secondary school students in classrooms where teachers made time to develop formative assessment strategies. In this study, 24 mathematics and science teachers were selected, trained with formative assessment strategies/techniques, and given a choice to choose which strategies to use. They assessed

students with assessment instruments. The authors acknowledged that the test lacked curricular validity (did not measure necessary concepts nor were aligned with the curriculum).

Observations and results of the interquartile range in effect size on test results revealed that teachers’ practices were slow to change, and any observable change was towards the end of the

year. Wiliam et al. (2004) attributed the quality of formative assessment to the teachers’ expertise level and suggested that using embedded formative assessment improved students’

achievement in externally mandated assessments. However, Wiliam et al. (2004) cautioned the acceptance of their results based on different units of comparison. They suggested further research on students’ achievement in classrooms where teachers formally embed assessments

into their lessons.

Yin et al. (2008) conducted a similar study to that of Wiliam et al. (2004) but embedded a formative assessment aligned with the curricular goal and summative assessment with a different outcome. Six experimental teachers taught a formative assessment embedded curriculum. The purpose of this quantitative exploratory study was to determine whether the embedded formative assessment task can improve students’ motivational beliefs, students’ achievement, and

conceptual change. The results of a motivational belief questionnaire and achievement

influence on students’ motivation, achievement, and conceptual change. Yin et al. (2008) noted

that trained teachers varied in degree of implementation of embedded formative assessment which, affected students’ outcomes differently. Their results revealed the difficulty of

implementing formative assessment even after providing teachers with training on how to use the information to improve teaching and student learning. They cautioned about solely embedding formative assessment into the curriculum without inviting teachers to participate in its design and provide follow up in-progress training.

In a similar study, Tomanek et al. (2007) explored science teachers' reasoning associated with task selection or evaluation of factors used as a planned formative assessment. The goal was to identify teachers' knowledge and beliefs that ground their assessment decisions. Data were collected from 24 first and 27 second-year teachers in a Science Teacher Preparatory Program (STPP) and 41 experienced science teachers who usually collaborate with the STPP faculty as mentors. They used formative assessment probes as a data collection instrument. However, the reliability and validity of their instrument was questionable and used as a pilot tool. Descriptive analysis of probes for teachers thinking revealed that task selection and evaluation was based on two themes: (1) "Characteristics of the tasks" and (2) "characteristics of students or the

curriculum" (Tomanek et al., 2007, p. 1119). They found no relation between being able to judge the level of thinking demanded of the task and teachers' training or experience. However,

selecting tasks based on students or curriculum characteristics was related to teaching

experience. They implied that (1) teachers reasoned with factors that sometimes work against the selection of tasks with the potential of assessing students' understanding of concepts. (2)

Prospective and experience teachers must be afforded the opportunity to question their beliefs about assessment and interpretation of evidence of student understanding.

These studies highlight teachers' reasoning about selecting assessment tasks and the challenges in creating and implementing embedded formative assessment. Wiliam et al. (2004) and Yin et al. (2008) findings on the relationship between teacher enactment and student achievement indicated that teachers had variable expertise and affected students' success differently. While Wiliam et al. (2004) reported a significant improvement in students' achievement with embedded assessment tasks, Yin et al. reported the contrary. Both studies found that experience alone is not enough for the effective implementation of formative

assessment. Tomanek et al. (2007) reported that training or experience had no relationship with being able to judge the level of thinking of tasks. These studies suggest that teachers need more training in constructing or participation in co-constructing and enacting formative assessment tasks.

Complexity of Classroom Conversations and Response Trajectory

Evidence in learning science indicates that affinity is more towards talking science than reading science (Lemke, 1990; Roth, 2005). The dynamics of talk provides an interactive medium for exploring and knowing about the world scientifically. Therefore, whole-class discussions create opportunities for students and peers to question and provide a rationale for scientific claims and for the teacher to solicit, monitor, and enhance students learning. Classroom talk is vital to many approaches to learning but difficult to coordinate because of the challenges to manage rich classroom discourse and inquiry activity in unison (Anderson et al., 2007). These daily assessment conversations or instructional dialogues were embedded in an activity currently taking place in the classroom. It can allow the teacher to gather information about the status of students’ conception, language use or communication skills, mental models, or used as strategies

students’ complete conceptions to the whole class. Adding another component of curriculum-

embedded assessment as written explanation (silent talk) could provide a broader definition of feedback that addresses each student’s need (Furtak & Ruiz-Primo, 2008). Thus, formative

assessment as a tool can be employed at any level of student-student-teacher interaction during daily classroom talk to improve students’ conceptual understanding and assist teachers

continuously gain insight about students’ level of understanding (Bell & Cowie, 2001; Furtak &

Ruiz-Primo, 2008). This section described studies that demonstrate the impact of formal (planned) and informal (unplanned) formative assessment in the classroom using various kinds of discourses.

Focusing on discussions between students,Anderson et al. (2007) conducted a study to understand how research efforts to document discursive classroom routines informed the process of advancing participation in scientific inquiry and performance on high-stake achievement. Using a situated lens, Anderson et al. (2007) explored “individual student performance on

individual test items as examples of specific types of discourse, allowing a coherent examination of transfer of understanding across very different ways of knowing” (p. 1742). Videotape

discussions from groups of 11th _{and 12}th _{grade students and their pre-and posttest results were} analyzed. The results highlighted students’ engagement in classroom conversation, the role of

answer rubrics, and the teacher facilitation that was better coordinated to scaffold more

productive discursive trajectory classroom talk in the second year. Anderson et al. (2007) noted an improvement in teacher providing informal feedback during classroom conversations and more productive discussions with the use of answer rubrics. Also, a varying degree of learning was reflected in group discourse on quiz feedback and on gains in the examination and test. The implication from this study is that the role and nature of group discourse couple with teacher

intervention affect the quality of students’ movement along the discursive trajectory from

informal formative feedback conversation to more formal assessment activities.

Looking at discourses between students and teachers, Ruiz-Primo and Furtak (2007) explored how students develop an understanding of concepts during daily whole-class

conversations. They used the Elicit, Student respond, teacher Recognize and Use (ESRU) model to distinguish the quality of informal assessment practices across teachers and to determine whether this quality can be linked to student performance. Videotapes of classroom

conversations were collected from three middle school teachers trained in the implementation of the FAST (Foundational Approach for Science Teaching) curriculum. The analysis of discourse transcripts indicated a range of informal assessment frequencies, from incomplete (ERS or IRE/F) to complete conversation cycles (ESRU). They made the inference that better informal assessment practices could be linked to better student performance, and the ESRU model was useful in capturing differences in teachers’ informal assessment practices. However, this

conclusion is drawn from data for only a single teacher.

Furtak and Ruiz-Primo (2008), in a similar study, added a written portion to students’ responses to classroom talk to capture students’ complete conception of knowledge status, and

compare this status with the learning goal. Their study examined the relative utility of the formal and informal functions of four types of formative assessment prompts in eliciting middle school students’ ideas about sinking and floating through written responses and classroom discussions.

Videotape of written assessment and discussion implementation were collected from four Romance project teachers. Data analysis indicated that all prompts elicited a high percentage of students’ ideas at the expected levels, but below-level conceptions were expressed more in

writing as compared to discussions. Many more students provided responses containing multiple conceptions in writing as compared to students’ responses in whole-class discussions, and some

prompts were more successful than others in eliciting a range of students’ ideas. Furtak and Ruiz-Primo (2008) noticed that teachers were not using whole-class settings efficiently to elicit students’ conceptions and suggested that teachers need more familiarity with prompts for

effective implementation. They recommended that teachers should base their judgment on students’ conception of both written work and classroom discussion. Besides, future research

should explore in greater depth ways teachers implement different assessment prompts and the extent of feedback provided to students about the conceptions elicited in the prompts.

Ruiz-Primo and Furtak (2006) compared how four middle school science teachers from the same experimental group used questioning as an informal formative assessment method to measure students’ learning. The purpose was to understand how informal assessment looks like in the context of scientific inquiry, the different levels of informal assessment practices, and whether these different levels were linked to levels of student learning. Data were collected from videotapes of classroom conversations and responses from pre-posttest assessments, and

embedded assessments prompt. The analysis results indicated that 1) teachers used complete cycles only 26% of the time and most cycles were 95% epistemic in nature 2) the pattern of change in students’ post-test results reflected a change in teacher’s informal formative

assessment practice 3) teachers using complete ESRU cycle had students with high performance. Ruiz-Primo and Furtak (2006) conclude that it is important to provide teachers with tools with which to respond in immediate and effective ways. They suggested that future studies were needed to revise scientific inquiry domains but did not indicate challenges/limitations with their study. Also, future studies could use ESRU coding techniques to explore discourse in the context

of embedded assessment. Ruiz-Primo and Furtak (2006) asserted that evidence from their study could be used to design assessment courses for preservice teachers and professional development on informal formative assessment strategies.

Hickey et al. (2012) study intended to promote meaningful participation in the discursive construction of shared domain knowledge and improved achievement with a design strategy made up of three different levels of assessment (informal/close, semi-formal/proximal, and formal/distal level). Through a situated lens, Hickey et al. (2012) rationalized that using different forms of assessment makes using and aligning them easy, such that an assessment can serve a formative function in one situation and summative function in another. Data were collected through informal observations and videotapes of feedback conversations then aligned to establish