Questions, Claims and Evidence: Teaching Argumentation in the NGSS through the use of a Science Writing Heuristic Brian Hand University of Iowa Discussion format Theoretical perspectives Randomized field trail results Classroom aspects Next Generation Science Standards Scientific and Engineering Practices 1. Asking questions (for science) and defining problems (for engineering) 2. Developing and using models 3. Planning and carrying out investigations 4. Analyzing and interpreting data 5. Using mathematics and computational thinking 6. Constructing explanations (for science) and designing solutions (for engineering) 7. Engaging in argument from evidence 8. Obtaining, evaluating, and communicating information Science The advancement of science is about a process of construction and critique Scientist negotiate with each other Scientists do not advance science through information transfer the – if this was the case who gave the first scientists information to pass on? - who gives the current generation of scientist the “new” knowledge Science Argument Is a central component of science Requires participants to negotiate meaning both publicly and privately Is bound by a structure linking questions, claims, evidence and rebuttals Constructed knowledge is tested against nature Our working definition Argument Based Inquiry is inquiry that is intended to build students grasp of scientific practices while motivating an understanding of disciplinary big ideas. Construction and critique of knowledge are centrally located through an emphasis on the epistemological frame of argument by engaging them in posing questions, gathering data, and generating claims supported by evidence. This perspective of argument builds on the work of Walton who suggests that argument is a logical contribution to the resolution of unsettled knowledge. This more general perspective on argument is valuable as it recognizes the use of argument as a learning tool; thus the immersion of students in argument throughout their inquiry. Douglas Walton University of Toronto Argument Deals with unsettled knowledge Trying to persuade others Explanation Deals with settled knowledge To inform others Argument Made up of questions, claims and evidence What is a claim? What is evidence? What is the relationship between these elements? Connections Question Claims Claims Evidence Argument Deals with questions, claims and evidence There must be connections between questions, claims and evidence There has to be strong coherence between the various components Arguments require reasoning – not something to be simply learned Critical Issues that need to be engaged with We do not pay enough attention to ideas such as data, evidence, explanation Researchers use such ideas – as published in articles Data/evidence Claims, evidence, reasoning Evidence and explanations Data and evidence Is this distinction important? Students have trouble separating these two “Data does not speak” We have to do something with data to get to evidence Evidence and Reasoning If we have to do something to data to get evidence – what is it? Critically we have to reason about the data – we have to make critical decision about What data points to use? Are there patterns? • If we remove reasoning from evidence we have data Relationship Data Reasoning Evidence Evidence and explanation Simply question – if evidence is not an explanation – what is it? Is not evidence a reasoned explanation about particular data points and how they fit together? All evidence is explanatory, but not all explanations are evidentiary Two essential components of science Language - there is no science without language - means that we have have to pay attention to all the different discourses/representations associated with science Argumentation - is a critical process that is central to the way in which science knowledge is constructed The Science Writing Heuristic approach is based on earlier Halliday work (70’s) You learn about language while you learn through using language Means that students learn about argument while they learn through using argument Importance? These distinctions are not trivial There is a distinctly different orientation to the learning of argument based inquiry Is it something “done to” students or something students should “be immersed” in? The Science Writing Heuristic Templates Teacher’s template Exploration of pre-instruction understanding Pre-laboratory activities Laboratory activity Negotiation I - individual writing Negotiation II - group discussion Negotiation III - textbook and other resources Negotiation IV- individual writing Exploration of post-instruction understanding Student’s template Beginning questions or ideas What are my questions about this experiment? Tests and Procedures What will I do to help answer my questions? Observations What did I see when I completed my tests and procedure? Claims What can I claim? Evidence What evidence do I have to support my claim? How do I know? Why am I making these claims? Reading How do my ideas compare with others? Reflection How have my ideas changed? Randomized Field Trial Involves 48 grade 3-5 buildings in Iowa 24 treatment, 24 control Divided into 5 clusters within the state 8 days of inservice – 5 in summer, 3 during the year Follow up monitoring within school Collection of teacher video – one per semester/year Collection of Iowa Test of Basic Skills/Iowa Assessment data Implementation of Cornell Critical Thinking test pre/post at grade 5 level Critical Thinking Improvement Scores Year 1 Critical Thinking Improvement Scores Year 2 Effect sizes for each cluster Transfer – what do we mean Domain Specific Knowledge Domain General Knowledge Transfer Argumentbased Inquiry Cornell Critical Thinking Test Data Used The data used for the following analysis' are the paired 3rd - 5th grade and paired 4th - 6th grade national standardized scores on the ITBS and Iowa Assessments for school year pairs 2006-07 with 2008-09, 2007-08 with 2009-10, 2008-09 with 2010-2011, and 2009-10 with 2011-12. The data is paired by student The associated demographics are also used Equivalence of ITBS and Iowa Assessments ITBS and Iowa Assessments share three sections that were taken by all schools in the study: Reading, Mathematics, and Science. The Math I and Math II Scores from the ITBS relate with the Math Comprehension score from Iowa Assessments For each combination of subject, grade, and year the National Standardized Scores were standardized. For the ITBS Math Scores they were added, then standardized Mixed Models For each of the four groups of students as described previously, for each of the three test types (RC, M, and SC), mixed models were fit The predicted value was the change in test score (DRC, DM, or DSC) Note: Scores standardized as previously mentioned The base-level fixed effects were pretest score (RC, M, or SC), ASN, BLK, HSP, FRL, ELL, GAT (All students only), SED (similarly), and TRT Mixed Models Continued The interactions included in the models were TRT with all the other fixed effects. The two random effects included in the modes were UNIT (the unit for which treatment or control was assigned, usually a school building) and DIST (the block of units used to control variability in the assignment of treatment and control) For each model non-significant variables were removed (tscore < 1.66 (relating p-value .1)) unless they were base fixed effects who had a significant interaction term or the TRT fixed effect. All Students - Reading DRC ~ RC + TRT + SEM + DSEM + ASN + BLK + SED + GAT + FRL + ELL + (1|UNIT) + (1|DIST) Notes TRT - Weak Positive (t = 1.66) No significant interactions All Students - Mathematics DM ~ M + TRT + DSEM + GEN + ASN + BLK + HSP + SED + GAT + FRL + ELL + TRT:GEN + TRT:BLK + TRT:SED + TRT:GAT + TRT:FRL + TRT:ELL + (1|UNIT) + (1|DIST) Notes TRT – Very Very Strong Positive (t = 11.23) TRT:GEN – Weak Negative (t = -2.03), TRT:BLK – Weak Positive (t = 1.97), TRT:SED – Very Strong Positive (t = 3.44), TRT:GAT – Very Very Strong Negative (t = -7.34), and TRT:ELL – Weak Postive (t = 2.10) All Students - Science DSC ~ SC + TRT + DSEM + GEN + BLK + HSP + SED + GAT + FRL + ELL + (1|UNIT) + (1|DIST) Notes Not significant TRT No significant interactions All Students - Comments There is small evidence to support students do better with Reading Comprehension in the Treatment group There is very strong evidence to support that students do better with Mathematics in the Treatment group The female disadvantage is reduced with treatment The African American disadvantage is reduced with treatment The special education disadvantage is reduced with treatment The gifted and talented advantage is reduced with treatment The free and reduced lunch disadvantage is reduced with treatment English language learners in treatment are as well off as non-English Language Learners in the Control group There is no evidence to support improvement is Science Comprehension associated with Treatment Special Education Conclusions There is no evidence to support that treatment leads to a change in Reading Comprehension Males receive an advantage in Reading Comprehension with treatment Free and reduced lunch students receive a further disadvantage with treatment There is very strong evidence to support that treatment leads to an improvement in Mathematics There is no evidence to support that treatment leads to a change in Science Comprehension Free and reduced lunch students receive a further disadvantage with treatment Gifted and Talented Conclusions There is no evidence of an effect of Treatment on Reading Comprehension There is weak evidence to suggest a negative effect of Treatment on Mathematics The female disadvantage is reduce in the treatment group to the point were a female in the treatment group is better off then in the control group The free and reduced lunch disadvantage is reduced in the treatment group to the point where a FRL student in the treatment group is better off then in the control group There is no evidence of an overall effect of Treatment on Science Comprehension The Hispanic advantage almost disappears in the treatment group Traditional Conclusions No evidence of an overall effect of Treatment on Reading Comprehension Treatment seems to decrease the Asian Advantage Treatment seems to decrease the free and reduced lunch disadvantage There is strong evidence of an improvement of Mathematics scores because of Treatment An African American in the treatment group has the same advantage as a Caucasian in the control group An English Language Learner in the treatment group is better off then a nonELL in the control group A free and reduced lunch student in the treatment group is almost as well off as a non-FRL in the control group No evidence of an overall effect of Treatment on Science Comprehension The Asian advantage decreases with the Treatment group Classroom Conditions How well do teachers implement? Is there a difference between treatment and control? Do they have more science teaching in school? Is there a carry over to other subjects? Can we see a shift in how the classroom environment looks? Does this impact language events? Teacher Implementation Scores Year 1 120 100 # of Videos 80 TRT 60 CTRL 40 20 0 0 - .25 0.5 - 0.75 1 - 1.25 1.5 - 1.75 2 - 2.25 2.5 - 2.75 3 Number of lessons per week Minutes/science lesson 60 50 % teachers 40 TRT Percent 30 CTRL Percent 20 10 0 0-15 16-30 31-45 46-60 60+ Transfer of approach into other disciplines Classroom environment Teacher-student talk Writing Does implementation matter? Do we understand what is happening? Sources of meaning Personal (Intuitive) Complexity of Reasoning Developing Single Reasoning Fuzzy Understanding Diagrammatic Representation Contextualized/ Perception-based Single Reasoning Alternative Explanation Developing Chain of Reasoning Comparing Ideas Chain of Reasoning Consolidating Ideas Developing Reasoning Network Further Negotiation Scientific (Reflective) Coherence Reasoning Network Research funded by a grant from the US Department of Education through the Institute of Education Sciences, award number R305A090094-10.