Session 2: Specifying the Conceptual and Operational Models and the Research Questions that Follow Mark W. Lipsey Vanderbilt University IES/NCER Summer Research Training Institute, 2008 Workshop on randomized controlled trials • Purpose: Increasing capacity to develop and conduct rigorous evaluations of the effectiveness of education interventions • Caveat: “Rigorous evaluations” are not appropriate for every intervention or every research project involving an intervention – They require special resources (funding, amenable circumstances, expertise, time) – They can produce misleading or uninformative results if not done well – The preconditions for making them meaningful may not be met. Critical preconditions for rigorous evaluation • A well-specified, fully developed intervention with useful scope – – – – basis in theory and prior research identified target population specification of intended outcomes/effects “theory of change” explication of what it does and why it should have the intended effects for the intended population – operators’ manual: complete instructions for implementing – ready-to-go materials, training procedures, software, etc. Critical preconditions for rigorous evaluation (continued) • A plausible rationale that the intervention is needed; reason to believe it has advantages over what’s currently proven and available • Clarity about the relevant counterfactual– what it is supposed to be better than • Demonstrated “implementability”– can be implemented well enough in practice to plausibly have effects • Some evidence that it can produce the intended effects albeit short of standards for rigorous evaluation Critical preconditions for rigorous evaluation (continued) • Amenable research sites and circumstances: – cooperative schools, teachers, parents, and administrators willing to participate – student sample appropriate in terms of representativeness and size for showing educationally meaningful effects – access to students (e.g., for testing), records, classrooms (e.g., for observations) IES funding categories • Goal 2 (intervention development) for advancing intervention concepts to the point where rigorous evaluation of its effects may be justified • Goal 3 (efficacy studies) for determining whether an intervention can produce worthwhile effects; RCT evaluations preferred. • Goal 4 (effectiveness studies) for investigating the effects of an intervention implemented under realistic conditions at scale; RCT evaluations preferred. Specifying the theory of change embodied in the intervention 1. Nature of the need addressed – what and for whom (e.g., 2nd grade students who don’t read well) – why (e.g., poor decoding skills, limited vocabulary) – where the issues addressed fit in the developmental progression (e.g., prerequisites to fluency and comprehension, assumes concepts of print) – rationale/evidence supporting these specific intervention targets at this particular time Specifying the theory of change 2. How the intervention addresses the need and why it should work – – – content: what the student should know or be able to do; why this meets the need pedagogy: instructional techniques and methods to be used; why appropriate delivery system: how the intervention will arrange to deliver the instruction Most important: What aspects of the above are different from the counterfactual condition What are the key factors or core ingredients most essential and distinctive to the intervention Logic models as theory schematics Target Population Intervention Proximal Outcomes Distal Outcomes Positive attitudes to school 4 year old pre-K children Exposed to intervention Improved pre-literacy skills Learn appropriate school behavior Increased school readiness Greater cognitive gains in K Mapping variables onto the intervention theory: Sample characteristics Positive attitudes to school 4 year old pre-K children Exposed to intervention Sample descriptors: basic demographics diagnostic, need/eligibility identification nuisance factors (for variance control) Improved pre-literacy skills Learn appropriate school behavior Increased school readiness Greater cognitive gains in K Potential moderators: setting, context personal and family characteristics prior experience Mapping variables onto the intervention theory: Intervention characteristics Positive attitudes to school 4 year old pre-K children Exposed to intervention Independent variable: T vs. C experimental condition Generic fidelity: T and C exposure to the generic aspects of the intervention (type, amount, quality) Improved pre-literacy skills Learn appropriate school behavior Increased school readiness Greater cognitive gains in K Specific fidelity: T and C(?) exposure to distinctive aspects of the intervention (type, amount, quality) Potential moderators: characteristics of personnel intervention setting, context e.g., class size Mapping variables onto the intervention theory: Intervention outcomes Positive attitudes to school 4 year old pre-K children Exposed to intervention Focal dependent variables: pretests (pre-intervention) posttests (at end of intervention) follow-ups (lagged after end of intervention Improved pre-literacy skills Learn appropriate school behavior Increased school readiness Greater cognitive gains in K Other dependent variables: construct controls– related DVs not expected to be affected side effects– unplanned positive or negative outcomes mediators– DVs on causal pathways from intervention to other DVs Main relationships of (possible) interest • Causal relationship between IV and DVs (effects of causes); tested as T-C differences • Duration of effects post-intervention; growth trajectories • Moderator relationships; ATIs (aptitude-Tx interactions): differential T effects for different subgroups; tested as T x M interactions or T-C differences between subgroups • Mediator relationships: stepwise causal relationship with effect on one DV causing effect on another; tested via Baron & Kenny (1986), SEM type techniques. Formulation of the research questions • Organized around key variables and relationships • Specific with regard to the nature of the variables and relationships • Supported with a rationale for why the question is important to answer • Connected to real-world education issues • What works, for whom, under what circumstances, how, and why? Session 3: Describing and Quantifying Outcomes Mark W. Lipsey Vanderbilt University IES/NCER Summer Research Training Institute, 2008 Outcome constructs to measure Identifying the relevant outcome constructs follows from the theory development and other considerations covered earlier in Session 2 – What: proximal/mediating and distal outcomes – When: temporal status– baseline, immediate outcome, longer term outcomes – What else: • possible positive or negative side effects • construct control outcomes not targeted for change Aligning the outcome constructs and measures with the intervention and policy objectives Instruction Assessment Policy relevant outcomes (e.g., state achievement standards) Alignment of instructional tasks with the assessment tasks Instructional tasks, activities, content Identical Analogous (near transfer) Generalized (far transfer) Basic psychometric issues Validity (typically correlation with established measures or subgroup differences) Reliability (typically internal consistency or test-retest correlation) – standardized measures of established validity and reliability – researcher developed measures with validity and reliability demonstrated in prior research – new measures with validity and/or reliability to be investigated in present study Special issue for intervention studies: sensitivity to change Achievement effect sizes from 97 randomized education studies Type of Outcome Mean Effect Measure Size Standardized .09 test, broad Number of Measures 29 Standardized test, narrow .32 127 Focal topic test, mastery test .50 263 Data from which measurement sensitivity can be inferred • Observed effects from other intervention studies using the measure • Mean effect sizes and their standard deviations from meta-analysis • Longitudinal research and descriptive research showing change over time or differences between relevant criterion groups • Archival data allowing ad hoc analysis of, e.g., change over time, differences between groups • Pilot data on change over time or group differences with the measure Variance control and measurement sensitivity Variance control via procedural consistency and statistical control using covariates for e.g., pre-intervention individual differences and differences in testing procedures or conditions Issues related to multiple outcome measures Correlated measures: overlap and efficiency Factor Analysis of Preschool Outcome Variables Factor Loadings Subtest Letter Word Identification Quantitative Concepts Applied Problems Picture Vocabulary Oral Comprehension Story Recall Pre-K Pretest Pre-K Posttest Kindergarten Follow-up .60 .82 .82 .75 .82 .53 .69 .82 .80 .76 .79 .55 .73 .78 .75 .67 .74 .61 Correlated change may be even more relevant Factor Analysis of Gain Scores for Pre-K Outcomes Factor Loadings Subtest Pre to Post Post to Follow-up Pre to Follow-up Basic School Skills Letter Word Identification Quantitative Concepts Applied Problems .74 .66 .54 -.19 .14 .08 .73 .70 .47 -.06 .06 .16 .79 .74 .40 -.15 .13 .41 Complex Language Picture Vocabulary Oral Comprehension Story Recall .09 .16 -.08 .77 .75 .37 .14 .17 -.16 .48 .72 .68 -.04 .13 -.01 .74 .69 .37 Handling multiple correlated outcome measures • Pruning– try to avoid measures that have high conceptual overlap and are likely to have relatively large intercorrelations • Procedural– organize assessment and data collection to combine where possible for efficiency • Analytic – create composite variables to use in the analysis – use multivariate techniques like MANOVA to examine omnibus effects as context for univariate effects – use latent variable analysis, e.g., in SEM Practicality and appropriateness to the circumstances • Feasibility– time and resources required • Respondent burden– minimize demands, provide incentives/compensation • Developmental appropriateness– consider not only age but performance level, possible ceiling and floor effect • For follow-up beyond one school year, may need measures designed for a broad age span to maintain comparability • May need to tailor measures or assessment procedures for special populations (disabilities, English language learners)