Outline• Definitions of Research/Evaluation• Purposes of Evaluation / Research• Science and Scientific Management• Research/Evaluation as Process• Types of Research / Evaluation
Definitions• Evaluation = Process of judging the merit or worth of something• Research – application of scientific methods to answer questions – controlled inquiry directed at increasing knowledge/establishing truth• Evaluation Research - combine the two
Science• Body of Knowledge • Method of Inquiry• systematic • logical• abstract – induction• – deduction general• parsimonious • self-corrective • empirical
Scientific Management• Application of scientific principles to management and decision making – systematic information gathering – empirical, objective, self-corrective
Process -- Steps Research Evaluation• define problem • describe program• objectives/hypotheses • evaluation criteria• literature review • program scoping• research methods • evaluation methods• gather data/analysis • gather data/analysis• conclusions • conclusions
Types of Evaluation by Program Stage• formative (conceptualization/design)]• process (implementation)• summative (outcomes, impacts, efficiency)
Types - By Approach• Standards – norm-based – criterion-referenced• Goals and objectives• Impacts or effects
Evaluation Criteria• Effort - qnty and qlty of inputs• Performance - qnty and qlty of outputs• Adequacy - meet needs?• Efficiency - benefits/costs• Equity - distributional issues, fairness
Process Evaluation• Identifies how and why program works – attributes – recipients – conditions – effects • single or multiple • intended or side effects • timing & duration, long/short term • cognitive, affective or behavioral
Research Process Define Problem, Research Objectives HOW? What? Overall Method Who?•Concepts •Survey •Experiment •Population•Variables •Case Study •Sampling •Secondary Data•Measures Data Gathering Analysis Application
Proposal Format 1. Problem Statement - define program to be evaluated/problem to be studied, users & uses of results. Justify importance of the problem/study. 2. Objectives : Concise listing . In evaluation studies, the objectives usually focus on the key elements of program to be evaluated & the evaluation criteria. These are the study objectives NOT the program objectives. 3. Background/Literature Review - place for more extensive history/structure of program. Focus on aspects most relevant to proposed evaluation. Discuss previous studies or the relevant methods. 4. Methods - details on procedures for achieving objectives - data gathering and analysis, population, sampling, measures, etc. Who will do what to whom, when, where, how and why? 5. Attachments - budget, timeline, measurement instruments, etc.NOTE: Most “programs” must be narrowed to specific components to be evaluated.Think of a “Program of studies” rather than a single evaluation study. The proposalshould define this specific study & how it fits into a broader program of studies.
Purposes of Proposal• Communicate with Client• Demonstrate your grasp of problem• Plan the study in advance, so others can evaluate the study approach – will it work? – have you overlooked something? – will results be useful to client? – Can we afford it?
Sample Objectives1. Estimate benefits and costs of program2. Estimate economic impacts of program on local community (social, environmental, fiscal).3. Determine effects of program on target population.4. Describe users and non-users of program5. Assess community recreation needs, preferences6. Determine market/financial feasibility of program7. Evaluate adequacy or performance of program
Methods Choices• Overall Approach/Design – Qualitative or Quantitative – Primary or secondary data – Survey, experiment, case study, etc.• Who to study - population, sample – individuals, market segments, populations• What to study - concepts, measures – behavior, knowledge, attitudes• Cost vs Benefit of Study
Definition & Measurement“measurement is the beginning of science, … until youcan measure something, your knowledge is meager andunsatisfactory” Lord Kelvin Nominal/Conceptual Definition - define concept in terms of other concepts, links concepts without tying them to real world Operational definition - equates definition with measurement, specify procedures/operations to generate the concept.
Levels of MeasurementLevel Characteristic ExampleNominal Unordered Race, gender categoriesOrdinal Ordered categories Sm, med.lg Hardness scaleInterval Consistent distance Temp in fahrenheit between categories or CelsiusRatio Natural zero Temp in Kelvin
Sampling• Always define study population first• Use element/unit/extent/time for complete definition • element - who is interviewed • sampling unit - basic unit containing elements • extent - limit population (often spatially) • time - fix population in time
Types of Sampling Approaches• Probability vs non-Probability• Judgment, Simple Random, Systematic• Stratify or Cluster (Area Sample)• Time Sampling
Sample size• Based on four factors • Cost/budget • Accuracy desired • variance in popln on variable of interest • subgroup analysis planned• Formula: n= Z2 σ2 / e 2 • n= sample size • Z indicates confidence level (95% = 1.96) ∀ σ = standard deviation of variable in population • e = sampling error
Computing 95% confidence interval• N= 100 , sample mean = 46%, use p= 50/50,• sampling error from table = 10%• 95% CI is 46% + or - 10% = (36, 56)• N=1,000 sample mean =22%• sampling error from table = 2.5%• 95% CI is 22% + or - 2.5% = (19.5, 24.5)
Research Designs/Data Collection ApproachesHow ....Where Household On-Site LaboratoryGatheredPersonal Surveys Surveys, Focus GroupsInterview Field ExpmtsTelephone/ Surveys Computer ComputerComputer Interviews InterviewsSelf-Admin. Surveys, ExperimentsQuest. Field ExpmtsObservation NA Observable Observable& Traces Characteristics CharacteristicsSecondary NA Internal NASources Records
Major Design Types• Surveys• Experiments• Observation• Secondary Data• Qualitative Approaches – Focus Group – Case Study
General Guidelines on when to use different approaches1. Describing a population - surveys2. Describing users/visitors - on-site survey3. Describing non-users, potential users or general population - household survey4. Describing observable characteristics of visitors - on-site observation5. Measuring impacts, cause-effect relationships - experiments
Guidelines (cont)6. Anytime suitable secondary data exists - secondary data7. Short, simple household studies - phone8. Captive audience or very interested population - self-administered survey9. Testing new ideas - experimentation or focus groups10. In-depth study - in-depth personal interviews, focus groups, case studies
Primary or Secondary Data• Secondary data are data that were collected for some purpose other than your study, e.g. government records, internal documents, previous surveys• Choice between Primary /Secondary Data – Costs (time, money, personnel) – Relevance, accuracy, adequacy of data
Qualitative vs Quantitative Quantitative QualitativePurpose Gen’l Laws Unique/Individual case Test Hypotheses Understanding Predict behavior Meanings/IntentionsPerspective Outsider-Objective Insider-SubjectiveProcedures Structured Unstructured formal measures open ended measures probability samples judgement samples statistical analysis interpretation of data
Qualitative vs Quantitative Approaches Qualitative Focus Group In-Depth Interview Case Study Participant observation Secondary data analysis Quantitative Surveys Experiments Structured observation Secondary data analysis
Survey vs ExperimentSurvey - measure things as they are, snapshot of population at one point in time, generally refers to questionnaires (telephone, self-administered, personal interview)Experiment - manipulate at least one variable (treatment) to evaluate response, to study cause-effect relationships (field and lab experiments)
STEPS IN A SURVEY1. Define problem and study objectives2. Identify information needs & study population(s)3. Determine basic design/approach - cross sectional vs longitudinal - on-site vs household vs other - self-admin. vs personal interview vs phone - structured or unstructured questions4. Questionnaire design5. Choose sample (frame, size, sampling design) 6. Estimate time, costs, manpower needs, etc.
Survey Implementation7. Proposal & “Human subjects” review8. Line up necessary resources9. Pre-test instruments and field procedures10. Data gathering and follow-up procedures11. Coding, cleaning and data processing12. Analysis: preliminary, then final.13. Communication and presentation of results.
Characteristics of a true Experiment 1. Sample equivalent experimental and control groups 2. Isolate and control the treatment 3. Measure the effect
Pre-test/Post-test with ControlR MB1 X MA1 Experimental groupR MB2 MA2 Control group R denotes random assignment to groups X denotes the treatmentMeasure of effect = ∆ Expmt gp - ∆ Control gp = (MA1-MB1) - (MA2-MB2) = with vs without
Threats to Internal validity• * Pre-measurement (Testing) : effect of pre- measurement on dependent variable (post-test)• * Selection: nonequivalent experimental & control groups, (statistical regression a special case)• * History: impact of any other events between pre- and post measures on dependent variable• * Interaction: alteration of the “effect” due to interaction between treatment & pre-test.• Maturation: aging of subjects or measurement procedures• Instrumentation: changes in instruments between pre and post.• Mortality: loss of some subjects
Three Audiences/styles• Researchers – research journal style – Technical, methods, statistical tests• Managers – business style – Results and implications• Public – newspaper style – Interesting, no jargon, highlights
Research vs Business Reports• Written/Research • Oral/Business – Problem – Objectives – Objectives – Key Results & – Methods Recommendations – Results – Justify from study – Discussion – Brief methods – Discussion
Reminders• Final Exam is Friday Dec 15, 7:45-9:45 am, this room• Final Papers due by Wednesday Dec 13• See YaYen Sun to finish lab work by end of week.