Week 12: Quantitative research designs SOWK 621: DeCarlo Contents • Video mini-lectures • Survey design • Experimental design • Design & methods • Virtual office hours • Questions This Photo by Unknown author is licensed under CC BY-NC-ND. Survey design Video mini-lecture Design: What your study will do • You’ve read your literature (lit review) • You’ve created your question (research question) • You’ve got your people (sampling) • You’ve figured how to measure your variables (measures) • Now you need to figure out what you’ll do with your participants when they come into your study • This is called design • Quantitative: experiment, survey, secondary data analysis, chart reviews, etc. • Qualitative: interview, focus group, secondary data analysis, etc. Strengths Strengths and limitations of surveys Limitations Cost-effectiveness Inflexibility Generalizability Lack of depth Reliability For cross-sectional designs, difficulty with time order Versatility Longitudinal designs are great Phone, mail, internet, and inperson surveys all have issues Time Time & administration: Types of surveys • Longitudinal: measure people over time • Trend, panel, cohort, retrospective • Cross-section: measure only one time Administration • Phone • In-person • Electronic • Mail • BRUSO: “brief,” “relevant,” “unambiguous,” “specific,” and “objective.” • Questions are based on operational definitions • But also include other variables and characteristics Questionnaires: A tool for surveys • Concise, easy to understand • Address the “most knowledgeable people” about a topic • Think back to sampling • Clear wording • Double negatives, double-barreled questions and answers, jargon, slang • Neutral wording • Leading language, social desirability • Pretesting is key Questionnaires: A tool for surveys • Closed-ended questions • Mutually exclusive and exhaustive response options • Fence-sitters and floaters • Filter questions, Matrix questions • Group your questions by theme • Ordering is important, though tricky • Think about the time needed to complete the questionnaire • Look professional Experimental design Video mini-lecture Experimental design: Probably not your design • It is my design, though • I have an intervention to test • Program evaluation is similar to experimental design Experimental design: Step by step Logic of experimental design • Maximize internal validity • Changes in DV are the result of IV (intervention) • Control for confounding/extraneous variables • Laboratory settings, treatment fidelity • Consistent treatment • Control vs. experimental group • Comparable groups • Random assignment • Avoid selection bias • Blinded • Avoid placebo effect, resentful demoralization, compensation • Pretest/posttest • Establish baseline, time order Experimental designs vary in rigor • Classic experimental designs (very rigorous) • Random assignment, control & experimental group, pretest & posttest • Many variables (dissembling, alternate treatment, etc.) • Quasi-experimental designs (somewhat rigorous) • No random assignment, may use comparison group or match cases • Time-series design • Non-experimental designs (not that rigorous) • No random assignment, may not have comparison group, or pretests • Still valuable Design & methods (Quant) Video mini-lecture Let's look at the Design & Methods (proposal) • I filled it out. And I'll walk you through it. Virtual office hours Design & methods Agenda • Calendar • Questions • Design & methods Calendar • Questions? Design & Methods • What have I seen so far?