GU Research Mod 7 Ch 9

The flashcards below were created by user MeganM on FreezingBlue Flashcards.

  1. What would have happened to the same people exposed to a causal factor if they simultaneously were NOT exposed to the causal factor.
  2. Involves DOING something to study participants.
  3. Experimenters manipulate the independent variable by administering a _______ to some and withholding it from others, or administering a different ________.
    • treatment (intervention)
    • treatment
  4. When is an approach with different doses or intensities of tx where all participants get some tx, but experimental grp gets an intervention that is more intense more "attractive"?
    when there is a desire to analyze dose-response effects
  5. Testing whether a larger dose is associated with larger benefits, or whether a smaller dose would suffice.
    dose-response effects
  6. A tendency for vatiation in ove variable to be r/t variation in another.
  7. Used to describe the relationship amongst variables; causality is NOT a consideration.
    Descriptive Correlational
  8. The AIM of a Univariate Descriptive design.
    • to describe the frequency of occurrence of a behavior or condition;
    • NOT to study the relationships
    • (Non Exp)
  9. Prevalence Design.
    • (Non Exp)
    • used to estimate prevalence rate of some condition at a PARTICULAR PT IN TIME
    • uses cross-sectional design
  10. Incidence Design
    • (Non Exp)
    • longitudinal (OVER TIME)
    • estimates frequency of developing NEW cases
    • must establish who is at risk of becoming a new case
  11. Start with a presumed cause & go forward in time to presumed effect.
    • Prospective/Cohort Design
    • (Non-Exp)
  12. Information about the potential causes is obtained BEFORE the effect/outcome
    • PROspective (cohort) Design
    • (Non-Exp)
  13. Group exposed to a phenomenon with potential health consequences is compared with a nonexposed group; researcher does NOT intervene.
    • Natural Experiment 
    • (Non-Exp)
  14. Tests whether the nonexperimental data conform to the underlying model to justify causal inferences.
    • Path analysis (similar techniques)
    • (Non-Exp)
  15. A design which allows researchers to manipulate multiple variables simultaneously.
    • Factorial 
    • (Experimental)
  16. People are randomly assigned to a specific combo of conditions.
    • Factorial Design
    • (Experimental)
  17. Factorial design allows to test not only ______ effects but also _______ effects.
    • main
    • interaction
    • (Experimental)
  18. What is Crossover Design?
    • Exposing same people to > 1 condition/tx/intervention
    • (Experimental)
  19. Type of design that can address multiple research questions.
    • Factorial
    • (Experimental)
  20. Crossover design is only effective if there is no expected what?
    • Carry-over effect
    • (when exposed to 2 different interventions, could be influenced by 1st intervention)
  21. How do you ensure no carry-over effect in a Crossover design?
    Make sure the treatments/interventions are not related
  22. What makes quasi-experimental designs different?
    • There is NO randomization, and 
    • sometimes NO comparison group
    • Use the word "comparison" group instead of "control" group
  23. Data is collected over an extended period of time and the intervention is introduced several times during that period of time.
    • Time Series 
    • (quasi)
  24. In a ______ design, one group only or in combo with a nonequivalent comparison group are used.
    • Time Series 
    • (quasi)
  25. A _______ group needs to be equivalent (as much as possible) to the intervention group.
  26. What is designed to perform the function of equalization of counterfactual and intervention groups?
    random assignment of people to one group or the other
  27. _________ prevents those who enroll participants from knowing upcoming assignments.
    allocation concealment
  28. ________ randomization may enhance the feasibility of conducting an experiment.
  29. __________ reduces the risk of contamination treatments (co-mingling of people in groups).
    Cluster randomization
  30. If a cluster randomized design is used, it is good to first _____ units along a dimension of importance BEFORE randomizing.
  31. Stratification is also called ________.
  32. What is permuted block randomization?
    Rather than having a randomization schedule for the entire sample, randomization occurs for BLOCKS of participants (ex: 6 at a time)
  33. Cluster randomization randomly assigns clusters of participants to different ________.
    treatment groups
  34. Which should you do first - randomize or obtain consent?
    Randomize (if after consent, people not assigned to their preferred condition may opt out of study)
  35. Who is blinded in a double-blind study?
    those delivering and those receiving the intervention
  36. Designs that involve collected data multiple times from two groups.
    Mixed designs
  37. Data collection at multiple post intervention points.
    • Repeated measure designs
    • (Experimental)
  38. The two ______ variables in a factorial design are the factors.
  39. How is a crossover design different than a factorial design?
    • Factorial: different people are randomly assigned to different tx (infant feeding)
    • Crossover: same people are exposed to more than one tx
  40. Which type of within-subjects design has the advantage of ensuring the highest possible equivalence among participants exposed to different conditions?
    Crossover design
  41. A placebo-type effect caused by people's expectations.
    Hawthorne Effect
  42. Control trials without randomization
  43. Collecting data on a new pt care facilitator (PCF) at baseline, before change, and after change; comparing this to a different hospital that is not instituting the PCF.
    • Nonequivalent Control Group pretest-posttest
    • (Quasi)
  44. In ________ experiments, the word comparison is used instead of control (group).
  45. Creating a single propensity score that captures the conditional probabilty of exposure to a tx given various preintervention characteristics.
    propensity matching
  46. When comparison data are gathered about a group of people B4 implementing the intervention
    Historical comparison group
  47. Single-subject studies use _______ designs to gather info about intervention effects based on a single pt under controlled conditions.
    time series
  48. What is compared in a dose-response design?
    the outcomes of those receiving different doses of tx (not as a result of randomization)
  49. When researchers study the effect of a potential CAUSE that they cannot manipulate, they use _______ designs to examine relationships b/w variables.
  50. In a retrospective design, the researcher begins with the _______ variable and then examines wheter it is correlated with one or more previously occuring _______ variables.
    • dependent (the effect)
    • independent (potential causes)
  51. What is a prospective design called in medical circles?
  52. A type of study that may involve multiple variables, but the primary purpose is to describe the status of each and not to relate them to one another. (like women's experiences with menopause)
    • Univariate Descriptive Studies 
    • (non-exp)
  53. Two types of descriptive study.
    • Prevalence
    • Incidence
  54. An estimated risk of "caseness" in one group compared with another. (dividing rate for one group by rate for another)
    Relative risk
  55. In _________ studies, researchers work with preexisting groups that were not formed at random, but rather through self-selection (selection bias).
  56. _______ research is often strong in realism and seldom critisized for artificiality.
Card Set:
GU Research Mod 7 Ch 9
2015-10-28 00:03:39
Research Polit Beck quantitative
Module 7, Ch 9 Quantitative Research Design
Show Answers: