quantitative research design

The flashcards below were created by user jam110007 on FreezingBlue Flashcards.

  1. key features of quantitative research designs
    • - interventions
    • - comparisons 
    • - control 
    • - masking/blinding
    • - time frames 
    • - relative timing
    • - location
  2. interventions key questions
    • will there be an intervention?
    • what specific design will be used
  3. interventions broad design options
    • experimental [randomized control trial] 
    • quasi-experimental [controlled trial without randomization]
    • nonexperiemental [observational]
  4. comparisons key questions
    what type of comparisons will be made to illuminate relationships?
  5. comparison design options
    within-subjects design 

    between subject design
  6. within-subject design
    same people compared at different times or under different conditions
  7. between-subjects design
    different people are compared [men and women]
  8. control key questions
    how will confounding variables be controlled?

    which specific confounding variables will be controlled
  9. masking/blinding key question
    from whom will critical information be withheld to avert bias?
  10. time frames key questions
    • how often will data be collected?
    • when, relative to other event, will be collected?
  11. relative timing key question
    when will information on independent and dependent variables be collected-looking forward or backward in time
  12. location key question
    where will the study take place?
  13. causality
    many [if not most] quantitative research questions are about causes and effects 

    research questions that seek to illuminate causal relationships need to be addressed with appropriate designs

    how easy is it to infer cause and effect?
  14. counterfactual
    is what would have happened to the same people exposed to a cause if they simultaneously were not exposed to the cause - something to be discussed and not tested
  15. effect
    an effect represents the different between what actually did happen when exposed to the cause and what would happen with the counterfactual condition
  16. three key criteria for making casual inferences
    the cause must precede the effect in time - independent variable must come before the dependent variable 

    there must be a demonstrated empirical [testable] relationship between the cause and the effect

    the relationship between the presumed cause and effect cannot be explained by a third variable; another factors related to both the resumed cause and effect cannot be the real cause - would be a confounding variable
  17. additional criteria for causality was raised by who?
    • Bradford-hill, in connection with debate about evidence linking smoking to lung cancer. for example
    • - biologic plausibility 
    • - coherence
  18. biologic plausibility
    the casual relationship should be consistent with evidence from basic physiologic studies
  19. coherence
    the evidence about the existence of a relationship should come from multiple sources
  20. designs that support casual inference
    research designs vary in their ability to support the criteria for causal inference 

    • experimental designs offer the strongest evidence of whether a cause [an intervention] results in an effect [a desired outcome]
    • - that's why they are high on the evidence hierarchies for questions about causes and effects
  21. components of a true experiment
    • - manipulation 
    • - control 
    • - randomization
  22. true experiement: manipulation
    • the researcher does something to some subjects
    • - introduces an intervention or treatment
  23. true experiment: control
    the researcher introduces controls, including the use of a control group counterfactual
  24. true experiment: randomization
    also called random assignment

    the researcher assigns subjects to groups at random

    typical assignment is to an experimental group or a control group 

    the purpose is to make the groups equal wit regard to all other factors except receipt of the intervention
  25. post test only [or after-only] design
    outcome data collected only after the intervention 

    • symbolic representation 
    • R X O
    • R    O

    • R= randomization 
    • X = receipt of intervention
    • O = observation/measurement of dependent variable
  26. pretest-posttest [before-after] design
    outcome data collected oth basline and after the intervention 

    • symbolic representation 
    • R O X O
    • R O    O
  27. crossover design
    • subjects are exposed to 2+ conditions in random order 
    • subjects serve as their own control 

    • symbolic representation 
    • R O Xa O Xb O
    • R O Xb O Xa O
  28. factorial design
    • more than one independent variable is experimentally manipulated 
    • example: a 2x3 design = 2 levels of 2 independent variable, 3 levels of a second
    • example: in a 2x2x2 design, 2 levels of 3 independent variables are manipulated
  29. the experimental condition
    must be designed with sufficient intensity and duration that effect might reasonably be expected 

    attention must be pain to intervention fidelity [or treatment fidelity], that is, whether the treatment as planned was actually delivered and received
  30. possible control group conditions [counterfactuals]
    • ™No intervention is used; control group gets no
    • treatment at all.

    ™An alternative intervention is used (e.g., auditory vs. visual stimulation).

    ™A placebo or pseudo-intervention, presumed to have no therapeutic value, is used.

    ™Standard methods of care (“usual care”) are used.

    • ™A different dose or intensity of treatment or only
    • portions of it are administered.
  31. attention control
    extra attention, but not the active ingredient of the intervention
  32. delayed treatment ["wait-listed controls"]
    the intervention is given at a later date

    • symbolic representation 
    • R O X O     O
    • R O    O  X O
  33. quasi experiments
    involve an intervention but lack either randomization or control group
  34. two main categories of quasi-experiment designs
    nonequivalent control group designs 

    within-subject designs
  35. nonequivalent control group designs
    those getting the intervention are compared with a randomized comparison group
  36. within-subjects design
    one group is studied before and after the intervention
  37. nonequivalent group designs: if pre-intervention data are collected 
    • then the comparability of the experiment and comparison groups at the start of the study can be examined 
    • -  nonequivalent control group before-after design 
    • -  symbolic representation 
    • - O1  x  O2
    • - O1      O2
  38. nonequivalent control group designs: without pre-intervention data
    it is risky to assume the groups were similar at the outset

    nonequivalent control group after-only design is much weaker

    • symbolic representation 
    • X O1
    •    O1
  39. within-subject quasi-experiment: one-group before-after designs
    typically yield extremely weak evidence of casual relationships

    • symbolic representation 
    • O1  X  O2
  40. within-subject quasi-experiment time series design
    gather pre-intervention and post-intervention data over a longer period

    • symbolic representation 
    • O1 O2 O3 O4 O5 O6 O7 O8
  41. evaluation of quasi-experiments
    may be easier an more practical then true experiments, but

    they make it more difficult to infer causality 

    usually there are several alternative rival hypotheses for result
  42. non-experimental research
    if there is no intervention, the study is nonexperimental [observational]

    not all independent variables ["causes"] of interest to nurse researchers can be experimentally manipulated 

    i.e. gender cannot ever be manipulated
  43. correlational designs
    • ™Cause-probing questions (e.g., prognosis or harm/etiology questions) for which manipulation
    • is not possible are typically addressed with a correlational design.

    ™A correlation is an association between variables and can be detected through statistical analysis.
  44. Retrospective Design
    • ™In a retrospective correlational
    • design, an outcome in the present (e.g., depression) is linked to a hypothesized cause occurring in the past (e.g., having had a miscarriage).
  45. case–control design
    a retrospective design ™in which “cases” (e.g., those with lung cancer) are compared to “controls” (e.g., those without lung cancer) on prior potential causes (e.g., smoking habits).
  46. prospective correlational design
    ™a potential cause in the present (e.g., experiencing vs. not experiencing a miscarriage) is linked to a hypothesized later outcome (e.g., depression 6 months later).

    ™This is called a cohort study by medical researchers.

    • ™Prospective designs are stronger than retrospective designs in supporting causal
    • inferences—but neither is as strong as experimental designs.
  47. Descriptive Research
    ™Not all research is cause probing.

    ™Some research is descriptive (e.g., ascertaining the prevalence of a health problem).

    ™Other research is descriptive correlational—the purpose is to describe whether variables are related, without ascribing a cause-and-effect connection.
  48. Cross-sectional design
    Data are collected at a single point in time.
  49. Longitudinal design
    ™Data are collected two or more times over an extended period

    ™Longitudinal designs are better at showing patterns of change and at clarifying whether a cause occurred before an effect (outcome).
  50. ™Controlling external factors
    –Achieving constancy of conditions

    –Control over environment, setting, time

    –Control over intervention via a formal protocol
  51. ™Controlling intrinsic factors
    –Control over subject characteristics
  52. Controlling Intrinsic Factors

    ™Subjects as own controls (crossover design)

    ™Homogeneity (restricting sample)


    ™Statistical control (e.g., analysis of covariance)
  53. Characteristics of Good Quantitative Research Design
    • Statistical conclusion validity
    • Internal validity
    • External validity
    • Construct validity
  54. Statistical conclusion validity
    ™the ability to detect true relationships statistically
  55. Internal validity
    • the extent to which it can be inferred that the independent variable caused or influenced the
    • dependent variable
  56. External validity
    • ™the generalizability of the observed relationships
    • across samples, settings, or time
  57. Construct validity
    the degree to which key constructs are adequately captured in the stud
  58. Threats to Statistical Conclusion Validity
    ™Low statistical power (e.g., sample too small)

    ™Weakly defined “cause”—independent variable not powerful

    ™Unreliable implementation of a treatment—low intervention fidelity
  59. Threats to Internal Validity
    • Selection threat
    • history threat
    • maturation threat
    • mortality threat
  60. selection treat
    threat to internal validity

    baises arising from pre-existing differences between groups being compared
  61. history treat
  62. history threat
    threat to internal validity 

    other events co-occurring with casual factor that could also affect outcomes
  63. maturation threat
    threat to internal validity 

    processes that result simply from the passage of time
  64. mortality threat
    threat to internal validity 

    differential loss of participants from different groups - typically a threat in experimental studies
  65. threats to external validity
    ™Inadequate sampling of study participants

    • ™Expectancy effect (Hawthorne effect) makes effects observed in a study unlikely to be
    • replicated in real life.

    • ™Unfortunately, enhancing internal validity can sometimes have adverse effects on external
    • validity.
Card Set:
quantitative research design
2013-12-04 23:49:36
3215 final

3215 final
Show Answers: