Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
49 Cards in this Set
- Front
- Back
threats to validity
|
reasons that an inference could be wrong
threats to validity are ruled out by exercising various types of control |
|
the best control method is
|
randomization to treatment conditions
|
|
homogeneity
|
is the use of a homogeneous sample to eliminate variability on confounding variables
|
|
pair matching is when
|
participants on key variables are matched to make groups more comparable
|
|
statistical control is the
|
process of removing the effect of a confound variable statistically through analysis of covariance
|
|
two disadvantages of of homogeneity, stratifying,matching and statical control are
|
researchers must know in advance which variables to control
can rarely control all the variables |
|
four types of validity affect the rigor of a quantitative study
|
statistical conclusion validity
internal validity construct validity external validity |
|
statistical conclusion validity concerns the
|
validity of inferences that there is an empirical relationship between variables (most often, the presumed cause and the effect)
|
|
threats to statistical conclusion validity icnlude
|
low statical power
low precision factors undermine a strong operationalization of the independent variable |
|
statistical power is the
|
ability to detect true relationships among variables
|
|
precision is the
|
exactness of the relationships revealed after controlling confounding variables
|
|
intervention fidelity concerns
|
the extent to which the implementation of a treatment is faithful to its plan
|
|
intervention fidelity is enhanced through
|
standardized treatment protocols
careful training of intervention agents monitoring of the delivery and receipt of the intervention manipulation checks treatment adherence avoiding contamination of treatment |
|
internal validity concerns
|
inferences that outcomes were caused by the independent variable rather than by factors extraneous to the research
|
|
threats to internal validity include
|
temporal ambiguity
selection history maturation mortality testing instrumentation |
|
temporal ambiguity is the
|
lack of clarity about whether the presumed cause preceded the outcome
|
|
selection is the
|
preexisting group difference
|
|
history is the
|
occurrence of events external to an independent variable that could affect outcome
|
|
maturation
|
changes resulting from the passage of time
|
|
mortality
|
effects attributable to attrition
|
|
testing
|
effects of a pretest
|
|
instrumentation
|
changes in the way data are gathered
|
|
internal validity can be enhanced through
|
judicious design decisions
analysis |
|
intention-to-treat analysis
|
analyzing outcomes for all people in their original treatment conditions
|
|
per-protocol analysis
|
analyzing outcomes only for those who received the full treatment as assigned
|
|
which is preferred intention to treat analysis or per-protocol analysis?
|
intention to treat
|
|
contruct validity concerns
|
inferences from the particular exemplars of a study (e.g. the specific treatment, outcomes, people and settings ) to the higher order construct that they are intended to represent
|
|
if the operationalization of a construct fails to incorporate all of the relevant characteristics of the construct or if it includes extraneous content what happens to construct validity?
|
it is threatened
|
|
examples of threats to construct validity are
|
subject reactivity
researcher expectancies novelty effects compensatory effects treatment diffusion |
|
external validity concerns
|
inferences about the extent to which study results can be generalized that is about whether relationships observed in a study hold true over variation in people, settings, outocme measures, and treatments.
|
|
efficacy studies emphasize
|
internal validity
|
|
effectiveness studies emphasize
|
external validity
|
|
RE-AIM framework is a model for
|
designing and evaluating intervention research that is strong on multiple forms of study validity
RE-AIM = reach, efficacy, adoption, implementation and maintenance |
|
selection effects
|
might be impossible for investigators to have an adequate number of individuals or adequate access to a target population
|
|
Reactive effects
|
participants reacts in a certain way to the research method that affects the results
|
|
Measurement effects
|
prior exposure
know that they are being assessed |
|
what kind of relationship do internal and external validity have
|
inverse
|
|
what is mono-method bias
|
he way an investigator might ***** a key variable in the research design.
Having only one particular way to assess a given variable. |
|
statistical conclusion validity can be undermined by
|
low statistical power
excessive homogeneity exert too much experimental control lack of treatment fidelity whatever interventions are used in a randomized control trial, need to be used and applied identically not doing this creates competing hypothesis |
|
steps in the research process are
|
Establish the Research Problem
Form a research question Develop a Hypothesis Design the study that is appropriate and answers the questions Data Analysis statistical analysis Interpretation |
|
the more connected the steps of the research process are the more evidence we have of
|
FIT
|
|
FIT is related to the extent of
|
scientific rigor = the more connected this steps are the more evidence we have of
|
|
simple research hypothesis
|
has minimum number of varibales
|
|
complex research hypothesis has
|
many variables
|
|
independent variable is
|
predictor variables, the effects of independent variable predicts the changes we see in the dependent variable
|
|
dependent variable is
|
outcome or response variables
outcome or response we see in the dependent is a function of change in the independent variable |
|
extraneous variables are ones that
|
influence can be broad
problematic |
|
extraneous variables can impact the relationship between
|
independent and dependent variables
directly impact the variances in the independent variable impact the dependent variable that are independent from the independent variable |
|
extraneous variables can create
|
competing hypothesis
which means that the dependent and independent variables are not actually creating our results |