Change the setting of the study control the study

Change the setting of the study
Control the study findings
 Distort the study findings
Reduce chance of error
Correct!
Bias distorts the findings from the real and the true. Changing the setting, controlling the study findings, and reducing the
chance of error can affect the generalizability of the research results.

NURS4110 sp2019
o
What is control?
 Implemented throughout the design
 Improved accuracy of findings
 Increased control in quasi-experimental research
 Greatest in experimental research
The nurse researcher is explaining the purpose of control in a study design. This would include which of the following?
Establish the credibility of the researcher.
Highlight design flaws.
 Increase the probability that the results are true to reality.
Interfere with the validity of the findings.
Correct!
As control increases, the likelihood that the study findings are an accurate reflection of reality increases.
o
What is manipulation?
 A form of control
 Implementation of a treatment or intervention
 The independent variable is controlled.
 Must be careful to avoid introduction of bias into the study
 Usually done only in quasi-experimental and experimental designs
 Internal and external validity
Internal “occurring inside the study” External “occurring on the outside”
Was the research done right?
Does it avoid confounding? (more than one possible
independent variable acting at the same time).
Less chance for confounding, higher internal validity.
High validity allows you to choose one explanation over
another.
How well data & theories from one setting apply to another
Generalizability
Cannot make inferences without external validity! Why? It can’t
be applied to the bigger picture without external.
Threats to internal validity:
History- when an event occurs at same time of treatment and
Threats to external validity:
Interaction bw how the subjects were selected & treatment

NURS4110 sp2019
changes participants behavior
Maturation- participants change over time
Testing- participants performance after multiple tests is
different than the initial testing
Instrumentation- changing measurement methods during the
study
Mortality- attrition, withdrawals, dropouts throughout that may
lead to unequal groups at the end of the study
Selection interactions- the selection method may interact with
one or more of the others causing bias
Pretesting may cause them to act differently than normal
Experimental setting- may react differently based off of where
they are
Multiple treatments/interventions- may lead to limited
generalizability bc early treatments may do the trick
The research study has a threat to internal validity when which of the following occurs?
 A valid rival hypothesis exists.
Findings are generalized beyond the target population.
The results of statistical analysis are incorrectly interpreted.
There is a mismatch between the conceptual and operational definitions.