exam ii: ch8 Flashcards
blueprint or detailed plan for conducting a study
research design
what provides the basis of a research design? (3)
purpose, review of literature, and framework
cause and effect relationship between variables
causality
cause + effect relationship between interrelating variables = change in dep variable
multicausality
the likelihood of accurately predicting an event
probability
slanting of findings away from what is true/expected, can distort the findings
bias
what type of research has the strongest control?
experimental
having the power to manipulate factors for desired outcome; method to reduce bias, improve accuracy, implemented throughout design
control
form of control used in quasi-experimental/experimental studies; implemental of a treatment or intervention
manipulation
what are the 5 elements of a strong design?
controlling…
- environment
- equivalence of subjects
- tx
- measurement
- extraneous variables
measure of the truth/accuracy of the findings obtained from a study
study validity
type of validity that begins with the fit between the conceptual and operational definitions of variables
construct validity
type of validity that measure the extent to which the study findings are a true reflection of reality, rather than the result of extraneous variables
internal validity
concerned with the extent to which study findings can be generalized beyond the sample used in the study.
external validity
concerned with whether the conclusions about relationships or differences drawn from statistical analysis are an accurate reflection of the real world
statistical conclusion validity
explain the following threats to construct validity:
- inadequate definition of variables
- mono-operation bias
- mono-method bias
- experimental expectancy bias
- social desirability
Inadequate definition of variables: lack of clear + operational definitions may = not congruent/inappropriate measurements
Mono-operation bias: each variable measured with one instrument/scale
Mono-method bias: only one measurement method is used to measure study variable
Experimental expectancy bias: researcher’s expectations or bias may influence data collection
Social desirability: participants select answers based on wanting researcher to like them/based on behaviors that are perceived as positive
explain the following threats to internal validity
- nonrandom participant selection
- nonrandom assignment to groups
- participant attrition
- unrelated event affecting the findings
- participant maturation
Nonrandom participant selection: participants selected by nonrandom sampling – unique characteristics that may influence study findings
Nonrandom assignment to groups: participants are placed in groups using nonrandom sampling in an interventional study, characteristics of the groups may influence the findings
Participant attrition: percentage of participants withdrawing from longitudinal study is >20%. The participants withdrew may be different from those who complete
Unrelated event affecting the findings: unrelated event that may affect (history, ex covid)
Participant maturation: participants become more experienced, wiser, or tired over the course of a study, which may influence the data collected
explain the following threats to conclusion validity
- low statistical power
- unreliable measurement methods
- intervention fidelity concerns
- extraneous variables in study
Low statistical power: conclusion is drawn that there are no sig differences or relationships when one exists (type 2 error)
Unreliable measurement methods: scales/measures used in a study are not consistently measuring study variables
Intervention fidelity concerns: not consistently implemented, may make detecting differences difficult
Extraneous variables in study: setting not controlled in study, which may affect responses
explain the following threats to external validity
- homogeneity in noninterventional study
- interaction of sample and setting
- interaction of selection + intervention
- interaction of history and intervention
Homogeneity in a noninterventional study: sample recruited was unique and does not reflect larger population
Interaction of sample and setting in the noninterventional study: the sample and the setting combined in a way that resulted in a unique sample that does not reflect larger population
Interaction of selection + intervention: participants decline to participate because of the commitment or effort the test requires. the sample does not reflect larger
Interaction of history and intervention: event that occurs in the setting affects the implementation of the intervention and data collection
type of study that seeks to gain more information about concepts, variables, or elements in a particular field of study.
descriptive studies
type of descriptive design used to examine variables in a single sample
simple descriptive design
type of descriptive design that examines differences in variables in 2+ groups that occur naturally in a setting
comparative descriptive design
examine relationships between or among two or more variables in a single group in a study.
correlational design
type of correlational design that describes variables and examines relationships
descriptive correlational design
type of correlational design that predicts the value of one variable based on the values obtained for another variable/variables
predictive correlational design
type of research design that has untreated control group with pretest/posttest
quasi experimental design
groups in comparative descriptive studies
study groups
uses large # of subjects to test a treatment’s effect and compare results with a control group who did not receive the treatment
- randomization is ESSENTIAL
- multiple geographic locations
- subjects come from reference population
randomized controlled trial
t/f: research designs should be developed to reduce likelihood of bias
true
