Exam 2 Flashcards
Epistemology
the theory of knowledge; the branch of philosophy that investigates how people know what they know.
Positivist Paradigm
Objective "real reality" driven by natural laws context minimized aims to describe, predict, explain & control neutral observer experimental method truth sought via replicable observation
Constructivist Paradigm
Subjective multiple realities influenced by culture & environment context emphasized aims to describe, understand, transform active participant dialogic, transformative truth determined by individual and group
What paradigm is qualitative research?
Constructivist
*Phenomenology
Method: Individual’s LIVED experience
Foundation: Philosophy
Bracketing
Identify & set aside personal biases
*Grounded Theory
Method: Used to develop/CONSTRUCT** a theory
Foundation:
Data collection: Constant comparison*
Modifiability
ability to change theory in light of new data
Grounded theory data collection & analysis uses a ___ construct
Constant Comparative Construct
Ethnography
Method: description of cultural groups or subgroups
Foundation:cultural anthropology
Data collection: Fieldwork/ field notes
*Emic View
Insider’s view; view of the participants or members of the social group
*Etic View
Outsider’s view; researcher’s interpretation of views about the human social life in a social science perspective
*Ethnography utilizes which kind of data collection
Field notes / Field work
Historical Method
Method: Systematic compilation of data to describe some past event
Foundation: philosophy, art, science
Basic components to research
- Identify the phenomenon
- Structuring the study
- Gathering the data
- Analyzing the data
- Describing the findings
*Credibility
TRUTH OF FINDINGS are judged by participants and others within the dicipline
*Auditability
Accountability as judged by the ADEQUACY OF INFORMATION leading the reader from the research question and raw data through various steps of analysis to the interpretation of findings.
*Fittingness
FAITHFULNESS TO EVERYDAY REALITY of the participants, described in enough detailso that others in the discipline can EVALUATE IMPORTANCE FOR THEIR OWN PRACTICE, RESEARCH & THEORY DEVELOPMENT
*Confirmability
Findings that reflect implementation of CREDIBILITY, AUDITABILITY and FITTINGNESS standards.
*Trustworthiness
Rigor or goodness of data
Triangulation
Combining methods, theories, dat sources, or investigators to converge on a single construct.
*Constructivist
Truth determined by the individual or cultural group
*Positivist
Truth sought via replicable observation
Purpose of research design:
- Avoid bias
- Provide plan to aid in solving problem/question/hypothesis
- Allows researcher to apply different levels of control so it can be said that the independent variable REALLY changed the dependent variable and not something else
Internal Validity
Asks if it is the independent variable (or something else) that caused or resulted in the change in the dependent variable.
Threats to internal validity
Selection Maturation Testing Mortality Instrumentation
External Validity
Questions the conditions under which the findings may be generalized. Deals with the ability to generalized the findings outside of the study.
-across types of persons, settings, & times.
Threats to external validity
Selection effects (Who) Reactive Effects (where-how) Measurement effects (how-when-what)
3 Characteristics of a true experiment or RCT/Experimental Design Features
Randomization
Control Group
Manipulation
Types of Experimental designs (Level II)
True experimental design
Solomon four-group design
After-only design
Experimental Design - Advantages/Disadvantages
- MOST appropriate for testing cause-&-effect relationships
- Provides highest level of evidence for single studies
- Not all research questions are amenable to experimental manipulation or randomization
- Costly
- Threats to internal validity not controlled by RCT: Subject mortality, Testing
- Difficult logistics in field settings
Hawthorne effect
just participating in atrial alters the behavior of the subjects, which can effect validity of study/results.
Types of Quasi-experimental design types (Level III)
- Nonequivalent control group design
- After-only nonequivalent group design
- One-group (pretest, post-test design)
- Time series design
Quasi-experimental advantages and disadvantages
- Practical and more feasible, especially in clinical settings
- Some generalizability
- UNABLE to amke clear cause-and-effect statements
- May not be able to randomize
- May control some extraneous variables: A priori by design, statistically
Experimental & Quasi-experimental studies offer
Strongest evidence for interventions
Non-experimental designs (Level IV & V)
(are still RESEARCH)
- No manipulation of an independent variable (independent variables have already occurred)
- Concepts of control still observed
- Cohort- subjects of a specific group that are being studied
Corhort
subjects of a specific group that are being studied
Random Assignment controls for?
Selection Bias
Another name for study weakness is:
Limitations
The threat to internal validity that is most difficult to control is?
Mortality
The degree to which study results can be applied to the larger population refers to :
External Validity
Which type of design controls for the reactive effects of the pretest?
Solomon four-group
*A true experiment requires:
Control
Manipulation
Randomization
*What must be present in a quasi-experimental research?
Manipulation of a variable
*The paradigm guiding a qualitative researcher is:
Contructivism
*Population
a well-defined set that has certain properties
- People
- Animals [BENCH SCIENCE]
- Objects
- Events
*Target population
The overall group of subjects or events to which the researcher is interested in generalizing conclusions
*Accessible Population
The group of people available to the researcher from which to pick a sample
*Sampling
a process of selecting a portion or subset of the designated population to represent the entire population
*Sample
a subset of sampling units from a population
*Sampling frame
a list of all units of the population
*Element
the most basic unit about which information is collected
*Sampling strategies - Probability
RANDOM SELECTION
each subject has an equal chance of being chosen to be in the sample
*Sampling strategies - Non-Probability
NON-RANDOM
*Probability sampling - Simple random
-Names out of a hat; table of random numbers
-LEAST bias
=population representativeness maximized
*Central Limit Theorem
Given any parent population, not necessarily normal, having mean u and standard deviation o, the sampling distribution for fixed n which is generated from this population will be approximately normally distributed with mean u and standard deviation of (insert equation)
Need what three assumptions for confidence
- Normally distributed
- Equality of variances
- Independence of data
