Intro and research design Flashcards
Parametric statistics
One branch of the statistical tree, identified primarily by an assumption of “normality” to the distribution that is formed by the data (an assumption of symmetry).
Non-parametric statistics
Assumptions of symmetry of distribution cannot or should not be made. (use chi-square to analyze data)
Descriptive statistics
describe
Inferential statistics
Attempt to answer “why”
Order of operations
Parentheses Exponentiation Multiplication Division Addition Subtraction
Population
All the people targeted by a study (N)
Sample
A subset of the population (n)
Representative
the extent to which the variables in the population are found in the same proportion in the sample.
Generalization
the extent to which you can apply the findings from your sample to the population.
Parameters
Numbers that apply to populations. (population IQ data, for example)
Statistics
Numbers generated on the sample.
Methodology
HOW WE DO RESEARCH; PROCESS (class)
book: the study of research methods. Addresses the broad tasks of how tasks are done, not the doing. Methodology is the blueprint for research whereas the method is the actual means of executing that blueprint. Involves the overarching principles of the research process
Method
WHAT WE DO (class)
book: the adaptation of the methodology to the actual doing. Methodology is the blueprint for research whereas the method is the actual means of executing that blueprint. Method brings methodology down one level to pragmatic considerations of how you are going to do it.
Experimental design
defined by random assignment of subjects across predetermined control and/or treatment conditions.
Quasi-experimental design
a quasi-experimental variable must be preexisting in subjects because, by the nature of the variable, a researcher cannot ask subjects to assume the risks that go with the behavior (i.e. smoking)
Correlational design
examines the potential for relationships between variables that might logically seem to be related. It does not establish causal factors.
Survey design
uses surveys to collect data
Single-subject design
studies only one subject. (an intervention may be measured with an ABAB design.)
Qualitative design
do no use mathematical data and are instead based on observations. Ex: Jane Goodall’s work with chimpanzees.
Operalization
making sure you can accurately measure what it is you want to measure. (using a scale to measure weight, for example)
Characteristics of a research question
- Parsimonious: very carefully stated (stingy, even)
- Precise
- Objective
- Able to operationalize: ask the question in such a way that the question clearly defines what variables will be compared and that the variables under investigation are measurable.
- Rational
Should be stated as a hypothesis
Why do we research?
Curiosity, and to advance the species: protection, comfort, sustainability
Simple pattern of research methods
question, data collection, data analysis
Statistics (simple, general definition)
Using basic math computation and some symbols to calculate probability and variance, probability of error/chance, magnitude of difference between groups
Tenets of Science
- Nature is lawful
- Nature can be identified and understood through systematic investigation
- Behavior is deterministic (cause and effect)
- Data should be defined, empirical, objective, systematically gathered, controlled, verifiable
Qualities of data in research
Data should be defined, empirical, objective, systematically gathered, controlled, verifiable
The Scientific Method
- Ask questions
- Read the lit
- Design methods
- Collect and analyze data
- Answer the question
- Share your findings
Forming a research question
- start broad, then narrow it down
- make your question precise, objective and able to fit into a hypothesis
Independent variable
treatment level or group; intervention
Dependent variable
outcome
What is included in a methods section?
- participants and setting
- materials
- procedures
- design, variables, analyses
“Answer the question”
Your data will only answer your research question (do no overgeneralize)
What is the most important step in research?
Sharing your findings! Through a peer reviewed paper, conference. You have a social responability!
Can you use the word “proven” with your study results?
NO! Hypothesis is either supported or not, never proven.
What is the hardest part of research design?
Planning your study: making sure your methods will answer your research question and are supported by previous research.
