Week 4 - Research and Statistics Flashcards
Scientific Method
Founded on empiricism - the assumption that the truth is best discovered through sensory experience
Empirical evidence
evidence that is publicly confirmable
Hypothesis
A prediction, generally phrased as an “if…then” statement
Theory
The best explanation we have given the facts at hand, including previous hypothesis testing
Population
The group that is ultimately of interest to the researcher
Sample
A subgroup of the population
6 ways that data may be gathered depending on the research question
- Archival research
- Observation
- Surveys
- Interviews
- Psychological tests
- Biodata
Basic research
research done for the sake of curiosity, theory development, or knowledge expansion
Applied research
research done to help solve a practical problem
Descriptive studies
Used simply to describe phenomena or populations of interest (who?, what?, where?, how?) - descriptive studies don’t always use statistics, but they can
Inferential studies
Used to explain and predict (when?, why?) - Often involve the use of inferential statistics, which are used to draw conclusions about the probability of the research results occurring by chance
Qualitative study
A textual description of interview data or observations
Case study or series
Presents a rich summary of facts, observations, and tests pertaining to at least one individual or group
Correlational study
Identifies the strength of association between two or more variables of interest EG - age is positively correlated with vocabulary - relationship between hours of study and GPA
Correlation coefficient
A statistic that expressed the size and direction of relationship between two variables - the larger the absolute value, the stronger the relationship
Positive correlation
When values of one variable change in the same direction as valued of the other variable; e.g. r=.76
Negative correlation
When high values of one variable are associated with low values of the other variable; e.g r= -.32
Inferential Research
Inferences have to do with cause and effect relationships - experiments are the best way to prove. the existence of cause-effect relationship
Variable
Any factor that varies in amount or kind
Independent Variable
A factor that is manipulated by the researcher
Dependent Variable
A factor that the experimenter measures to determine the impact of the IV - outcome variable
Control group
Not exposed to the conditions expected to create change
Random assigment
An equal chance of being assigned to the control group or the experimental group
Experimental control
Other systematic methods of ensuring equal treatment across groups and the studies ( ex strict procedures, no distractions)
Placebo effect
Provides no active effect
Quasi-experimental studies
Demonstrate that naturally existing factors that differentiate two groups lead to differences in measured variables
Longitudinal studies
Measure the same variables in the same person at different ages
Cross-sectional studies
Measure a trait across as number of age groups/ categories at the same time
Sequential studies
Combine cross-sectional and longitudinal studies, and are therefore the most efficient design
Mean
Average
Median
Midpoint in rank-ordered data
Mode
Score appearing most often
Normal distribution
Bell-shaped curve
Standard deviation
Degree to which scores in ordered distribution are spread out