Research methods Flashcards
(46 cards)
Variable
Any phenomenon that can take on more than one value.
Continuous variable
Has a continuum of possible values.
Categorical variable
Can only take on fixed values in groupings or categories.
Extraneous variable
An external variable not being investigated that may affect IV or DV and cause inaccurate conclusions.
Confounding variable
Produces effects that might be confused with the effects of the independent variable.
Independent variable
The variable manipulated by the experimenter.
Dependent variable
The response measured by the experimenter.
Experimental research
Researcher manipulates aspects of a situation and examines the response to assess cause and effect. May have external validity issues but provides strong evidence to establish causation.
Generalisability
If research results from the sample can be applied to the entire population of interest. findings must be valid.
Measure
Way of assessing a variable, must be reliable and valid.
Reliability
Measure’s ability to produce consistent results.
Validity
Measure’s ability to assess the variable of interest or assess what it is intended to measure.
The scientific approach
Uses empirical methodologies such as observation and experimentation to gain knowledge. Three main goals are description, prediction and understanding.
Quantitative research
Experiments or surveys that provide quantified data.
Qualitative research
Interviews, observations and case studies to understand phenomena. Has holistic view, inductive approach. Has three key perspectives: positivism (objective research that assumes a universal truth), interpretivism (subjective research in natural setting) and critical theory (taking a moral-political approach to examine how power is expressed in society).
Theory
Systematic way of organising and explaining observations.
Hypothesis
Prediction of relationship between two or more variables.
Random sampling
Every member of the population has equal probability of being selected.
Stratified random sampling
Sample selected that proportionately represents all subpopulations. Selection is only random within groups.
Correlation coefficient
measures/ quantifies the association between two variables and ranges from -1.0 to +1.0. A correlation of 0 means two variables are unrelated. High correlation positive or negative means scores on one variable is a good predictor of the other variable. Measured with Pearson’s r = co-efficiency.
Positive correlation
The higher the score is on one variable, the higher the other variable will score.
Negative correlation
The lower the score is on one variable, the higher the other variable will score.
Informed consent
Participants must be informed of the purpose of the study and the nature of the treatments before they agree or refuse to participate.
Deception
Some experiments need deception but participants must be debriefed, can withdraw at any time, and the research is important and cannot be conducted without deception.