Research Methods Flashcards
Independent variable
A variable that is hypothesized to cause or lead to variation in another variable
Dependent variable
A variable that is hypothesized to vary depending on or under the influence of another variable
Validity
that state that exists when inferences about empirical reality are correct
Measurement validity
When a measurement measures what we think it is measuring
Causal validity
When the conclusion that A lead to or causes B is correct
Ordinal measurement
A measurement of a variable in which the numbers indicating a variable’s value specify only the order of the cases permitting greater than and less than distinctions (1st, 2nd, 3rd place)
Interval measurement
Measurement of a variable in which the numbers represent fixed measurement units but there is not absolute zero (time of day)
Ratio measurement
Measurement of a variable in which the numbers represent fixed measurement units and there IS an absolute zero (inches on a ruler)
Reliability
A measurement procedure yields consistent scores when the phenomenon being measured is not changing
How to assess reliability
Test-retest, Cronbach’s Alpha for inter-item reliability, inter-observer reliability
How to improve reliability
Use an existing tool, pre-test the tool, do a cognitive interview for a survey/interview, establish inter-rater reliability for observtions
Three features of a true experiment
two (or more) groups–one that is an experimental group and one that is a control group. Random assignment to the two groups, variation in the independent variable before assessing the change in the independent variable
External validity
Does the treatment hold for subgroups across populations, settings etc?
Elements of Quantitative Methods
Emphasis on scientific method, testing a hypothesis, objectivity, need to quantify. Pro: allows the study of large populations and results using stats and probability
Elements of Qualitative Methods
Appreciates subjectivities, multiple perspectives, values depth over breadth and search for more holistic meaning, usually means natural settings. Pro: more intimate understanding of people, places, cultures
Mixed methods
Can permit capitalizing on best of the quantitative and qualitative traditions. Allows for inductive (bottom up) and deductive (top down) reasoning. Allows for triangulation of data sources.
Type 1 Error
Incorrect rejection of the null hypothesis (false positive)
Type 2 Error
Incorrect acceptance of the null hypothesis (false negative)
Internal Validity
Did the experimental treatment make a significant difference in this specific instance?
Threats to internal validity
Selection, history, maturation, testing
Construct validity
Is the study measuring the construct that it is intended to measure?
Threats to construct validity
Experimenter expectancies, novelty/disruption effects, Reactivity to experimental situation
Logic Model
A graphic depiction of rationale and expectations for the program: resources, activities, outputs, outcomes (short and long term)
Dusenbury et al. (2003)
Fidelity of implementation: adherence, dosage, quality of delivery, participant responsiveness and program differentiation.
Effect size
The difference in means between the treatment and control groups divided by the pooled standard deviation of the two groups. Important for revealing if a difference between groups is meaningful.
Lipsey et al. (2012)
Common benchmarks of effect size are not appropriate for education research where an effect of .30 is considered quite large.
Power
A measure of the likelihood of a researcher finding statistical significance in a sample if the effect exists. Power is a function of sample size, effect size and significance level