Research Methods Final

Questions

• Diffuse question

• Specific question

• Connection between the two

Diffuse question

Specific question

Connection between the two

Alternative Hypotheses

Logic

Method

Results

Inferences

IV & DV

Confound/Confounding Variable

an uncontrolled factor in an experiment that systematically varies with the independent variable (IV), and causally influences the dependent variable (DV)

Qu. Participants in an experiment interacted with a female confederate in both conditions. Is the confederate’s gender a confound?

NO, it doesn’t vary - gender can be a confound but since she is both conditions, it is NOT a confound

• Confounds need to be different in one of the conditions

Nuisance variable

is associated with the DV that is outside of the influence of the IV, adding noise but not bias.

Construct Validity

Degree to which your operationalizations, measures or experiments capture (or generalize to) the concepts you’re hoping to study. (Most important validity in research)

External validity

Another validity in research

Degree to which the conclusions in your study would hold for other persons in other places and at other times.

Focus on generalization

Reliability

Consistency of measurement values for a variable.

• Test-retest

• Internal Consistency

• Inter-rates Reliability

Test-retest

same measure, same individual at different points in time under same conditions

Internal Consistency

a set of measures that all get at the same underlying idea. If I score “high” on one half of the test, I should also score high on the other half of the test. (e.g. average inter-item; Split half; Cronbach’s alpha)

Inter-rater reliability

Do raters agree on observations?

Main Effects

A “____” is the effect of one of your independent variables on the dependent variable, ignoring the effects of all other independent variables

For example, in a study of diet and exercise on weight loss, the main effect of diet is the average difference in weight loss between two diets regardless of exercise levels.

Interaction

When the combined influence of factors is not simply additive. An _____ is present if the effect of one factor (e.g., drug) depends on the level of another factor (e.g., age).

IQ Test Main Effect of Teacher Expectations

If you want to see if there’s an ME of teacher expectations average across the high condition and the normal condition, and see if there is a difference

• Make sure to write explicitly that there is or isn’t a ME for each IV

IQ Test Main Effect of Student Age

• ME’s go in the direction of the higher average once you find the difference

IQ Test Interactions

Subtract from within and/or across group and if there is a difference there is an interaction

• For graphs if there is an intersection there is an interaction

IQ Test Main Effect & Interactions Plots

Correlational/Quasi-experimental design

Experimental design

Correlation is not causation

If a correlating variable changes, the other variable it is correlated to does not necessarily change - the correlated variables are only coinciding with each other, not causing each other

• Experimental design is needed to determine causality

IQ Test IV and DV

IV: Age, and teacher expectations 2 conditions each

• 7 and 15 years, High and normal

DV: Mean IQ test scores

Opportunistic Observations Pros

• Able to access data in quantity that is hard to get in lab studies

  • twitter and emotions data study

• Gives you access to people across space and the past

• Study psychological constructs with external validity that is hard in lab settings

• Helps generate new hypotheses and variables unexplored before

Opportunistic Observations Cons

• Can’t control for confounds

• Harder to isolate cause and effect lack of control