PSYCH 202 1st Edition Lecture 4Behavioral Measures:I. Does it really capture the concept?Experimenter or expert rating:II. Potential for confirmation bias• Seeing what we expect to seeReliability:I. Stability or consistency in measurement.Inter-rater (or inter-observer) reliability:III. Degree to which two (or more) raters or observers agree that they have seen the same thing.Test-retest reliability:I. Do you get the same results at 2 different timesValidityII. Does the measure assess the concept it is supposed to measure?III. Is what we are concerned about in translating ideas into variablesIf 2 variables are correlated (or “co-vary”), when one changes, the other does too. I. We can predict one variable from the other.Correlation coefficient:II. measure of how closely the values of two variables are related to each other.III. Correlation coefficient tells you two things:1. Direction of the relationship: Is the correlation positive or negative? a. positive correlation - when one number goes up, the other goes upb. negative correlation - when one number goes up, the other goes downThese notes represent a detailed interpretation of the professor’s lecture. GradeBuddy is best used as a supplement to your own notes, not as a substitute.2. Strength of the relationship: How far is the correlation coefficient from zero? a. Correlations can range from -1.0 to 1.0Zero correlation means no relationship between the 2 variablesCorrelation does NOT equal causality!!!I. However if there is a causal relationship, then there will also be a correlation• Seeing what we expect to seeCausal possibilities with a correlation:II. A causes B (watching violent TV causes aggressive kids)III. B causes A (aggression causes kids to watch violent TV)IV. C causes A and B (third variable explanation: lack of parental supervision causes violent TV watching AND aggressive kids)V. Correlation does not imply causationBut causation implies correlation:I. If something causes something else, it should be correlatedPlacebo Condition:II. Experimental condition that research participants think or expect will have some effect, but which in fact is inert (has no effects).Double Blind Study:I. neither person administering study or participant knows which condition the participant is
View Full Document