70 likes | 184 Views
Additional Vocabulary relating to experiments: . Measurement/Ordering: Assigning numbers or scales to variables in an experiment. Ex: Using a 5 point scale ranging from “never” through “rarely” “sometimes”…. Additional Vocabulary relating to experiments: .
E N D
Additional Vocabulary relating to experiments: Measurement/Ordering: Assigning numbers or scales to variables in an experiment. Ex: Using a 5 point scale ranging from “never” through “rarely” “sometimes”…..
Additional Vocabulary relating to experiments: Mean: technical term for average. You will compare (2) means if you are working with an experimental group and a control group.
Additional Vocabulary relating to correlations: Coefficient of correlation: (The symbol for the coefficient of correlation is a lower case “r”.) How closely are two variables related? For example, how closely is being anorexic related to being a perfectionist? The coefficient of correlation is expressed as a number between 0 and 1. Zero means no correlation and 1 means a perfect correlation.
Additional Vocabulary relating to correlations: Coefficient of correlation: The coefficient of correlation can either be (+) or (-) . The “strength” of this relationship is what is indicated by the number 0-1. *This is very important!
Coefficient of correlation: Ex: The number of times a student is absent from class correlates -.40 with the final course grade. *It is a neg. correlation because as one factor goes up (the # of times class is missed) the other factor goes down (grades.) *A coefficient of .40 (on a scale of 0-1) indicates somewhat of a correlation/relationship between the two. Remember the scale ranges from 0, 0.1, 0.2, 0.3……..1.
Coefficient of correlation: *A coefficient of -.90 (on a scale of 0-1) would have indicated an extremely strong correlation between missing class and your final course grade. (You would be destined to fail.) In Psychology, anything above a .60 (+/-) correlation is considered to be quite high. Anything below .20 (+/-) is judged to be only minimally useful.
Correlations do not indicate a cause and effect relationship like experiments. Ex: There is a correlation between the softness of the asphalt in the streets and the number of heat stroke cases. Does soft asphalt CAUSE heat strokes? No, the SUN, a third variable causes heat strokes. There is simply a correlation between the two.