researchmeth

• one manipulated variable - IV
• one labile measured variable-DV

• two measured variables
• one called predictor (cause)
• one called predicted (effect)

• one stable measured variable (SV) treated as the IV
• one labile measured variable treated as DV

• a (typically unmeasured) variable that could be the cause of both the measured variables in a correlational study
• To fix, identifiy, measure, and co vary it out

a significant relationship that is not causal (in either direction)

• the correlation between one variable at one time and another variable at another time
• used to determine the more likely direction of causation
• if r2 X1Y2 > X2Y1 then X is probably the cause

• the correlation between two variables after the effects of a third variable (Z) have been removed
• used to test (and rule out) a third variable explanation
• if prXY.Z=rXY then Z is not a cause of both X and Y

the extent to which the results (from an experiment) will generalize to other situations

when the results from an experiment or study are unique to the situation

when the results from an experiment or study are unique to the subjects

when only easily-recruited subjects are used

when all members of the population have a definable probability of being sampled, but no attempt is made to match the group sizes

when the relative sizes of the groups in the sample are forced to be the same as those in the population

• Most popular type of non-proportional stratified random sampling
• when the sizes of the groups in the sample are forced to be equal

a structured set of items designed to measure attitudes, beliefs, values, or behavioral tendencies

• a small set of items designed to measure a particular attitude, belief, value, or behavioral tendency
• Likert: sets of strongly agree thru strongly disagree items
• Guttman: sets of ascending questions
• Thurstone: check all that apply (that are worth varying points)
• Semantic Differential: indicate position between opposite pairs

• studying behavior in everyday environments without getting involved
• key threat: reactivity (secondary observer bias)

• studying behavior from within the target group
• key threat: standard experimenter bias (secondary observer bias)
• note: Participant observation is not often possible since no consent observation can only occur when and where there is no reasonable expectation of privacy

• when the beliefs or expectancies of the observers influence what is recorded
• inter-coder reliability must be at least .90

• in general, when the beliefs and or expectancies of the experimenter end up altering the results
• standard experimenter bias: occurs when the experimenter behaves differently when collecting data in different conditions (experimenter reactivity)
• defenses: remove experimenter or double-blind
• observer bias is when only the recording of data is altered by the beliefs of the observer
• defenses: checklists and or partial sampling

when you take only one sample and then divide the subjects into the groups after-the-fact

• when you take separate samples for each of the groups
• note: this is another good example of the effort vs quality trade-off

• (aging) when you follow the same subjects over time
• major unique threat is time-frame (zeitgeist) effects

• (aging) when you take separate samples for each age group (at the same time)
• major unique threat is cohort effects

Solution to both longitudinal and cross sectional: run a hybrid study and verify same results either way

• Domain: are you studying hidden, internal variables or behavior? This determines whether to use a survey or observation
• Sampling: how much does this depend on who is measured? determines whether some fancy method (stratified, proportional, random) must be used
• Causal Logic: is one of my variables either highly stable or the passage of time? determines whether one of the specific designs is used

• it could be that X causes Y or that Y causes X
• to determine, run a cross-lagged study