Flashcards in Quantitative Observations Deck (25):
Aspects of an observational setting that cause people to behave the way they think an observer wants or expects them to behave.
A technique for gathering scientific knowledge by unobtrusively observing people in their natural environments.
Two Issues with Naturalistic Observation
1. If an event does not occur naturally or very often.
2. If an event only occurs through interaction.
Surveys are good for observing:
Attitudes, beliefs, then behaviours, in that order.
Measures to Avoid Demand Characteristics
Privacy, anonymity, involuntary reactions, blindness.
No one is watching or looking.
Removing name, giving number. The participant cannot be held accountable for their actions, and are more likely to act naturally.
Example of Involuntary Reaction
Pupils dilating when excited.
Keep true purpose of research study hidden through cover stories or filler items/measures (lying to participant).
Observer Expectations Can Influence:
-How something is observer.
-How someone behaves.
Double Blind Study
An observation whose true purpose is hidden from both the researcher and the participant. Can be facilitated through the use of computers or research assistants.
Used to determine relation and causation, and explain why. The co-relationship or pattern of co-variation between wo variables, each which has been measured several times.
A property whose value can vary and change. Example, insulted or not insulted and agreed to give time or not.
Two Types of Correlation Coefficient
Positive and negative correlation.
More-more or less-less (spinach increasing longevity).
More-less or less-more (bacon decreasing longevity.
Measures direction and strength of relationship between two variables.
Perfect positive relationship.
Perfect negative relationship.
A correlation observed between naturally occuring variables.
Example of: all variables that are causally related are correlated, but not all variables that are correlated are causally related.
Three possibilities (X=watching violent TV, Y=aggressiveness, Z=lack of parental supervision).
o X can cause Y,
o Y can cause X, or
o Z can cause X and Y.
Third Variable Correlation
Two variables (X and Y) are correlated only because each is causally related to a third variable (Z).
Matched samples and matched pairs.