Variables, Design and Hypotheses Flashcards
What is experimental design?
-Vary an IV as the rest remains constant
-Measure changes in DV
-Changes that occur in IV tend to further cause changes to DV
-Infer causality
What is a quasi-experimental design?
-Remains similar to experimental
-IV is unable to be manipulated
What is a correlational design?
-No manipulations
-Measuring two or more variables to see how related they are to each other
-Can’t infer causality as there may be a third variable that could impact results
State information around IV’s
-An experiment can have one or more IVs and this is manipulated to allow to test for an outcome
-Each IV should have two or more levels
State information around DV’s
-Experiment will have one or more DVs and this is what will be measured
-Should be operationalised to specify how it’s going to be measured
-Different ways to measure DV’s; nominal, ordinal, interval and ratio
What is nominal data?
-Non numerical e.g. preferred travel method being bus/car/train
What is ordinal data?
-Numbers in a certain order e.g. education level
What is interval data?
-Values that have specific differences between them
-Lack a zero point (include negative)
-E.g. IQ, temperature
What is ratio data?
-Values that have a defined zero mark
-E.g. height and weight
What are confounding variables?
-Variable that wasn’t manipulated but could influence results
-Aim to be minimised as much as possible
What is a between subjects design?
-Ppts only take part in one level of the IV
-Independent groups
-Random assignment of ppts can help to account for individual differences
What is a within subjects design?
-Ppts perform all levels of IV
-Repeated measures
What are order effects?
-Repeated measures causes this to occur
-Can create fatigue, practice or boredom
-Causes results to differ to what they should be
-To prevent this, it’s best to use randomisation of trials and counterbalancing
What is a matched subjects design?
-Ppts are matched with someone else who has similar demographic characteristics e.g. IQ or socio-economic status
-Pair are tested as one individual over the two levels of IV
What is experimental/research hypothesis?
-Conceptual idea that tries to explain an observation
-Based on original theories
What is statistical hypothesis?
-Specific statement that we can use to collect data and test our hypothesis
-Also known as a prediction
What is null hypothesis?
-States that there is no difference between two variables
-Parametric statistics/Non-parametric statistics
What is alternative hypothesis?
-Logical alternative for the null hypothesis
-Predict that there will be a significant difference/relationship between variables
-Either directional/non directional
Describe the process of inferential testing
-Conduct Null Hypothesis Significance Testing
-Only reject null hypothesis
-If p<0.05, there is less than 5% probability that the results happened by chance, significant results found, can reject null hypothesis
-If p>0.05, there is more than 5% probability that the results happened by chance, non significant results found, failed to reject null hypothesis
