Flashcards
statistical theory
Flashcards do conjunto (23)
- 1One tailedUsed if your hypothesis was directional
- 2Two tailedUsed if hypothesis was non-directional (probability levels double as they're looking for a significant increase and decrease)
- 3Experimental hypothesis (HE)Researchers begin research by writing hypothesis; an experimental hypothesis predicts the effects of the IV on the DV in an experiment
- 4Research hypothesis (HR)For non-experiment research, it predicts the outcome of a study
- 5Directional hypothesisStates the direction of the difference (higher, lower etc). Tend to be used when the results of previous research suggests a particular outcome.
- 6Non-directional hypothesisDoes not state the direction of the difference.
"There will be a significant difference" (Experiment)
"There will be a significant relationship"
(correlation) - 7Statistical testDetermines which hypothesis is true and thus whether we accept or reject the null hypothesis
- 8Null hypothesis (H0)States there IS NOT a significant difference between variables. This is accepted when there's >5% probability that results are due to chance
- 9Alternative hypothesis (H1)This is used after the null hypothesis is rejected. This states there IS a significant difference between variables
- 10FalsifiabilityA theory can not be considered scientific unless it admits the possibility of being proved untrue
- 11UnrelatedIndependant groups is unrelated as ptts in each condition are different
- 12RelatedRepeated measures and matched pairs are related as ptts are matched on some variable that is important for the investigation
- 13Chi squaredA test for difference or correlation between 2 variables or conditions, data should be nominal and use an unrelated design
- 14sign testA test of difference between 2 conditions; data should be at a nominal level using a related design
- 15Mann-whitneyA test for a significant difference between 2 sets of data must be ordinal levels with an unrelated design
- 16WilcoxonA test for significant difference between 2 sets of data, data should be ordinal levels using a related design
- 17Spearman's RhoA test for a correlation when data is at an ordinal level
- 18Unrelated t testA test for difference between 2 sets of data, data must be at interval level with an unrelated design
- 19Related t dataA parametric test for difference between 2 sets of scores; data must be interval level with a related design
- 20Pearson's rA parametric test for a correlation when data is at interval levels
- 21Nominal datacategories of data
(eg. male or female) - 22Ordinal datasubjective (eg. scale of 1-10)
- 23Interval datadata based on objective measuring scales (most powerful measure, used in parametric tests)
Informações
- Descrição
- A2 Psychology (Paper 2) FlashCards sobre statistical theory, criado por Emily Moss em 24-02-2017.
- Criado por
- Emily Moss
- Visualizações
- 2
- Avaliação