Inferential Statistics Flashcards
What are descriptive statistics?
- Descriptive statistics-
- Descriptive statistics summarizes or describes the values of a data set and describe any difference in results obtained.
what are three categories of descriptive statistics?
*Descriptive statistics consists of three basic categories of measures: measures of central tendency, measures of variability (or spread), and frequency distribution.
*Measures of central tendency describe the centre of the data set (mean, median, mode).
*Measures of variability describe the dispersion of the data set (range, standard deviation).
*Measures of frequency distribution describe the occurrence of data within the data set (count).
What is Inferential test of significance?
Show if there’s a significant causal effect/cause-and-effect relationship between IV and DV.
Rest on concept of probability of data being due to random chance factors or something else, determines whether to accept null hypothesis because it tells us whether results were due to chance or not.
When should alternative hypothesis be accepted?
Accept the alternative hypothesis if there’s a significant causal effect (Normally 0.5) between conditions (IV and DV) instead of being due to variation or other factors.
When should null hypothesis be accepted?
Accept the null hypothesis if the inferential tests show that there’s not a significant difference
What is probability and chance?
Probability-Likelihood of event occurring
Likelihood that results we generate are due to chance
Chance- Level of significance, how sure we want to be
What does it mean if Null hypothesis is accepted?
If the probability of the results is due to chance we are assuming that difference between data is too small to be significant and show real effect, null hypothesis would be maintained/ accepted.
What does it mean if alternative hypothesis is accepted?
If the probability of the results is due to real difference/relationship we’re assuming that difference between data is significant enough to show real effect, maintained/ accepted alternative/experimental hypothesis
What is the normal level of significance and what does inferential test of significance tell us about it?
The normal level of significance (probability) in psychology is equal to or less than 0.05/ 5%.
This means. That we accept results if there’s a 5% (or less than) probability that they are due to chance
The inferential test of significance tells us whether it meets 0.05/ 5% probability threshold or not.
Will tell us if we met level/threshold of significance based on results
When is Experimental Hypothesis accepted?
If the inferential test shows that our results are significant and that there’s causal effect between IV and DV, we can accept experimental/alternative hypothesis due to 95% confidence in accuracy of prediction, 5% confidence that results are due to chance.
When is Null Hypothesis accepted?
If the inferential test shows that our results are not significant we can accept null hypothesis due to 95% confidence in inaccuracy of prediction (That it’s wrong) and that results are due to chance or 5% confidence in accuracy.
What is Type 1 Error?
A type 1 error means you have incorrectly concluded that alternative hypothesis is correct when results lacked significance, meaning that null hypothesis should be accepted. May occur due to level of significance being higher than 0.05 and therefore experimenter may conclude/ support alternative hypothesis. For example- Level of significance may be 0.1 which is too lenient, leads to 10% of error/chance. Normally it’s 5% error/chance
What is type 2 error?
A type 2 error means you have incorrectly concluded that alternative hypothesis isn’t correct, accepted null hypothesis. If result of a test is equal to or less than 0.01, probability of chance should be accepted as it’s highly significant, if we set 0.01 as accepted level of significance, a number of alternative hypotheses will be rejected when there was real effect.
May occur due to level of significance being lower than 0.05 (Stricter level of significance) and therefore experimenter may support null hypothesis and reject alternative.
For example- Level of significance is 0.02 which is stricter, leads to 2% chance of error/chance. Normally it’s 5% error/chance
