Unit 1 - Inferential Statistics Flashcards
What is a normal distribution curve?
Is when all measures of central tendency occur at the highest point in the curve such that half the results are above this midpoint and half are below it. It will be what is expected from a normal target population, covering a range from lowest to highest.
How does normal distribution of data link to standard deviation?
Standard deviation is a measure of dispersion it assumes that there has been a normal distribution of data.
How does normal distribution of data link to the use of parametric inferential statistical test?
Having normally distributed data is one of the three conditions that need to be satisfied for a research to be able to use parametric inferential statistical test.
What is a skewed distribution curve?
If the distribution was representing data from skewed or unusual data set. The mean would be different from mode…
What is a negatively skewed distribution?
If the mean is to the left of the mode (i.e. lower than the mode score) then there is a negative skew to the results.
What is a positively skewed distribution?
If the mean is to the right of the mode (i.e. higher than the mode score) then there is a positive skew to the results.
What is probability?
Is the likelihood of something happening, having an idea of the probability of IV on the DV.
What is the standard level of significance?
P < 0.05 - the probability of results being due to chance is less than 5%. So their alternative hypothesis is supported.
What happens if the standard level of significance is NOT reached?
It would be expressed as P > 0.05, so their null hypothesis would be supported.
What happens if the standard level of significance is EXCEEDED?
It would be expressed as P < 0.01 (99%) so their alternative hypothesis is strongly supported.
What is a type 1 error?
(false positive). This happens when the alternative hypothesis is accepted and the null hypothesis is rejected, and the behaviour shown was really due to chance.
What is a type 2 error?
false negative). This happens when the rejection of the alternative hypothesis and the acceptance of the null hypothesis, when really the independent variable is having a significant impact on the dependent variable.
What does this symbol ‘=’ mean?
Equal.
What does this symbol ‘
Less than.
What does this symbol ‘
Much less than.
What does this symbol ‘>’ mean?
Greater than.
What does this symbol ‘»’ mean?
Much greater than.
What does this symbol ‘∝’ mean?
Proportionality = a correlation has a linear property.
What does this symbol ‘~’ mean?
Approximately.
Why are statistical tables of critical values used?
To compare the calculated value, based on their data, with the critical values mathematicians have identified to be the value at which data is significant.
What factors impact critical values?
Significance level (P < 0.05), sample size (n) and sometimes whether research is investigating a one or two tailed hypothesis.
What is the criteria for parametric tests?
1) the data is interval or ratio
2) the data is normally distributed
3) similar variances (homogeneity of variance)
What are the parametric tests?
- Unrelated t-test
- Related t-test
- Pearson’s Product Moment
What are the non-parametric tests?
- Chi-Squared
- Binomial Sign Test
- Mann Whitney U Test
- Wilcoxon Signed Ranks
- Spearman’s Rho
What is the key criteria for non-parametric tests?
- What level of data is being used (nom, ord, int)
- If it is a test of difference or correlation
- If it is a test of difference, is it repeated or independent design.