inferential testing Flashcards
3 factors to think about when choosing inferential statistical test
- the design for the study
- the level of data
- whether a difference of correlation is being measured
how would you know if the test if an unrelated design
if the test an independent group design
how would you know if a test is a related design
if its either a repeated measures design or a matched pairs design
what are the 8 situational tests
-sign test
- Wilcoxon
- related t-tests
- unrelated t-tests
- Chi-squared test
- Mann-Whitney
-Spearman’s Rho]
- Pearson’s R
when will you use a sign test
if its producing Nominal Data and is either used in matched pairs design or a repeated measures
when would you use a Wilcoxon test
if the data is ordinal and its either a matched pairs or repeated measures design
when would you use a related test
data that is been shown as interval, and either shown in a repeated measures or a matched pairs design
when would you use a Chi-Squared test
data that is being shown as nominal, an unrelated test, i.e. independent measures design or through the test of correlation
when would you use a Mann-Whitney test
data that is ordinal, the test of difference through an independent design (unrelated design)
when would you use a unrelated test
- interval data being used
- test of difference through a independent measures design
when would you use Spearman’s Rho
- data is being ordinal
- used to test correlation
when would you use Pearson’s R
- data that is interval
- through testing correlations
what is the critical value
the numerical boundary that stands between accepting or rejecting the null hypothesis when a hypothesis is being tested
define significance
how sure you are about a correlation or difference existing
define probability
how likely is is for an event to happen
what is the usual level of significance used within probability value tables
- 0.05%
- the p value can be usually equal or less than 0.05%
what are type 1 errors
- known as optimistic
- incorrect rejection of the null hypothesis which is actually true ( a false positive)
what are type II errors
- pessimistic
- the failure to reject the null hypothesis that is false (a false negative).
steps to work out a critical value tables
1) identify hypothesis, one-tailed or two tailed
2) select the significance level relevant to the hypothesis
3) find N or DF
4) identify critical value
5) read table rule, located underneath the table
6) compare the calculated value with the critical, then write a statement about it
