Topic 15: Research Methodology: Data analysis: Inferential analysis Flashcards
1
Q
What is an inference?
A
A statement about a population based on information contained in a sample from that population.
2
Q
What are the 2 types of hypotheses used?
A
- The null hypothesis(H0) stating that there is no relationship/effect/difference between the dependent and independent variables.
- The alternative hypothesis(Ha) will always then state the suspected relationship between the DV and IV.
3
Q
What are the 2 types of univariate hypotheses testing?
A
- Relational hypothesis testing- how does change in DV vary with change in IV? (use if both DV & IV are continuous data)
- Hypotheses about differences- how does DV vary from one group to another(IV) (use if DV is continuous data, while IV is categorical)
4
Q
What are Type 1 (alpha) and Type 2 (beta) errors?
A
- Type 1 errors occur when a True null hypothesis is rejected as false (study finds there is a relationship when in reality there isn’t; false positive) (There should be less than 5% chance of that happening)
- Type 2 errors occur when a False null hypothesis is accepted by the study as true (study finds there is no relationship when in reality there is; false negative) (There should be less than 20% chance of that happening)
5
Q
What are the 12 steps of inferential statistical analysis?
A
- State practical problem
- State statistical problem (identify DV, IV & objectives)
- State the null hypotheses(H0)
- State alternative hypotheses(Ha)
- Select tolerable alpha level (usually 5%)
- Select tolerable beta level (usually 20%)
- Determine sample size
- Select appropriate hypotheses tests (beyond scope of module, p-value is given)
- Run appropriate hypotheses tests (beyond scope of module, p-value is given)
- Interpret statistical finding (If p-value>=alpha value(5%) then H0 stands. If p-value<alpha value(5%) then reject H0 and accept Ha)
- Interpret practical results and link back to Step 1
- Write conclusions
