Jazmeen Discussion
Hypothesis testing is used to assess the plausibility of a hypothesis using sample data. A null hypothesis states that there’s no significant difference or effect between specified variables, while an alternative hypothesis claims that there’s an effect in the variables (Gerstman, 2015). My selected research question is, to what extent does gender influence the length of hospital stays for MI patients?
Null hypothesis: The length of hospital stay (LOS) for MI patients is not influenced by gender.
Alternative hypothesis: Gender influences the LOS of hospitalization for MI patients.
Statistical null hypothesis: H0: μ1 = μ2
Statistical alternative hypothesis: Ha: μ1≠ μ2 . This could also be represented as Ha: μ1 > μ2 or Ha: μ1 < μ2.
Alternative phrasing
Null hypothesis: there is no correlation between gender and LOS for the hospitalization of MI patients.
Alternative hypothesis: There is a correlation seen between gender and LOS of hospitalized MI patients. You could also predict that female MI patients will have a longer LOS, or vice versa for males.
References
Gerstman, B. (2015). Basic Biostatistics: Statistics for Public Health Practice (2 ed.). Burlington, MA: Jones & Bartlett Learning.
Jackie Discussion:
4-1 Discussion: Choosing Hypotheses
Null hypotheses are best defined as there is no specific difference between variables, whereas, in alternative hypotheses, there is a difference/ relationship between the variables (Gertsman, 2015).
My chosen health question: To what extent do the ages of MI patients vary by gender?
The null hypothesis for this question is as follows; there is no extent to which the ages of MI patients vary by gender.
The alternative hypothesis is; there is an extent to which the ages of MI patients vary by gender.
Statistical Notation:
Null Hypothesis: H0 = µ1 = µ2
Alternative Hypothesis: Ha = µ1 ≠ µ2 OR µ1 ≤ µ2 OR µ1 ≥ µ2
There is also another which I could word the hypotheses whilst maintaining the same meaning as my original wording and will support this with alternative wording.
Null: there is no direct association between gender and the ages of MI patients.
Alternative: There is a direct association between gender and the ages of MI patients. For example, one could anticipate that males over 50 have more risk of having a MI than females over 50, whereas females over 50 could have more risk of having a MI over males.
References
Gerstman, B. (2015). Basic Biostatistics: Statistics for Public Health Practice (2 ed.).
Burlington, MA: Jones & Bartlett Learning.
