Hypothesis Flashcards
Mastery
- Null means…
- The null hypothesis, AKA …
- The null hypothesis states that you expect…
- You either … or … a hypothesis…
- A hypothesis is not a theory but could … if …
- Null means empty, no value, or NO CHANGE.
- The null hypothesis, AKA the default hypothesis
- The null hypothesis states that you expect no change, or the currently value is accepted.
- You either accept it or reject a hypothesis…
- A hypothesis is not a theory but could become one if given sufficient supporting evidence.
- The Alternative Hypothesis states…
- The stated hypothesis in a study is…
- You then either … or … the alternative hypothesis based on…
- The Alternative Hypothesis states something is, will, or has , changed.
- The stated hypothesis in a study is normally the alternative
hypothesis, or Alt Hypothesis, or - You then either accept or reject the alternative hypothesis based
on the data you gather in your study
Reject the null
* α = …
* Up to you…
* α>.05 - 1 in 20…
* α> 0.01 - 1 in 100 …
* α > 0.001 - 1 in …
* α≈ 2.87×10−7 or 1 in 3 million (standard for theoretical physics) 5 sigma
- α = acceptable error.
- Up to you to define:
- α>.05 1 in 20 (standard level in physiology and psychology)
- α> 0.01 1 in 100 (more statistically rigorous, better physiology standard)
- α > 0.001 1 in 1000 (typical new engineering standard)
- α≈ 2.87×10−7 or 1 in 3 million (standard for theoretical physics) 5 sigma
- Both Alternative and Null Hypothesis can also be stated with an equation
H-alpha*: μ1 < μ2
H-alpha* : μ1 > μ2
* Reads:
H-nought* : μ1 = μ2
* Reads:
H-alpha* : μ1 > μ2
* Reads: the alternate hypothesis is that the population
mean is GREATER than the expected value
- Reads: the alternate hypothesis is that the population
mean is LESS than the expected value
H-nought* : μ1 = μ2
* Reads: the null hypothesis states that the population
value will be equal (no change) from the expected
value.
Does consumption of dark chocolate improve cardiovascular health.
* What would your Null Hypothesis from this statement be?
*
* What would your alternate Hypothesis be?
*
- What would your Null Hypothesis from this statement be?
- People in my sample who consume dark chocolate will have the same cardiovascular health scores as people who received a placebo
- What would your alternate Hypothesis be?
- People who consume dark chocolate will have improved cardiovascular health
scores when compared with placebo
- A type I error is…
- Fire alarm … and …
- You indicate you…
This is referred to as … (the chance of making a
type I error)
We use alpha to set our P value.
Example: an α level of 0.05 means there is a …% chance of rejecting the null hypothesis when it is true
- A type I error is the rejection of a true null hypothesis (you state something has changed when it did not)
- Fire alarm rings and there is no fire.
- You indicate you detect a change (and rejected the null hypothesis) when a change did not occur
This is referred to as or **alpha𝞪 **
(the chance of making a type I error)
We use alpha to set our P value.
Example: an α level of 0.05 means there is a 5% chance of rejecting the null hypothesis when it is true
- A type II error is…
- Fire alarm … when…
- Something has….
- WHAT GREEK SYMBOL
AND WHAT DOES IT CALCULATE
- A type II error is the failure to reject a false null hypothesis (you state something did not change, but it did)
- Fire alarm fails to ring when there is a fire
- Something has change but you did not recognize the
change - β or Beta is the calculated likelihood of creating a type II error (not recognising a change when it occurred)
- Alpha and Beta are related but calculated differently
Lactate supplementation for concussion victims
H- ALPHA:
H - NOUGHT:
H- ALPHA: Hypothesis: ‘Lactate supplementation will improve symptoms for people who have suffered a
concussion’
H - NOUGHT: Null Hypothesis: ‘lactate supplementation will have no effect on concussion symptoms’
- Rough Alternate Hypothesis: ‘Lactate supplementation will decrease recovery time for people who have suffered a concussion’
- Refined alt. Hypothesis: (Stated or test hypothesis)
- :
- Null Hypothesis
- ‘
- Rough Alternate Hypothesis: ‘Lactate supplementation will decrease recovery time for people who have suffered a concussion’
- Refined alt. Hypothesis: (Stated or test hypothesis)
- : Administering supplemental sodium lactate to individuals who have visited a hospital for a concussion, will decrease head and neck pain, dizziness, and sensitivity to light and noise, as reported by standardized survey, when compared with regular treatment protocols (RTP).
- Null Hypothesis
- ‘Supplemental lactate provides no improvement when compared to RTP. The mean outcomes will not differ between the two treatment groups.’
- Was there an intervention?
- Yes, sodium lactate as a treatment
- Was there a control group?
- Yes, one group received standard care and nothing else
- Was the control group randomly determined?
- Yes, people were placed in one or the other groups randomly
- This would be classified as an …
BUT NOT TRUE, WHY
- Was there an intervention?
- Yes, sodium lactate as a treatment
- Was there a control group?
- Yes, one group received standard care and nothing else
- Was the control group randomly determined?
- Yes, people were placed in one or the other groups randomly
- This would be classified as an RCT, but it is not a “true” RCT as the exposure to concussion was not (could not) be controlled.
- Survey all children in sports program correlate concussion with type of sport.
- Follow children in different sports programs over five years measure occurrence of concussion.
- Find children / teen agers who have experienced concussion determine what sports they played, match them with peers without concussion from same sport and others.
Determine what they experienced in the past that contributed to their concussion that
matched peers did not.
do rugby players get more concussions than soccer players?
- Survey all children in sports program correlate concussion with type of sport.
- Cross sectional survey a slice in time
- Follow children in different sports programs over five years measure occurrence of concussion.
- Cohort longitudinal, prospective, five year time frame
- Find children / teen agers who have experienced concussion determine what sports they played, match them with peers without concussion from same sport and others. Determine what they experienced in the past that contributed to their concussion that
matched peers did not. - Case-control, retrospective, backward in time
- Hypothesis The sport you participate in relates to your likelihood of experiencing a concussion
- Null Hypothesis :
Can we do an intervention? ….
We must do …. study
Can we group subjects?
What is the temporality? ….
- Hypothesis The sport you participate in relates to your
likelihood of experiencing a concussion - Null Hypothesis : Sport participation does not relate to your likelihood of experiencing a concussion
Can we do an intervention? ×
We must do an observational study
Can we group subjects? - group people in rugby, soccer, basketball.
What is the temporality? - how does it move through time.
