Probability and Statistical Significance Flashcards
What are the two way studies can have errors?
Caused by chance (Random Error)
Not Causes by chance (Bias or Systematic error)
Random Error
Due to Chance
Does not bias a study
Systematic error
Due to Bias
Error that is inherent to the study method being used and results in a predictable and repeatable error for each observation
‘Statistically significant’
if a study result is unlikely to have occurred by chance
A Chance occurrence
Something that happens unpredictably without discernible human intention or with no observable cause: Caused by chance or random variation
Random Variation
there is an error in every measurement. If we measure something over and over again, we will get slightly different measurements each time and a few measurements may be extreme
Statistical inference
if we measure something only once, how sure are we that our measurement has been caused by chance
What are the two methods used for estimating how much random variation there is in a study and whether the result was likely to have been caused by chance?
Confidence Intervals
P-values
Confidence Intervals
Estimates how much random variation there is in our measurement
The range of values where the true value of our measurement can be found (assuming the study wasn’t biased)
For some measures it can estimate whether the measure was likely to have been caused by chance
P-values
Used to estimate whether the measure was likely to have been caused by chance or not
If the 95% Confidence Interval odds ratio DOES NOT INCLUDE ONE….
the odds ratio is statistically significant
When the odds ration is equal to 1
there is no association
Odds Ratio is greater than one and confidence interval doesn’t include one
Positive association
Statistically significant
Odds Ratio is greater than one, but confidence interval includes one
No association
Not statistically significant
Odds Ratio is less than one and confidence interval doesn’t include one
Negative association
Statistically significant
Odds Ratio is less than one and confidence interval includes one
No association
Not statistically significant
What value must the P-value be less than to be statistically significant?
0.05
Null Hypothesis
The ‘hypothesis of no association’ - this is no association between exposure and disease
Alternative hypothesis
The ‘research question’ - that there is an association between exposure and disease
If the p-value is less than 0.05
Reject the null hypothesis and accept the alternative hypothesis
If the p-value is greater than 0.05
Accept the null hypothesis and reject the alternative hypothesis
Type I (alpha) error: False Positive
Rejecting the null when it is NOT false
p=0.05
Type II (Beta) error: False Negative
Not rejecting the null when it is false (an association truthfully exists)
Conventionally set at 0.20
Categorical
The variable is broken into discrete categories
