Studied by 0 people
alpha level
robability of finding your result due to chance, the probability of committing a Type 1 error
type one error
probability of finding your result when the null hypothesis is true.
Critical value
a number you can look up in a t or f table that signifies the boundary of accepting/rejecting the null hypothesis based on parameters of your data and your chosen a level
P value
the probability of finding our result when the null hypothesis is true
R value- measures the strength and direction of a linear relationship between two variables
R2 value
measures how well the regression line represents the data, the proportion of the variance in the dependent variance in the dependent variable that can be predicted or explained by the independent variable.
Standard error
measures how closely the population mean varies around your sample mean.
Type one error
It is a false positive, the probability of a type one error is equal to alpha level, rejecting null hypothesis when it is true, finding a difference when there isn’t on in reality.
The variation can be found from the r2 value- if the r2 value is 0.247
then the variation will be 24.7%.
If p is less than 0.05 (significant value)
This is true positive but there is still risk of committing type one error
if p value is more than 0.05 which is alpha
This is true negative because p value is greater than the alpha, we could be showing a type 2 error
As power increase
the probability of type 2 error decreases
type one error
false positive
type two error
false negative
probability of avoiding a type 2 error
power, it is also the probability of making a correct decision. More power is needed to detect smaller
things that impact power
More power is needed to detect smaller differences. Smaller sample size requires a large amount of power to detect differences. The higher the sample size, the less power we need. Lower sample size means more power. Sample variance is deviation. Smaller variance means higher amounts of power. More sample variance will lower our power. Alpha is inversly related.
type one error
rejecting null hypothesis when it is true
type one error
it is a false positive, finding a positive when there isn’t one in reality
type one error
the probability of a type one error is equal to alpha level
risk of committing type one error [reject null, truth about null hypothesis is true]. ALPHA
observe difference when none exist
risk of committing type two error [accept null, truth about null hypothesis is false]
fail to observe difference when one exist.
increasing the standard deviation (variance) of a data set will serve to
decrease power
Researchers are looking at the effects of exercise on the strength of the biceps brachii. Fifteen subjects are randomly divided into 3 groups: no exercise, low dose (3x per week for 3 weeks) and high dose (6x per week for 6 weeks).
The statistical analysis yields a p value of 0.15. What should the researchers be concerned about?
Not enough power because the P value of 0.15 means there is a 15% chance that the observed results are due to random variation rather than true effect.
power
is probability of correctly rejecting a false null hypothesis. An increase in variance (or standard deviation) generally leads to more spread out data, which can make it harder to detect a true effect, thus decreasing power
Increasing the variance of data set typically
decreases the power of a statistical test because it increases the noise relative to the signal. THE POWER INCREASES WITH LARGER SAMPLE SIZES AND LOWER VARIABILITY.
What does the lack of statistical significance suggest in relation to Power?
The study may not have enough power to detect a true effect
larger effect size
This means larger differences between the groups being compared. This increase power.
What does Lower variability (smaller standard deviation) within groups do in relation to power?
It increases power
Increasing the sample size of a data set will serve to
increase power. This is because increasing the sample size reduces variability in estimates, leading to more precise results and increased statistical power, larger sample sizes make it easier to detect true effects, increasing the likelihood of rejecting a false null hypothesis0 INCREASE POWER
increasing the power of a test?
allows researchers to detect smaller differences
Decreasing the standard deviation (variance) of a data set will serve to?
increase power
increasing the variability of the data and decreasing the number of subjects tested will always?
decrease the power of a statistical test.
The power of a test
The ability to reject the null hypothesis when it is false, The ability to find an effect when an effect exists, The ability to detect significant differences
what are things that will always increase the power of a statistical test?
increase sample size, increase difference between the means, decrease variability
