Module 2 Flashcards
What value do probabilites have to be between?
- 0 and 1
- 0 and 100%
- 10:1
What does a probability of 1 mean?
- the event is certain to occur
What does a probability of 0 mean?
- the event is certain not to occur
What is the probability distribution?
- describes the probability of every event
- applies to both categorical and continuous data
What must the total area under a normal curve equal?
- 1 or 100%
What is standard normal distribution?
- normal distribution with mean = 0 and sd = 1
What happens to the normal distribution curve when the mean increases?
- moves across the x axis indicating a higher mean
What happens to the normal distribution curve when the sd decreases?
- the distribution becomes skinnier implying a smaller range in values
What are the nice rules for conversion of normal distribution?
- 68% of values are within 1 sd of mean
- 95% of values are within 2 sd of mean
- 99.6% of values are within 3 sd of mean
What is statistical inference?
- make inferences about population based on data from sample
What is sampling variability?
- variation in results from different/random samples of the same population
To make an inference about the population does the sample need to be representative?
- yes
What happens to a histogram when the sample size increases?
- becomes narrower
- mean becomes less susceptible to single values
- sd decreases
What is standard error (se)?
- standard deviation of the sample means
- measures amount of variability in the sample mean
What happens to standard error as the sample size increases?
- decreases
What is the se equation?
se= standard deviation of sample/ square root (size of sample)
What is standard deviation
- measures amount of variability in sample
What is the central limit theorem?
- if raw data is not normally distributed, means of samples taken from the distribution will be normally distributed
What should results be reported with?
- caveat
e. g. based on a study og 1000 mothers from Brisbane
Is it easier to prove or disprove a hypothesis?
- easier to find evidence against a hypothesis
What is a Null Hypothesis (H0)?
- nothing is happening
e. g. the new drug doesn’t change blood pressure
What is an alternative hypothesis (H1 or Ha)?
- something is happening
e. g. the new drug changes blood pressure
What is a one-tailed hypothesis?
- has a direction
e. g. new drug lowers blood pressure
What is a two-tailed hypothesis?
- doesn’t have a direction, just a change (could be positive or negative)
e. g. new drug changes blood pressure
What type of hypothesis does research questions have?
- alternative
What are the 6 steps to hypothesis testing?
- state null and alternative hypothesis
- collect data
- summarise data appropriately
- what would data look like if the null hypothesis was true
- calculate discrepancy between groups
- make decision and conclusion
What is test statistic?
- discrepancy between the data and what is expected under the null hypothesis
what is the equation for test statistic?
(observed-expected)/precision
How do you calculate an overall estimate of se using both samples?
= square root (se1 squared + se2 squared)
How does the value of the test statistic determine?
the p-value and ultimatelt whether the difference is statistically significant
What is the p-value/significant level?
- probability of observing the data if the null hypothesis is true
- lower p-value indicate the null hypothesis is less plausible
What p-value is said to make results statistically significant?
5% or <0.05
What is a true positive?
- rejected null hypothesis and shown true effect of H0 is not true
e. g. guilty in jail
What is a type 2 error/false negative?
- the null hypothesis is not rejected but it is not really true
- <0.2
e. g. guilt but no jail
What is a type 1 error/false positive?
- the null hypothesis is rejected but there is evidence it has a true effect
- <0.05
e. g. not guilty in jail
What is the power equation?
= 1 - type 2 error
What is a confidence interval (Cls)?
- specify boundaries within which we believe the population values will fall
What is the equation for 95% Cl for the mean?
= mean + or - 1.96xse
interpretation: there is a 95% probability that this interval contains the true population mean
What is 99% of the standard normal distribution within?
-2.58 to 2.58 se of the mean
What is 95% of the standard normal distribution within?
-1.96 to 1.96 se of the mean
What p-value determines when the Cl must be reported?
p<0.05
What hypothesis does the p-value tell you about the likelihood of?
null hypothesis
