inference about one prop Flashcards
point estimates
estimates of our population parameters
use the sample mean to estimate the population mean
use sample standard deviation to estimate the population standard deviation
use the sample proportion to estimate the population proportion
point estimates don’t account for sampling variability —> samples vary
symbols for sample mean, population mean, sample standard deviation, population standard deviation, sample proportion, and population proportion
sample mean : x bar
population mean : weird u
sample standard deviation : s
population standard deviation : beta
sample proportion : phat
population proportion : p
what is the basis of hypothesis testing
whether the point estimate is
1. typical value ( could have happened by chance alone)
2. unusual (happened other than chance)
null hypothesis
Ho
“nothing of interest” statement
any differences we see in the sample results is due to chance alone
always contains equal sign
p-value
the probability of obtaining a value of the statistic as extreme or more extreme as the observed statistic when the null hypothesis is true
gives a sense of how unusual our study results are
this is the number obtained from looking at number of simulations that simulate the same sample proportion as the observed divided by the nubmer of total simulations
Alternative hypothesis
Ha
“something interesting” statement
differences between sample results compared to status quo is difficult to explain as due to chance alone
small p-value
there is evidence that the observed results aren’t by chance alone when null hypothesis is true
reject null
the smaller the p-value, the stronger the evidence against the null hypothesis
large p-value
not enough evidence to reject the null
dail to reject the null
p-value small enough to reject null
there is statistically significant enough evidence against the null hypothesis when the p-value is less than some reference value a
a. 0.01, 0.05, 0.10
if p-value is < a, reject null
two sided hypothesis
Ha has p not equal to a number (so not p<n>n) meaning values can be greater or lesser than the population proportion p-value and reject the null</n>