Question paper Flashcards
In a normal population 68 percent of the population lie with in one standard deviation of the mean
true
the standard error of mean is usually greater than standard deviation
false
the SD is always larger because SE = SD/ root of n
five number summary comprises of
range (max,min) ,median, upper and lower quatriles
if the units of measurements are chanhed from mmol/l to mol/l, the t statistic will not change
true
the parameters underlying different samples from the same population will be different
false
if a 95 percent CI for the difference b/w two means of two normal populations does not contain 0 then a test of null hypothesis that the means are equal will give p>0.05.
false
if the 95% CI of the ratio contains the value 1, the p-value will be
gre
a study comparig two groups wrt the proportion of patients cured in each group should use a chi 2 test
true
a p-value <0.05 proves that the null hypothesis is true
false
less means significant and hence reject null hyothesis
a regression analysis allows the population mean of the response variable to vary according to the value of the independent variable
true
fishers exact test can be used to
compare 2 grps of unpaired binary data when some expected counts are below a given threshold
for small values
in a paired T test
the samples within the pairs are assumed to have ND
in a regression equation x and y are
independant and dependant variables
the symmetry of a normal population implies that the mean is equal to the median
true
chi 2 test can be aplied to a table of percentages
false
frequencies not percentages
when comparing the measurements made by two machines for the same continuous sets of variables, it is important to use a statistical method that reports the correlation b/w the two sets of measurements
false
because same continuous set of variables
