Session #6 Flashcards
what are the two types of quantitative data?
continous
discrete
type of quatintative data in which all values are possible in a range
continuous
ex: shear strength of porcelain
type of quantitative data in which only certain valuess are possible in a range
discrete
ex: possible number of teeth someone has
what are the two types of categorical data?
nominal
ordinal
type of categorical data where data falls into category, but no order to data
nominal
ex. presence/absence of oral cancer, or race/ethnicity
type of categorical data where that has a specific order to it
ordinal
how often do you brush your teeth? (never, seldom, always)
which is more sensitive to extreme values, the mean or the median?
mean
measure of how much the individual data points vary around the MEAN
standard deviation
describing if there is a linear relationship btw an independent variable (X) and a dependent variable (Y)
correlation
the value of a correlation coefficient lies between what numbers?
- 1 and 1
* the closer (r) is to 1 or -1 the stronger the relationship
the fraction of variation in Y explained by X
square of the correlation
the higher the r-squared the better the fit of the regression line
an explanation of certain observations
hypothesis
what do we use hypothesis testing for?
to tell if what we observe in the population is consistent with the hypothesis
what is the null hypothesis?
states that there is NO DIFFERENCE btw two groups being compared or NO EFFECT of a product or intervention
what is the alternative hypothesis?
- this is the one the researcher thinks is the “truth”
- states that THERE IS A DIFFERENCE btw two groups being compared or an effect of a product or intervention
can be directional or non-directional
the population mean for group 1 is the same as the population mean for group 2
H0 interpretation
the population mean for group 1 is different than the population mean for group 2
Ha interpretation
type 1 error
rejecting the null hypothesis that is actually true in the population
the level of statistical significance (alpha) is commonly set to _____ and is interpreted as what?
***this is only in type I errors
- o.o5
- the max chance (5%) of incorrectly rejecting the null hypothesis when it is actually true
type II error
failing to reject the null hypothesis that is actually false in the population
beta
the probability of type II error
power
1-B and is related to the sample size used in the study
the probability, assuming that the null hypothesis is true, of seeing an effect as extreme or more extreme than that in the study by chance
p-value
- reject the null hypothesis is P-value less than or equal to alpha
- fail to reject if P-value is more than alpha
—-the higher the p-value, the more leeway you are giving yourself to be wrong by chance
a range of values about a sample stat that we are confident that the true pop parameter lies
confidence interval
most common = 95%
statistical test that can be used to determine whether the mean value of continuous outcome variable differs significantly between two independent groups
t-test
type of t-test that can be used when the outcome variable of interest is only being examined in one group
one-sample t-test
type of t-test that can be used when subjects are matched in pairs and their outcomes are compared within each matched pair (including where observations are taken on the same subjects before and after giving and intervention)
matched-pair t-test
when examining categorical data, this test can be used to compare the proportion of subjects in each of two groups who have a dichotomous outcome
chi-squared test
statistical method that allows for comparison of several population means
ANOVA
***IS USED FOR CONTINUOUS VARIABLES WITH MORE THAN TWO GROUPS
what type of statistic does an ANOVA test use?
F-statistic
Determining if findings are important from a clinical standpoint
clinical significance
probability that chance is responsible of an observed difference
statistical significane
does a p-value say anythings about clinical relevance or quality of a study?
no
what are the main limitations of statistical inference?
- only tells about the role of chance or random error in making inference from your study population to the source populaiton
- does NOT tell about the role of bias or confounding
- STATISTICS DO NOT TELL YOU ABOUT CAUSALITY
systematic error in design, conduct or analysis of a study that results in a mistaken estimate of an exposure’s effect on disease
bias
systematic error in selecting subjects into one or more groups, such as cases and controls, or exposed or unexposed
selection bias
errors in procedures for gathering relevant info
info bias
situation in which a non-causal association btw a given exposure and an outcome is observed as a result of the influence of a third variable
confounding variable
what two things make a variable a confounder?
- it is a known risk factor of the outcome
- it is associated with the exposure but is not the result of the exposure
is confounding an “all or none” phenomenon?
no