Biostatistics for Dentistry Flashcards
True or False: Statistics are important because they allow us to understand information and make clinical decisions based on data.
True
What are two types of data?
- Quantitative
2. Categorical
Quantitative data can be divided into ____ and ____.
Continuous
Discrete
Continuous data has values that are ____.
all possible, no set range
Discrete data has values that are ______.
only possible within a range
Categorical data can be subdivided into _____ and _____.
Nominal
Ordinal
What is Nominal data ?
data falls into a category but has no order (race/ethnicity)
What is ordinal data?
data has a specific order within a category
never, sometimes, always
What are four ways to describe quantitative data?
Mean
Median
Mode
Standard Deviation
The ____ is sensitive to extreme values.
Mean
The ____ is less sensitive to extreme values.
Median
____ is the measure of how much the individual data varies around the mean.
Standard Deviation
What are three ways to describe categorical data?
- frequency
- Percentage
- Correlation
Correlation shows whether there is a _____ between an independent variable (x) and dependent variable (y).
linear relationship
A correlation coefficient (r) can lie between ___ and ___.
-1 and +1
A positive correlation coefficient value indicates ____.
as independent (x) increases,, dependent (y) increases
A negative correlation coefficient value indicates ____
as independent (x) increases, dependent (y) decreases
The closer (r) is to +1 or -1, the ______ the relationship.
stronger
The square of correlation (r^2) is the fraction of ______ in Y explained by X.
variation
ex. if r =0.9 then r^2 = 0.81
The higher the r^2 value, the _____ the fit of the regression line.
better
A _____ is an explanation for certain observations
hypothesis
The null hypothesis usually states that ___________.
there is no difference between two groups being compared or no effect of a product or intervention
What is the alternative hypothesis?
the one the researcher believes to be the truth; usually states that there is a difference between two groups being compared or an effect of a product
Results of testing a hypothesis can be ______ (1>2) or ______ (1=2)
directional
non-directional
Is the following statement null or alternative?
=the population mean for men is the same as the population mean for women.
Null (there is no difference)
What is a Type I error?
rejecting the null hypothesis when it is actually true
trying to act like there is a difference when there is NOT one
What is the level of statistical significance for Type I error?
alpha
Alpha is commonly set to ______ and is interpreted as the maximum chance (____%) of incorrectly rejecting the null hypothesis when it is actually true.
0.05
5%
What is a Type II error?
failing to reject the null hypothesis when it is actually false in the population
Saying there is no difference when there actually is a difference
The probability of a type II error is described as _____.
beta
How is “power” calculated?
1 - beta
True or False: Power is related to the sample size.
True
What is the “p value”?
the probability, assuming the null hypothesis is true, of seeing an effect (as extreme or more extreme than that in the study) by chance
_____ the null hypothesis if the p-value is LESS THAN OR EQUAL TO alpha.
Reject
Fail to reject the null hypothesis if the p-value is _____ than alpha.
greater
What are confidence intervals?
a range of values about a sample statistic that we are confident about that the true population parameter lies
What is the most common confidence interval?
95%
Confidence interval = 95%
If the data collection and analysis is repeated over and over, the confidence interval will _____ the correct value 95% of the time.
include
What are three ways to test statistics?
t-test
Chi Square
Anova
What is a t-test and when is it used?
a test used to determine whether the mean of a continuous outcome variable differs significantly between two independent groups
-used for a continuous outcome
True or False: When using a t-test, the alternative hypothesis may be directional or non-directional.
True
The ____ ____ t-test can be used when the outcome variable of interest is only being examined in one group.
One-sample
The ____ _____ t-test can be used when subjects are in pairs and their outcomes are compared within each pair (including where observations are taken on the same subjects before and after a given intervention).
Matched-pair
True or False: A t-test can measure up to 3 groups.
False, two groups only
Ex. Null: women = men; alternate: women don’t = men. alpha= 0.05, p-value-0.006.
What can be concluded?
p < alpha
reject the null hypothesis: conclude that there is a difference between men and women
What is a chi-squared test?
a test used during examination of CATEGORICAL data to compare the proportion of subjects in each of TWO groups who have a dichotomous outcome
ex: periodontitis in diabetics vs non-diabetics
null: no association, alternate: association present; if p-value is less than alpha, reject the null
What is an ANOVA used for?
Analysis of Variance:
a statistical method that allows for comparison of several population means (more than two!)
What is the null hypothesis when comparing 3+ groups? When using ANOVA, when can you reject the null hypothesis?
Null: means of all groups are equal
Reject when the p-value of F-Statistic is less than or equal to alpha
True or False: ANOVA is the only analysis that uses p-value.
False, it is the only one to use F-statistic
True or False: A p-value tells about clinical relevance and study quality.
False!
P-values have _____ significance, but not _____ significance.
statistical
not clinical
Statistical inference only tells about the ______ in making inference from your study population to the source population.
role of chance or random error
True or False: Statistics do not tell about causality.
True
What is bias?
systemic error in the design, conduct or analysis of a study that results in a mistaken estimate of an exposure’s effect on disease
Two important types of bias: _____ and ____.
selection bias- choosing patients
information bias- gathering info
A situation in which a non-causal association between a given exposure and an outcome as a result of the influence of a third variable.
Confounding
Confounding can be either a _______ or ____.
confounding variable
confounder
A variable is confouding if it is a known ______ of the outcome or if it is associated with ______.
- known risk factor of the outcome
2. associated with the exposure but not the result of
How do you evaluate confouding?
assess the measure of association within strata
True or False: Confounding can lead us to conclude a causal relationship when there is none.
True
