Biostatistics Flashcards by Garrett Gilbert

what is the interquartile range (IQR)?

-middle 50% . . from 25th-75th percentile

How well did you know this?

Not at all

Perfectly

what is within 1 standard deviation?
2?
3?

middle 68%

95%
99.7%

How well did you know this?

Not at all

Perfectly

there is a positive skew if what?

negative?

mean>median

mean

How well did you know this?

Not at all

Perfectly

What test is used to assess for equal variances between groups in interval data?

Levene’s test

How well did you know this?

Not at all

Perfectly

if interval data is not normally distributed then what?

use an ordinal test then

How well did you know this?

Not at all

Perfectly

what are the types of comparison/assessment that can be used?

Correlation
Regression (predicts an outcome)
Survival comparison (as time progresses . . something over time)
Group Comparison

How well did you know this?

Not at all

Perfectly

what comparison provides a quantitative measure of the strength and direction of a relationship b/t variables
-values range from -1.0 to +1.0

Correlation

How well did you know this?

Not at all

Perfectly

Nominal correlation test?

Contingency coefficient

How well did you know this?

Not at all

Perfectly

Ordinal correlation test

Spearman correlation

How well did you know this?

Not at all

Perfectly

Interval correlation test

Pearson correlation

How well did you know this?

Not at all

Perfectly

what does a Kaplan meier Curve represent?

proportion of events over time . . so a survival comparison

How well did you know this?

Not at all

Perfectly

Nominal Survival test?

Log Rank test

How well did you know this?

Not at all

Perfectly

Ordinal Survival Test?

Cox-Proportional Hazards test

How well did you know this?

Not at all

Perfectly

Interval Survival Test?

Kaplan Meier test

How well did you know this?

Not at all

Perfectly

Nominal Regression test?

Logistic Regression

How well did you know this?

Not at all

Perfectly

Ordinal Regression test

Multinomial logistic regression

How well did you know this?

Not at all

Perfectly

Interval Regression test?

Linear Regression

Nominal Group comparison, 2 groups, independent data

(Pearson’s) Chi-squared or Fisher’s Exact

Nominal Group comparison, 2 groups, related data?

McNemar

Nominal Group comparison, 3 or more groups, independent data

Chi-squared or Fisher’s exact

Nominal Group comparison, 3 or more groups, related data

cochran

What is the Bonferroni test of inequality used for

when a nominal group comparison with 3 or more groups are compared and it is found that there is a difference

Ordinal Group comparison, 2 groups, independent data

Mann-Whitney

Ordinal Group comparison, 2 groups, related data

Wilcoxon Signed Rank

Ordinal Group comparison, 3 or more groups, independent data

Kruskal-Wallis

Ordinal Group comparison, 3 or more groups, Related data

Friedman

When is a Student-Newman-Keul, Dunnet, or Dunn test used

when an ordinal group comparison with 3 or more groups is done and there is a difference

Interval Group comparison, 2 groups, independent data

Student T test

Interval Group comparison, 2 groups, related data

Paired T test

Interval Group comparison, 3 or more groups, independent data

ANOVA or MANOVA

Interval Group comparison, 3 or more groups, related data

Repeated measures ANOVA or Repeated measures MANOVA

what are the interval tests done when there are 3 or more groups and a difference is found

- Bonferroni - Tukey - Scheffe - Dunn - Dunnett - Student-Newman-Keul

how do you decide b/t a Chi-squared and a Fisher's exact test

- Chi-squared used for cell counts > 5 | - Fisher's exact test used with EXPECTED cell count of <5

what are the buzz words to know if data b/t comparison groups are related

- pre vs post - before and after - baseline vs end

what is the correlation test showing relationship or agreement b/t evaluators (consistency of "decisions" and "determinations")

Kappa Statistic

describe the kappa interpretation

+1: the observers perfectly "classify" everyone exactly the same way 0: there is no relationship at all b/t the observers' "classifications", above the agreement that would be expected by chance - 1: the observers "classify" everyone exactly the opposite of each other

Describe a Type 1 error

Rejecting the Null hypothesis when it is actually true and you should have accepted it

Describe a type 2 error

Not rejecting the Null hypothesis when it is actually False and you should have rejected it

what does the p value represent?

the probability of experiencing a type 1 error | -in other words it is the risk of being wrong if we reject the null hypothesis

what is often used now in place of the P value and what is the value of it

Confidence interval . . most of the time 95% -states that we are 95% confident that the "true" difference or relationship between the groups is contained within the confidence interval range

what happens if the CI crosses 1.0

there is NO difference or it's not significant