3.1 intro to biostatisitics Flashcards by George Gutierrez

The null hypothesis is either accepted or rejected based on what?

Statistical analysis

How well did you know this?

Not at all

Perfectly

What does the null hypothesis state?

Research perspective which states there will be no true difference between groups being compared

How well did you know this?

Not at all

Perfectly

What 3 primary levels for variables?

Nominal data, Ordinal data, and intervel/ratio data

How well did you know this?

Not at all

Perfectly

What are the 3 key attributes of data measurement?

Order/Magnitude. Consistency of scale/equal distances. Rational absolute zero.

How well did you know this?

Not at all

Perfectly

What is an ex of nominal?

Labeled variables without quantitative characteristics. ex. Lawyer, truck driver.

How well did you know this?

Not at all

Perfectly

what is an ex of ordinal?

Theres order and magnitude. ex. Pain scale

How well did you know this?

Not at all

Perfectly

What is an ex of interval/ratio?

ex living siblings (#) and personal age (in years)

How well did you know this?

Not at all

Perfectly

Define mode

most frequent

How well did you know this?

Not at all

Perfectly

Define median

middle number

How well did you know this?

Not at all

Perfectly

Define mean

average number

How well did you know this?

Not at all

Perfectly

Define variance

Average of the squared differences in each individual measurement value x and the groups X-

How well did you know this?

Not at all

Perfectly

Define Standard deviation

Square root of variance value.

How well did you know this?

Not at all

Perfectly

What are parametric tests?

Stat tests useful for normally distributed data

How well did you know this?

Not at all

Perfectly

How well did you know this?

Not at all

Perfectly

Is the curve Negative or Positive skewed.

How well did you know this?

Not at all

Perfectly

Is the curve Negative or Positive skewed?

How well did you know this?

Not at all

Perfectly

If the data is skewed what cant it be?

Interval

How well did you know this?

Not at all

Perfectly

What is kurtosis?

Tells how peaked a value is

How well did you know this?

Not at all

Perfectly

If theres more cluster in a graph, what type of Kurtosis is it?

Positive

How well did you know this?

Not at all

Perfectly

If theres less cluster in a graph, what type of Kurtosis is it?

Negative

How well did you know this?

Not at all

Perfectly

What are the 4 key questions to selecting the correct statistical test?

What data level is being recorded. 2. What type of comparison/assessment is desired. 3. How many groups are being compared? 4. Is the data independent or related (paired)?

How well did you know this?

Not at all

Perfectly

What are the two sub catagory questions to ask for What Data LEVEL is being recorded?

Does the data have order/Magnititude? 2. Does the data have an equal distance?

How well did you know this?

Not at all

Perfectly

What are the 4 catagory questions to ask for What type of Comparison Assessment is desired?

Correlation. 2. Regression. 3. Surival 4. Group comparison

How well did you know this?

Not at all

Perfectly

What does Correlation (r) do?

Provides a quantitative measure of the stength and direction of a relationship between variables.

How well did you know this?

Not at all

Perfectly

What is partial correlation?

A correlation that controls for confounding variables

For Pearson Correlation, a P\>0.05 means what?

Means there is no linear correlation, there may still be non linear correlations

How can Pearson Correlation control for confounding?

By being ran as partial correlation

What is a Kappa statistic?

a correlation test showing relationship of agreement between/consistency of decisions determinations

Pertaining to Kappa statistic, what do +1, 0, -1 mean

+1 classify everyone exactly the same way. 0 theres no relationship. -1 classify everyone exactly opposite of each other

What is regression?

measure of relationship between variables by allowing the prediction about the dependent, or outcome, variable DV knowing the value/category of independent variables IVs.

What are survival tests?

Compares the proportion of events over time, or time to events, between groups. Ongoing.

What curve do survival tests get represented by?

Kaplan-Meier curve

For Nominal data, when would you use Pearson's Chi-square test?

2 groups of independent data

For Nominal data, when would you use Chi-square test of independence?

More or equal to 3 groups of independent Data

For Nominal data, when would you use Fisher's Exact test?

Groups with more or equal to 2 groups with EXPECTED cell count of less than 5

When would you use Bonferroni test of inequality?

When we need to adjust the p value for # of comparisons being made. Very conservative

What are 3 key words for RELATED data?

Pre vs Post. Before vs After. Baseline vs End.

For Ordinal data and interval data, What is the Student-Newman-Keul test?

Compares all pairwise compasrisons possible. All groups must be equal in size

For Ordinal Data and interval data, What is the Dunnett Test?

Compares all pairwise comparisons against a SINGLE CONTROL

For Ordinal Data and Interval data, what is the Dunn Test?

Compares all pairwise comparisons possible. Useful when all groups are not equal size

For Interval data, what is the ANCOVA?

Compares the means of all groups or RELATED data against a SINGLE DV

For interval Data what are two other Post-Hoc Tests for groups 3+?

Tukey, Sheffee, and Bonferroni correction

What is the Tukey test?

slightly more conservative

What is the Sheffe test?

Less affected by violations in normality and homogeneity or variances. MOST CONSERVATIVE

What does the Bonferroni correction do?

Adjusts the p value for # of comparisons being made. Very conservative

What are 3 required assumptions of interval/ratio data?

Normally-distributed. Equal variances, and Randomly derived and independent.

For Interval/ratio data what is Levene's test?

used to asses for variance that they are equal

What 3 things do we do when Interval/Ratio data is not normally distributed?

Transform data to a standardized value. Use non parametric tests. Always run Descriptive stats and graphs.

What is Type 1 error?

we reject null, but shouldve accepted. Meaning you states there is a difference when there is not.

What is a Type 2 error?

accept null but shouldve rejected. Meaning You said there isnt a difference when there actually was.

How would one minimize type 2 error?

Power

The larger the sample size the

greater ability of detecting a difference if there is one. More power

What are the 3 rules for Sample size determination?

1. Minimum dif between groups deemed significant. 2. Expected variationg of measurement. 3. Type 1 and 2 Erroo rated and confidence Interval.

How is the p value determined?

Probability of observing a test statistic value as extreme or more extreme than actually observed if groups were similar.

What do statistical tests determine?

Possible error rate or chance in comparing dif between variables

Is the p value selected before or after a study starts?

BEFORE

what value should P be to be statistical significant?

0.05 or less.

What are 4 interpretations of a pre set p value?

Prob of making a type 1 error if null is rejected. Probability of claiming a dif between groups when there isnt. Prob of obtaining group differences as great or greater if the groups were actually the same/equal. Prob of obtaining a test statistic as higher if the groups were actually the same/equal.

How are Confidence Interval CI calculated?

Calculated at an a priori % of confidence that statistically the real dif or relationship resides.

What two things is CI based on?

Variation in sample (V/SD). Sample size (N)

What is the interpretation of a 95% CI point estimate?

95% confident that the true difference or relationship between teh groups is contained within the confidence interval range

What is the interpretation of a 95% CI without a p value?

If CI crosses 1 for ratios (OR/RR/HR) or 0.0 for absolute differences = NOT significant.