BIOSTATS: PARAMETRIC TESTS Flashcards
The Pearson product moment correlation must be used on relationships.
Linear
In order to visualize data points, correlation is commonly graphed as a .
Scattergram
In linear regression, more than one independent variable is used to predict the dependent (Y) variable.
Multiple
If data is not linear, the form of regression analysis used is called .
Simple nonlinear regression
Logistic regression can be binomial, ordinal, or .
Multinomal
In order to determine how well the “best-fit” line on a sample represents the population, a interval can be calculated for the slope range.
Confidence
The only way to reduce the chance of both type I and type II errors is to increase the .
Sample size
The (larger/smaller) the sample size, the closer the correlation coefficient is to the true population relationship.
Larger
If there is no correlation between two groups, the correlation coefficient is (value) .
Zero
A type (I/II) error occurs when a false positive result is obtained.
Type 1
A low alpha value (increases/decreases) the chance of committing a type I error.
Decreases
is a method that fits the best straight line through a scattergram of linear-related data points.
Simple linear regression
In evaluating a binary dependent variable, the likelihood of an event can be predicted using __________ regression.
Logistic
Logistic regression predicts the ____________ of particular outcomes.
Probability
The Trauma Injury Severity Score, used to predict the mortality of injured patients, was originally developed using regression.
Logistic
The correlation coefficient used on ordinal data is called the correlation.
Spearman rank-order
The chance of committing a type II error is equal to the value.
Beta
ANOVA measures effect size through a percentage called , which is analogous to the coefficient of determination.
ETA-squared
The correlation coefficient used on interval or ratio data is called the correlation.
Pearson product moment
The coefficient of is a value between zero and one that represents the fraction of the variation between variables, rather than the natural variation within each variable.
Determination
To increase the certainty that a (positive/negative) result is correct, the ideal power must be increased.
Negative
In logistic regression, the dependent variable is .
Categorical
If a placebo appears to have an effect when represented graphically, it is called .
Regression to the mean
A type (I/II) error occurs when a false negative result is obtained.
Type II
Logistic regression is used for predicting (independent/dependent) variables that are members of one of a limited number of categories.
Dependent
The power of a test is increased when the effect size is (increased/decreased) .
Increased
_______________ logistic regression is used to analyze a dependent variable that has more than two outcome categories.
Multinominal
In regression, correlations among nonlinear data with more than one independent (X) variable are determined.
Multiple nonlinear
A low alpha value (increases/decreases) the chance of committing a type II error.
Increases
When interpreting study results, the measure of the risk that a researcher is willing to take in rejecting a null hypothesis that is actually true is called .
Alpha
A perfect positive correlation has a correlation coefficient of (value) __________________.
One
In regression analysis, the line of best-fit cannot extend beyond the .
Data points
A coefficient is a number between negative one and positive one that describes the degree of correlation between one variable and another.
Correlation
, or the likelihood that a result is not a false negative, is equal to one minus the beta value.
Power