Study your flashcards anywhere!

Download the official Cram app for free >

  • Shuffle
    Toggle On
    Toggle Off
  • Alphabetize
    Toggle On
    Toggle Off
  • Front First
    Toggle On
    Toggle Off
  • Both Sides
    Toggle On
    Toggle Off
  • Read
    Toggle On
    Toggle Off

How to study your flashcards.

Right/Left arrow keys: Navigate between flashcards.right arrow keyleft arrow key

Up/Down arrow keys: Flip the card between the front and back.down keyup key

H key: Show hint (3rd side).h key

A key: Read text to speech.a key


Play button


Play button




Click to flip

43 Cards in this Set

  • Front
  • Back
Parametric Statistical Tests
Used for dependent variables measured on an interval or ratio scale

Dependent variable data must be continuous and normally distributed

Student's t-test
Paired t-test
Repeated measures Analysis of Variance
One-Tailed Test
A research hypothesis that one group will be higher/lower than another (directional)
Two-Tailed Test
A research hypothesis that one group will be different than another (non-directional)
What is the advantage of a Two-tailed test?
They are easier to reject the null hypothesis
What are the disadvantages of a two-tailed test?
Can lead to Type III error
What is Type III error?
There is a statistically significant difference between the gps, but the direction of the difference stated by the researchers based on their sample data is incorrect
What is an Independent Group?
Different subjects in each gp being compared (groups are mutually exclusive)

Comparisons are made between groups

Examples: Parallel designs, M vs FM, MSII vs. MSIII
What is a dependent group?
The subjects in each group being compared are the same.

Comparisons are made within a group or cohort at different points in time.

Examples: Pretest-Posttest, Before-After, Crossover designs
What is a student's t-test?
Test used to compare 2 independent groups for differences on an interval or ratio level dependent variable

Commonly used to compare characteristics of 2 tx groups at beginning of a study (to see if gps were equal at baseline) or outcomes at end of study (determine the effect of a tx)
What is a paired t-test?
Same as a student's t-test, but the two groups are dependent

Often used in before/after, pretest/post-test, and cross-over studies
What is Analysis of variance (ANOVA)?
Test used to compare 3 or more independent groups for differences on an interval or ratio level dependent variable

Used to compare treatment gps at both the beginning of a study (to test for equality) and at the end of a study (to test for effectiveness)
Why is it better to use ANOVA than t-tests for large studies?
ANOVA=1 possibility
t-tests=need to use multiple tests leading to many possiblities
What are the types of ANOVA tests?
One-way (1 IV, 1 DV)

Two-way (2 IV, 1 DV)

MANOVA (>1 IV, > 1 DV)
In ANOVA post-hot tests, p < alpha tells you what?
Does not tell you which groups are different from each other.
What is Repeated measures ANOVA?
Similar to ANOVA, but for 3 or more dependent groups

A-baseline vs. A-time1 vs A-time2 vs...
What is analysis of Covariance (ANCOVA)?
Tests for differences between gps in interval or ratio data, but also controls for differences between the gps in terms of covariates/confounding variables

May be used regardless of number of gps compared, or the relationship between the groups (independent, dependent)
What are Non-Parametric statistical tests?
Statistical tests used when dependent variables are measured on a nominal or ordinal scale

Also used when data from interval or ratio scale dependent variables are not normally distributed

Examples: Chi-square, Friedman
What is a Chi-square test?
Measures differences in proportions (percentages) between groups

Differences are between independent groups on a nominal level data

Commonly used to analyze contingency tables which can be used to determine equality of gps at baseline or differences between gps after treatment

Sample size should always be atleast an expected frequency of 5 subjects per data cell
What if there are less than five subjects per cell?
Fisher's Exact Test
What is McNemar's Test?
Similar to Chi-square, but for paired or matched data (dependent)
What is the Mantel-Haenszel Test?
Similar to Chi-square, but for stratified data

One contigency table for each level of stratification
What is the Mann-Whitney U Test?
Analogous to the student's t-test, but dependent variable is ordinal or not normally distributed
What is the Wilcoxon Signed-Rank?
Analogous to paired t-test, but dependent variable is ordinal or not normally distributed
What is the Kruskall-Wallis test?
Analogous to ANOVA, but dependent variable is ordinal or not normally distributed
What is the Friedman test?
Analogous to repeated measures ANOVA, but dependent variable is ordinal or not normally distributed
What is the Kaplan-Meier Survival Analysis?
A non-parametric test used to examine differences in the probability of an event occurring between treatment groups over time
What is the log-rank test?
Used to determine if the differences in probabilities between the groups are different over the entire time of the study
What is Pearson's r (aka Correlation Coefficient)?
Measures degree of relationship between two continuous variables

Data must be interval or ratio

Values for r range from -1 to +1

-1=perfect negative relationship
0=no relationship
+1=perfect positive relationship

Both p value and the test statistic (r) are needed to interpret result. Look at p value first. If p < alpha, then interpret the r value
What is Spearman's rho?
Similar to Pearson's r, but with ordinal (ranked) data
What is regression analysis?
Derives an equation that allows researchers to predict a value for a dependent variable given one or more independent variables

Dependent variable usually measured on an interval or ratio scale.
What if the dependent variable is ordinal or nominal in regression analysis?
Use logistic regreesion
What is the regression equation?
What is regression weights (m)
How much each individual IV(x) contributes to the regression equation (unstandardized)
What is Beta (-1 - +1 scale)?
Standardized regression wt. allows reader to assess the strength of each IV(x) in predicting the DV (y)
What is r^2 (0-1 scale)?
Total amount of variance (%) explained by the combo of all IVs included in the analysis (0-1)
What is the Cox Proportional Hazard Model?
Uses regression analysis to predict for each subject the time to the study end point (e.g. death)

While controlling for time and confounding variables, the Cox regression equation slop coefficients (m) can be translated into relative risk values for the end point for each level of the independent variable.
What is Relative Risk (aka hazard ratio)?
Measure of the relative frequency of an event when a specific factor is present
What is the relative risk equation?
Incidence rate of event among those exposed to a specific factor / "" those not exposed to the factor (control)
How is incidence calculated?
# who experience event / total # at risk
What sudy design should Relative risk only be used in?
Prospective study designs
How is the relative risk measured?
RR=1: variables are independent (no relationship)

RR<1: those exposed have lower incidence of event than those not exposed

RR>1: those exposed have higher incidence of event than those not exposed
How are confidence intervals used with relative risk?
Determination of statistical significance (instead of p values)

Always look at the CI before interpreting the RR
What type of studies are Odds ratios used for?
Retrospective study