Please enable JavaScript.
Coggle requires JavaScript to display documents.
USING STATISTICS TEST TO MAKE INFERENCES ABOUT POPULATIONS - Coggle Diagram
USING STATISTICS TEST TO MAKE INFERENCES ABOUT POPULATIONS
Testing for differences between groups
CHI SQUARE STATISTICS
Non parametric test. Used when analyzing nominal and ordinal data. Useful for finding differences between groups on demographic variables at the start of the study.
THE T STATISTICS
Used to determine whether there is a statistically significant difference between two groups
Correlated test
there are only two measurements taken on the same person (one group) or when the groups are related.
Independent t test
used when data values vary independently from one another.
Analysis of Variance ANOVA
Used when the level of measurement is interval or ratio and there are more than two groups or the variable of interest is measured more than two times
Testing for relationships among variables
CORRELATION AND PEARSON'S R
Correlations are evaluated in terms of magnitude, direction, and sometimes significance.
Direction:
refers to the way the two variables covary.
Magnitude:
refers to the strength of the relationship found to exist between twovariables.
MULTIPLE REGRESSION
Used to study the relationship of many independent variables on one dependent variable.
Multiple linear regression
is used when the dependent variable is measured at the interval or ratio level.
Multiple logistic regression
is used when the dependent variable is dichotomous.
Effect Size
Is the magnitude (or size) of an effect
MEAN DIFFERENCE
Examine the size of the difference in means between two groups in a study, all you have to do is subtract the means.
STANDARDIZED MEAN DIFFERENCE
Is generally calculated by taking the difference in mean outcome between the two groups (experimental minus control) and dividing it by the pooled standard deviation.
Risk Statistics
RISK RATIO OR RELATIVE RISK
Compares the probability of an out-come in two groups. One group is exposed to something (usually a risk factor), and the other group is not.
Expresses how much more (or less) likely it is for the exposed person to develop an outcome (relative to an unexposed person)
Relative risk =
Risk of outcome in exposed group/ Risk of outcome in unexposed group
RR = 1.0, there is no difference in risk between the groups:the exposure did not increase or decrease risks of an outcome.
RR > 1, it suggests an increased risk of that outcome.
RR < 1, it suggests a reduced risk in the exposed group.
ODDS RATIO
The odds of an outcome or event such as a disease. Odds ratios are used in logistic regression and in case-control studies.
Odds ratio =
Odds of an outcome in an experimental group /Odds of an outcome in control group
OR =1.0 means that the odds of an event in the groups are the same.
HAZARD RATIO
Ratio of the hazard rate of one group to the hazard ratein another group.
Hazard can be defined as the probability that an individual at any given time has an event (such as death) at that time (assume the individual was event-free up to then).
Hazard ratio =
risk of event in treated group/ risk of event in control group
HR is 1, the event rate or risk for each group is the same.
HR = 2, it means that at any time twice as many participants in the treatment/exposed groups are having an event proportionally to the comparison group.
HR < 1, then fewer participants in the treatment/exposed group are having an event proportionally to the comparison group.