Brown: Chapter 6 Flashcards
Several research designs are linked to inferential relationship statistics, such as:
Nonexperimental, cross-sectional relationship studies
Predictive studies
Case-control, cross-sectional predictive studies
Cohort studies in which data is prospectively or retrospectively collected over time
What is a Pearson product-moment correlation?
Describes the strength of the relationship between 2 continuous variables
Used in nonexperimental and cross-sectional relationship studies
What is a Spearman correlation?
Describes the strength of the relationship between 2 variables when at least one variable is rank-ordered
Used in nonexperimental and cross-sectional relationship studies
What is a linear regression?
Identifies predictors of a continuous outcome measure
Used in predictive studies
Several variables (predictors) are entered into a regression equation to determine how well they predict an outcome of interest (criterion); the outcome variable is continuous
What is an odds ratio?
Identifies predictors of a dichotomous outcome measure; best when the frequency of the outcome occurring is rare
Used in case-control and cross-sectional predictive studies
Best applied to cross-sectional studies in which the event or outcome has already occurred
What is a risk ratio?
Identifies predictors of a dichotomous outcome measure
Used in cohort studies in which data is either prospectively or retrospectively collected over time
Used to determine whether an event or outcome occurs
What is a hazards ratio?
Identifies predictors of a dichotomous outcome measure; often includes survival curve estimates
Used in cohort studies in which data is either prospectively or retrospectively collected over time
Provide information about the rate of change over time
What does this symbol mean: AR
absolute risk
What does this symbol mean: β
beta weight
What does this symbol mean: CI
confidence interval
What does this symbol mean: HR
hazards ratio
What does this symbol mean: N, n
number of participants in a study
What does this symbol mean: OR
odds ratio
What does this symbol mean: p
probability value; the likelihood of making a type I error
What does this symbol mean: r value
correlation coefficient
What does this symbol mean: R2
variance; the amount of variance accounted for in a correlational study
What does this symbol mean: R2 change
the amount of unique variance when a new variable or group of variables is added to a regression equation
What does this symbol mean: RR
risk ratio or relative risk
What is a correlation?
A correlation is a type of inferential statistic
that determines how much two or more
variables fluctuate together:
- Expressed in terms of strength, direction, and
significance
- Also represent effect sizes: the r value
What is explained variance?
the proportion of one data set that accounts for the spread of numbers in a second data set
What is a scatterplot?
a graph of plotted points that shows the relationship between two sets of data
What does this scatterplot show?
positive relationship
What does this scatterplot show?
no relationship
What does this scatterplot show?
negative relationship
What does this scatterplot show?
positive relationship
What does this scatterplot show?
negative relationship
What is a correlational matrix?
In a correlation matrix, all variables listed in the vertical left column are also listed across the top horizontal column to capture all relationships that exist between any two variables.
Studies to predict outcomes use…
regression equations that calculate the degree to which two or more variables predict a specific outcome
How are linear regression outcomes reported?
with β weights, R2, and R2 change statistics.
What is β weight?
the standardized regression coefficient
What is R2?
Coefficient of determination)
The amount of variance accounted for by a single predictor or a group of predictors
Linear regression allows for…
the removal of shared variance among predictors to determine which predictors contribute the most variance
When does multicollinearity occur?
when two or more predictor variables are highly intercorrelated and means that each predictor provides only a small amount of unique variance
The challenge of linear regression lies in choosing…
predictors: the amount of total variance
accounted for shows how effective the
researcher was in the selection of predictors.
What does this picture show?
Multicollinearity, shared variance, and unique
variance
What is logistic regression?
a statistical method for analyzing a dataset in which the outcome is measured with a dichotomous variable (only two possible outcomes)
With odds ratio, logistic regression is used to…
determine the association between a predictor or a set of predictors and a dichotomous outcome that has already occurred
Confidence intervals are frequently reported with…
odds ratios
What are risk ratios (relative risk)?
Used to compare one group that has been exposed to a condition and one group that has not
A risk ratio then tells you if your absolute risk is raised or lowered based on a predictor or factor.
When are risk ratios used?
Risk ratios are appropriate for cohort studies in which the participants are followed over time, after which they are divided into two groups based on a specific outcome.
Risk ratios also can be used in retrospective analyses in which the researcher uses existing data to look back on participants to determine if they progressed or declined
over time.
What is a hazards ratio?
Hazards ratios report the likelihood of an event happening at a specific time point; however, most hazards ratios report multiple time points within a single study.
Hazards ratio often includes the presentation of data over time in the form of Kaplan Meier curves that illustrate the duration of time for two (or more) groups to reach an outcome.
What is this graph?
Kaplan-Meier curves comparing survival according to level of physical activity
