interpretation of the posterior distribution and credible intervals is more

intuitive than the interpretation of p values and confidence intervals

a Bayesian approach allows for the explicit inclusion of prior knowledge and

expertise in the specification of the prior distribution

the Bayesian approach does not assume underlying normal distributions, so it can robustly handle skewed

or otherwise nonnormal distributions that may cause problems for classical statistical tests

A primary concern with Bayesian methods is with the specification of the prior distribution

Be very explicit in reporting the exact form of your prior distributions and your reasons for using those priors

The computations involved with Bayesian data analysis are considerably more complex than

those associated with, say, classic t tests or ANOVAs

Prior to gathering your data, understand the type of data you will collect

(e.g., means, counts) and what statistical methods are appropriate

As the first part of your analysis,

always compute descriptive statistics and create visual displays of your data.

Check the assumptions of your statistical tests and, in particular,

identify values that might be considered outliers

To examine patterns of means, you will typically use

t tests, one-way ANOVA, or factorial ANOVA

To examine the relationships between variables measured on interval or ratio scales,

you will most often use correlations or regression analysis

When inspecting counts or frequencies for categorical data,

you will typically use a chi-squared test

Once you have completed your data analysis,

create visual displays that capture any of your key results.

Help

Options

focusNode

Didn't know it?
click below

Knew it?
click below

Don't Know

Remaining cards (0)

Know

retry

shuffle

restart

0:00

Embed Code - If you would like this activity on your web page, copy the script below and paste it into your web page.

Normal Size Small Size show me how

psy400ch14p367-369

Advantages and Disadvantages of Bayesian Approaches

Term	Definition
interpretation of the posterior distribution and credible intervals is more	intuitive than the interpretation of p values and confidence intervals
a Bayesian approach allows for the explicit inclusion of prior knowledge and	expertise in the specification of the prior distribution
the Bayesian approach does not assume underlying normal distributions, so it can robustly handle skewed	or otherwise nonnormal distributions that may cause problems for classical statistical tests
A primary concern with Bayesian methods is with the specification of the prior distribution	Be very explicit in reporting the exact form of your prior distributions and your reasons for using those priors
The computations involved with Bayesian data analysis are considerably more complex than	those associated with, say, classic t tests or ANOVAs
Prior to gathering your data, understand the type of data you will collect	(e.g., means, counts) and what statistical methods are appropriate
As the first part of your analysis,	always compute descriptive statistics and create visual displays of your data.
Check the assumptions of your statistical tests and, in particular,	identify values that might be considered outliers
Compute effect sizes and, when appropriate,	confidence intervals.
If appropriate, consider using nonparametric tests or	robust parametric tests.
To examine patterns of means, you will typically use	t tests, one-way ANOVA, or factorial ANOVA
To examine the relationships between variables measured on interval or ratio scales,	you will most often use correlations or regression analysis
When inspecting counts or frequencies for categorical data,	you will typically use a chi-squared test
the Mann Whitney U test is commonly used for	examining ordinal data.
Once you have completed your data analysis,	create visual displays that capture any of your key results.

Created by: james22222222

"Know" box contains:
Time elapsed:
Retries: