Inferential Statistics

In most cases researchers want the _____ hypothesis to be true.

Alternative hypothesis

central limit theorem:

As sample size increases (to infinity) the plot of the means would take the shape of a normal curve

Statistical significance is easier to achieve when _____ is large; may not reflect a ______ significancw

SAMPLE SIZE; practical significance

4 factors that influence statistical power

-tradeoff between alpha and beta - Variance (power increases with reduced variance) - Sample size (larger sample greater power) - Effect size (greater difference between groups: greater power)

power= (equation)

1-beta

probability of making a type II error is denoted by:

beta

the greater the power the _____

less likely to make a type II error (like sensitivity)

degrees of freedom (df)

number of components that are free to vary

in statistics probabilities are used to determine if:

observed differences are representative to the population or if they could have occurred by chance

Hypothesis testing is used to help decide if :

observed effect reflects chance alone, or if we can argue with CONFIDENCE that these are REAL effects

assumptions from inferential statistics are based on 2 concepts of statistical reasoning:

probability & sampling error

type 1 error is:

rejecting Ho when it is TRUE (saying there is a difference when there is not)

Power analysis can be used to determin _____ or _____

sample size ; probability of type II error

Standard error of the mean depends on ______ and reflects ______

sample size; sampling error

Type II error

say there is not a difference (fail to reject Ho) when a true difference exists!

The larger the sample size the smaller the ______

standard deviation

Standard error of the mean

the standard deviation of a sampling distribution of means

Alpha is used to control _______

type 1 error

Rejecting the null hypothesis means:

unlikely that the observed findings are due to chance, more likely that a significant effect is present

Assumptions of nonparametric statistics

- not normal distribution - nominal/ordinal data

Assumptions of parametric statistics:

- random sample - group variance is homogenious (normal distribution) - interval/ratio data

T/F an alpha of 0.0001 is highly significant?

False. can only claim that it is significant not "highly significant" or "more significant"

inferential statistics allow us to _____ from a sample to a population

Generalize

generally parametric statistics have more _____

statistical power

Probabiliy=

the likelihood that any one event with occur, given ALL the possible outcomes (in the "long run")

What is Alpha?

the probability of change occurrance

Sample error

when characteristics of the sample do not reflect the characteristics of the population