Inferential statistics

One tailed t test

used if hypothesis is directional

two tailed t test

used if hypothesis is nondirectional

Calculated t value

used to determine the prob. that the difference occurred by chance -if p value is less than .05 then there is a sig. diff. between the two groups.

Post hoc comparisons( after f-ratio)

used to determine which of means differ sig.

Factorial analysis of variance

- an inferential stat. test that enables investigators to analysis the effects of several factors or indep. variables on the dep. variable.

Post hoc most freq. used after ANOVA

-Fishers LSD -Duncans new multiple test - newmans -kents -tukeys HSD -schelfers test

Statistical significance

-Involves Sub. a stats value for an observed relationship and comparing the stat. value to a distribution of other stat. values to determine how likely the value of interest could have occurred by chance.

ANOVA

-one way analysis of a variance -parametric inferential test -enables investigator to compare 2 or more group means -an extension of the t test -

Parametric tests

-require the est. of at least one variable. -measurements must be interval or ratio -Variable of concern must be normally distributed

ANCOVA ( analysis of covariance)

...Inferential stat. test That enables investigators to adjust statistically for group differences that may interfere with obtaining results that relate specifically to the effects of the independent variables on the dependent variables.

regression analysis

Often used to predict the value of one variable given info about another variable. can describe how 2 continuous variables are related. - systematic investigation of the relationship among a # of variables as possible.

Inferential statistics

Statistical tests whose goals is to determine the probability of an occurrence -Analyze data to est. the likelihood that different in the groups are the result of chance as opposed to the manipulation of variables.

Standard error( sampling error)

The discrepancy between the char. of the sample and the population. The discrepancies that occur when a small group(sample) is selected to represent the char. of a larger group(population)

Level of significance ( Alpha Level)

Whether null should be rejected depends on the level of error that can be tolerated. --- Determined to ID the prob. that the differnce between groups have occurred by chance rather than in response to the manipulation of the variables.

Multiple regression

another freq. used approach to analyzing data in which thee are multiple variables. Enables investigators to examine relationships among several variables.

Nondirectional Hypothesis

does not predict the kind of effect but can state a relationship between variable 1 and 2 -ex There will be a difference in performance of girls and boys -not defining kind of direction

Degrees of freedom(df)

mathematical concept that describes the number of events or observations that are free to vary.- for each stat. test there is a formula for calculating the df. Both t- value and df are used to find the sig level from the t distribution table. if t value is larger than tabled value using df the results are sig.

ANCOVA ( analysis of covariance)

often used to study groups that have not been randomly assigned but exist naturally

T-test(parmetric)

used when investigators wish to compare the means of 2 groups. calculated using a formula that incorporates the means for both groups and standard error of the difference between the means.

Two way ANOVA(factorial)

used with 2 indep. variables and one or more dep. variables

Directional Hypothesis

can predict the direction-effect of one variable on the other as pos. or neg.- ex- Girls performed better than boys

F-statistic/ ratio( ANOVA)

is calculated a 3 or 4 digit number that is used in conjunction with df to est. the level of sig. from the f- distribution. If f-ratio is sufficiently large then the conculsion is that there is a difference between at least 2 of the means drawn.

One way ANOVA

is used with 1 indep. and 1 dep. variable

Graph

x-axis= (horizontal) indep. variable y-axis=(vertical) dependent variable

Null Hypothesis

A statistical statement that predicts no difference between the groups of events or observations under study. Inferential stats are used in an effort to reject the null, showing that a difference does exist.

Type 2 error

Accepting the null hypothesis when it is in fact false. Example: Occurs when a researcher concludes that a given intervention did not have a positive outcome on a dependent variable when it actually did.

Level of significance ( Alpha Level)

P or alpha is usually .05 but can be .01 or .001 -How many times out of a 100 or 1000 the null might incorrectly rejected. The lower the A level the greater the confidence that the diff. between groups did not occur by chance.

Advantages of using factorial ANOVA

The test id more powerful in detecting differences and it permits the investigator to test the hypoth. relating to interaction.

covariance

The variable that could confuse the scores

Multivariate analysis

refers to a group of inferential stat. tests that enable the investigator to examine multiple variables simultaneously. Unlike other inferential stat. tests these tests permit the investigator to examine several dependent or independent variables simultaneously. -----attempt to reject the null at a given level of significance, tests include: Hutellings T, Manova, Mancova

Interaction

refers to the collective effect of the indep. variable on the dep. variable--- F ratio calculated for interaction effect along with corresponding level of sig.

Probability

refers to the likelihood that the difference between groups are the result of chance. Acceptable levels of chance must be specified in order to keep w/in the framework of inferential stats

type 1 error

rejecting the null hypothesis when it is true( no difference between groups)

correlations

relationships described in terms of probability----a relationship was significant