Statistics

Front 1

population

Back 1

the total number of observations sharing at least one trait in common

Front 2

parameter

Back 2

any measure obtained by having measured an entire population.
--Statisics is to sample as ___is to Population

Front 3

sample

Back 3

smaller number of observations taken fromt he total number making up the population

Front 4

statistic

Back 4

any measurement made on a sample

Front 5

outlier

Back 5

when one or two scores in a large random sample fall so far from the mean ( say 3 or 4 SD units away)

Front 6

bias

Back 6

a nonrepresentative sample it often due to

Front 7

central limit theorem

Back 7

when successive random samples are taken from a single pop., the means of these samples assume the shape of a normal curve. It applies only when all possible samples have been randomly selected from a single population

Front 8

sampling distribution

Back 8

every point on the abscissa of a sampling distribution of means

Front 9

M

Back 9

sample mean

Front 10

u

Back 10

population mean

Front 11

M - u

Back 11

sampling error

Front 12

Type 1 error

Back 12

error made when the researcher incorrectly rejects the null hypothesis ( because in fact it should have been accepted.) The alpha level sets the probability of this error occurring.

Front 13

Type 2 error

Back 13

error made when the researcher incorrectly accepts the null hypothesis ( because in fact it should have been rejected). The beta level sets the probability of this error occurring.

Front 14

z test

Back 14

method of hypothesis testing that can be used when the parameter values are normally distributed and the mean and the SD of the distribution are already known.

Front 15

One tail Test

Back 15

the use of only one tail of the distribution for testing the null hypothesis.

Front 16

Null Hypothesis

Back 16

assumption that results are simply due to chance. States that no real differences exist int he population from which the samples were drawn.

Front 17

Chi Square

Back 17

statistical test of significance used to determine whether or not frequency differences have occurred on the basis of chance. Data has to be in NOMINAL form ( actual number of cases) that fall into 2 or more discrete categories. A non parametric test.

Front 18

Beta Level

Back 18

the probability of committing a Type 2 error

Front 19

Alpha Level

Back 19

the probability of committing a Type 1 error. The level is usually set beforehand and should NOT be higher than .05

Front 20

Ha

Back 20

alternative hypothesis

Front 21

Ho

Back 21

null or chance hypothesis

Front 22

df

Back 22

degrees of freedom--based on how many values are free to vary once the mean and the sample size are set.

Front 23

point estimate

Back 23

the use of a sample for predicting a single population value. For example, the use of a sample mean for estimating "u" is a point estimate

Front 24

confidence interval

Back 24

are determined on the basis of a probability value of .95 ( 95% certainty) or .99 ( 99% certainty)

Front 25

significance

Back 25

indicates that the results of a study are not simply a matter of chance.

Front 26

effect size (ES)

Back 26

the difference between 2 population means in units of the population standard deviations. has 2 major tests for computing effect size: 1. d for t test 2. eta square for ANOVA

Front 27

alternate hypothesis

Back 27

the opposite of the null hypothesis. It states that chance has been ruled out.

Front 28

ANOVA

Back 28

Analysis of Variance- a statistical test of significance developed by Sir Ronald Fisher.
aka "F" ratio. The test is designed to establish whether a significant (nonchance) difference exists among several sample means.

Front 29

t ratio

Back 29

may be used for assessing the likelihood of a given sample being representative of a particular population. It is a test of the null hypothesis