Dea Ii Exam Ii

Front 1

what research design is considered the classic model for clinical practice?

Back 1

one group pre-test/post-test

Front 2

what are three basic criteria for a true experiment (also known as a randomized controlled trial)?

Back 2

(1) randomization
(2) minimum of two groups
(3) treatment that is controlled by the investigator

Front 3

this term can indicate independent groups of subjects that are compared to each other

Back 3

between

Front 4

this term can indicate repeated measures, aka a pretest and a postest on subjects

Back 4

within

Front 5

this effect asks, "does the effect of treatment vary based on whether you had a pretest or not?"

Back 5

pretest by treatment interaction effect

Front 6

this type of statistics is used to "infer" something about what you'd see in the population

Back 6

inferential

Front 7

do parameters describe samples or populations?

Back 7

populations

Front 8

do statistics describe samples or populations?

Back 8

samples

Front 9

a single numerical index that describes the members of a sample

Back 9

statistic

Front 10

a single numerical index that describes the members of a population

Back 10

parameter

Front 11

an idea (declarative statement) about the assumed relationship between two or more populations with regard to a particular dependent variable

Back 11

hypothesis

Front 12

together, these two types of statistical hypotheses express all possible relationships in a research study

Back 12

(1) research
(2) null

Front 13

this hypothesis assumes that the means of each group are not different from each other, and that the effect of the independent variable was the same on both groups

Back 13

null

Front 14

which hypothesis is tested during hypothesis testing?

Back 14

null

Front 15

this type of hypothesis states that a change in the independent variable will cause a change in the dependent variable, but the change could either be positive or negative (just that the treatment group will differ from the nontreatment group); also known as two tailed

Back 15

non-directional

Front 16

this type of hypothesis states that a chance in the independent variable will cause a certain type of change (either positive or negative, based on your prediction) in the dependent variable (treatment group will be better than the non-treatment group); also known as one tailed

Back 16

directional

Front 17

criterion level of acceptability (how much of a chance you're willing ot take in being wrong about your decision about the null)

Back 17

alpha level

Front 18

what type of error does setting an alpha level protect against?

Back 18

type I

Front 19

what is the alpha level usually set at?

Back 19

.05

Front 20

if the statistical value is incompatible with the null, do you accept or reject the null hypothesis?

Back 20

reject

Front 21

if the statistical value is compatible with the null, do you accept or reject the null hypothesis?

Back 21

accept

Front 22

if you reject the null hypothesis, it is less likely that the difference in means occurred by __

Back 22

chance

Front 23

the alpha level determines the start of the __ __, the place where we have to reach in order to reject the null hypothesis

Back 23

rejection region

Front 24

this type of error occurs when the null hypothesis is false, but the researchers do not reject, and instead accept the null hypothesis; also known as missed findings

Back 24

type II error

Front 25

this type of error occurs when the null hypothesis is true, but the researchers do not accept, and instead reject the null hypothesis; also known as false findings

Back 25

type I error

Front 26

this level is the probability of having a type II error

Back 26

beta

Front 27

this is the probability that the null is false and that you reject it; represented as 1-B; if you have enough of this, you can minimize type II errors

Back 27

statistical power

Front 28

what are five factors that affect statistical power?

Back 28

(1) alpha level
(2) difference size between the two groups
(3) variability among subjects within groups
(4) sample size
(5) number of tails of hypothesis

Front 29

does a low alpha level increase or decrease statistical power?

Back 29

decrease (more likely to have a type II error)

Front 30

does more variability in subjects increase or decrease statistical power?

Back 30

decrease

Front 31

what is the easiest way to increase statistical power?

Back 31

to increase sample size

Front 32

which type of hypothesis has more statistical power, 1 tailed or 2 tailed?

Back 32

1 tailed

Front 33

if p is less than the set alpha level, do you accept or reject the null?

Back 33

reject

Front 34

if p is greater than or equal to the set alpha level, do you accept or reject the null?

Back 34

accept

Front 35

what are 7 guiding questions to determine if the correct statistic was used?

Back 35

(1) what question did the investigators ask?
(2) what was the data type of the dependent variable
(3) how many groups were icluded in the study?
(4) how many IVs (factors and levels)?
(5) were measures repeated or independent?
(6) is the hypothesis 1 or 2 tailed?
(7) what was the distribution of the data (normal or not?)

Front 36

this is the ratio of between groups variability over within-group variability

Back 36

t ratio

Front 37

does high variability increase or decrease the t ratio?

Back 37

decrease

Front 38

do you want a bigger or smaller t ratio in order to maximize the different between the means relative to variance?

Back 38

bigger

Front 39

when there is 1 factor with 2 or more levels, and it is a between groups design, what statistical test would you use?

Back 39

one-way chi-square

Front 40

in a one-way chi square test, if the obtained value of chi square is greater than the critical value, do you accept or reject the null?

Back 40

reject

Front 41

for nominal data-if there are two factors and two or more levels for each factor, and the study is a between groups design, what statistical test do you use?

Back 41

two way chi square

Front 42

what are the degrees of freedom for a one-way chi square test?

Back 42

(a-1)

Front 43

what are the degrees of freedom for a two way chi square test?

Back 43

(a-1) x (b-1)

Front 44

for a two way chi square test, if the obtained value of chi square is greater than or euqal to the critical value, do you accept or reject the null?

Back 44

reject

Front 45

what type of data are the chi square tests used for?

Back 45

nominal

Front 46

for ordinal data-if there is one factor and two levels, and it is a between groups design, what type of statistical test do you use?

Back 46

Mann-Whitney

Front 47

for oridinal data-if there is one factor and two levels, and it is a within subjects design, what type of statistical test do you use?

Back 47

WIlcoxon

Front 48

for ordinal data-if there is 1 factor and more than two levels, and it is a between groups design, what type of statistical test do you use?

Back 48

Kruskal-Wallis

Front 49

for ordinal data-if there is one factor and more than two levels, and it is a within subjects design, what type of statistical test do you use?

Back 49

Friedman

Front 50

what are the four types of statistical tests used for ordinal data?

Back 50

(1) Mann-Whitney
(2) Wilcoxon
(3) Kruskal-Wallis
(4) Friedman

Front 51

for score data-if there is 1 factor and two levels, and it is a between groups design, what type of statistical test do you use?

Back 51

independent groups t-test

Front 52

for score data-if there is one factor and two or more levels, and it is a between groups design, what type of statistical test do you use?

Back 52

one way between groups ANOVA

Front 53

for score data-if there is one factor and two levels, and the design is within subjects, what statistical test would you use?

Back 53

paired t-test

Front 54

what are the degrees of freedom for a paired t-test?

Back 54

n-1 (n=number of pairs of scores)

Front 55

for score data-if there is one factor and two or more levels, and the design is within subjects, what statistical test would you use?

Back 55

one way within subjects ANOVA

Front 56

for score data-if there are two factors and two or more levels for both factors, and it is a between groups design, what statistical test would you use?

Back 56

two way between groups ANOVA

Front 57

for score data-if there are two factors and two or more levels for each factor, and the design is within subjects, what type of statistical test would you use?

Back 57

two way within subjects ANOVA

Front 58

for score data-if there are two factors and two or more levels for each factor, and the design is mixed (both within subjects and between groups), what statistical test would you use?

Back 58

two way mixed ANOVA