Research Test 2 Part 2

Front 1

internal validity

Back 1

threat to one group, not the whole experiment

Front 2

alternative explanations of study outcomes are called

Back 2

threats to internal validity

Front 3

4 parts of good quantitative research design

Back 3

statistical conclusion validity, internal validity, external validity, construct validity

Front 4

statistical conclusion validity

Back 4

ability to detect true relationships statistically

Front 5

internal validity

Back 5

extent it can be inferred the independent variable caused dependent variable

Front 6

external validity

Back 6

generalizability of observed relationships across samples, settings, or time

Front 7

construct validity

Back 7

degree that key constructs are adequately captured

Front 8

main thing in controlling confounding variables

Back 8

achieving constancy of conditions

Front 9

external factors

Back 9

conditions, environment, time, etc

Front 10

intrinsic factors

Back 10

subject characteristics

Front 11

5 methods of controlling intrinsic factors

Back 11

randomization, subject as own control, homogeneity, matching, statistical control

Front 12

main 2 methods for controlling intrinsic

Back 12

matching, analysis of covariance

Front 13

statistical control ie

Back 13

analysis of covariance

Front 14

randomization (2)

Back 14

recruitment, assignment

Front 15

researcher bias (2)

Back 15

double blind, more than one observer

Front 16

statistical analysis (4)

Back 16

test null HO, probability of type 1 or 2 error, effect size, statistic assumptions have been met

Front 17

type 1 error (3)

Back 17

think intervention worked when it did not, rejection of true null HO, caused by extraneous variable

Front 18

alpha or p value cannot be greater than

Back 18

0.05

Front 19

type 2 error (3)

Back 19

accept false null HO, failure to find significance when there is one, from insufficient power (sample size)

Front 20

which error is more serious

Back 20

type 1

Front 21

when we increase control of type 1 we

Back 21

decrease control of type 2

Front 22

7 threats to internal validity

Back 22

temporal ambiguity, selection, history, maturation, attrition (mortality), testing, instrumentation

Front 23

temporal ambiguity

Back 23

unclear which variable occurred first

Front 24

selection

Back 24

difference in experimental and control groups

Front 25

biggest threat to non experimental designs

Back 25

selection

Front 26

history

Back 26

events during data collection (Katrina, 9/11)

Front 27

maturation

Back 27

subjects change (smarter, sicker)

Front 28

way to help with maturation

Back 28

ANCOVA

Front 29

attrition (mortality)

Back 29

loss of subjects from study

Front 30

testing

Back 30

subjects get better at test (pre/post)

Front 31

instrumentation

Back 31

data collection instrument changes during data collection

Front 32

3 ways to control instrumentation

Back 32

inter-rater reliability, training to correct, calibration of instruments

Front 33

3 tx effects

Back 33

placebo, hawthorne, multiple tx

Front 34

hawthorne

Back 34

subject want to please researcher or behave differently because being studied

Front 35

hawthorne most concern when

Back 35

researcher in position of authority, leadership role (subject's physician)

Front 36

4 other concerns not controlled through random assignment

Back 36

diffusion of tx, resentful controls, equalization of tx, compensatory rivalry "john henry"

Front 37

diffusion of tx

Back 37

subjects communicating

Front 38

resentful controls

Back 38

control gets no tx (may drop out or score low), blinded tx (give fake tx)

Front 39

equalization of tx

Back 39

people admin tx to compensate the on tx group, double blind

Front 40

compensatory rivalry "john henry"

Back 40

control tries to compensate for lack of tx, blinding, separate groups

Front 41

3 parts to external validity

Back 41

generalizability, population validity, ecological validity

Front 42

population validity

Back 42

ability to generalize to other populations (children to adolescents)

Front 43

ecological validity

Back 43

setting, geographical location

Front 44

6 threats to external validity

Back 44

selection effects, time, history, novelty, experimenter effects, hawthorne effect

Front 45

selection effects

Back 45

sample should represent population

Front 46

time

Back 46

winter vs summer, morning vs night

Front 47

history

Back 47

results considered in context of time period, may not be generalizable to future

Front 48

novelty

Back 48

subject response just because it is new or unusual

Front 49

experimenter effects

Back 49

are there different effects if different researcher

Front 50

hawthorne effect

Back 50

internal or external

Front 51

increase external validity by

Back 51

replication replication replication

Front 52

3 ways to deal with threats

Back 52

eliminate threat, control threat, account for threat

Front 53

eliminate threat

Back 53

data collection by assistant if researcher is threat

Front 54

control threat

Back 54

make certain some subjects with threat are in each group

Front 55

account for threat

Back 55

"limitations"

Front 56

must have internal validity in order to have

Back 56

external validity

Front 57

impossible to develop "best practice" without

Back 57

organizing and evaluating research evidence through a systematic review

Front 58

3 forms of systematic reviews

Back 58

narrative qualitative integration, meta-analysis, metasynthesis

Front 59

meta-analysis

Back 59

statistical integration of results

Front 60

metasynthesis

Back 60

theoretical integration and interpretation of qualitative findings

Front 61

meta-analysis advantages

Back 61

objectivity, increased power, increased precision

Front 62

increased power

Back 62

reduces risk of type 2 error

Front 63

increased precision

Back 63

smaller confidence intervals

Front 64

objectivity

Back 64

eliminates bias in drawing conclusions

Front 65

meta-analysis criteria

Back 65

hypothesis identical across studies

Front 66

metasynthesis

Back 66

bringing together and breaking down findings

Front 67

3 general rules for quantitative analysis

Back 67

statistical tests are selected a priori, run all identified tests, report all tests run

Front 68

selective reporting is a source of

Back 68

bias

Front 69

type of analysis driven by (3)

Back 69

goals, assumptions of data, number of variables

Front 70

frequency distributions

Back 70

arrangement of numeric values from lowest to highest and how many times each value is obtained

Front 71

frequency distributions in 3 ways

Back 71

shape, central tendency, variability

Front 72

frequency distributions represented in (3)

Back 72

chart, table, graph

Front 73

3 characteristics of normal distribution

Back 73

symmetric, unimodal, not too peaked and not too flat

Front 74

normal distribution referred to as

Back 74

bell shaped curve

Front 75

mode

Back 75

most frequently occurring score

Front 76

median

Back 76

exact middle

Front 77

mean

Back 77

average

Front 78

only thing applicable to nominal value

Back 78

mode

Front 79

median used when

Back 79

want to compare

Front 80

what is less affected by extreme scores

Back 80

median

Front 81

what is affected by outliers or extremes

Back 81

mean

Front 82

homogeneity variability

Back 82

little

Front 83

hetergeneity variability

Back 83

great

Front 84

range

Back 84

highest value minus lowest value

Front 85

standard deviation (SD)

Back 85

average deviation of scores in a distribution

Front 86

correlation coefficients used for

Back 86

describing relationships between two variables

Front 87

the greater the absolute values of the coefficient...

Back 87

the stronger the relationship (r = -.45 > r = +.40)

Front 88

pearsons r is both

Back 88

descriptive and inferential statistic

Front 89

correlation tests that the relationship between two variables is

Back 89

not zero

Front 90

reliability by setting

Back 90

confidence limits

Front 91

probability by

Back 91

hypothesis testing

Front 92

statistical inference conclusions concerned with

Back 92

probability of drawing an erroneous conclusion

Front 93

confidence interval

Back 93

2 numbers defining an interval that we believe with identified level of confidence actually includes the estimated population paramete

Front 94

interval estimation

Back 94

range of values within a population value probably lies

Front 95

interval estimation includes

Back 95

confidence interval (CI) and confidence limits

Front 96

3 things significance testing can go between

Back 96

sample and population, 2 samples, 2 variables in sample

Front 97

hypothesis testing used to test differences between (3)

Back 97

means, proportions, variances

Front 98

level of significance

Back 98

preset standard that is considered significant in determining differences

Front 99

most common level of significance

Back 99

.05 (5% chance of making type 1 error) and .01

Front 100

for level of significance set a

Back 100

priori (alpha) (differen than chrombox alpha)

Front 101

priori alpha

Back 101

p value

Front 102

smaller p value show more evidence

Back 102

against null HO, significant

Front 103

larger p value show more evidence

Back 103

for null HO, not significant

Front 104

2 ways to thing about which stat is correct

Back 104

level of measurement, number of variables

Front 105

level of measurement

Back 105

nominal, ordinal, interval ratio, determines parametric or nonparametric

Front 106

number of variables

Back 106

bi-variant, multivariant

Front 107

parametric stats type of sample

Back 107

random from define population

Front 108

parametric stats dependent variable measured at

Back 108

interval ratio level

Front 109

parametric stats dependent variable is

Back 109

normally distributed

Front 110

parametric stats is

Back 110

more powerful

Front 111

parametric stats can make

Back 111

assumptions about population from which sample was drawn

Front 112

nonparametric stats do not require

Back 112

normal distribution

Front 113

nonparametric stats measures data

Back 113

on nominal or ordinal level

Front 114

t test tests the difference between

Back 114

two means (means for men vs women)

Front 115

t test ie

Back 115

is there a difference between effectiveness of tylenol and advil for HA? mean pain rating for each group

Front 116

t test for dependent (paired) groups: within-subjects test

Back 116

means for pt before and after surgery

Front 117

error of multiple comparisons

Back 117

each test 5% chance results due to standard error, 3 groups = 15% chance

Front 118

what avoids error of multiple comparisons

Back 118

ANOVA

Front 119

analysis of variance (ANOVA)

Back 119

tests the difference between more than 2 means

Front 120

5 ways of ANOVA variance analysis

Back 120

one way ANOVA (3 groups), multifactor (two way) ANOVA, repeated measures ANOVA (RM-ANOVA) within subjects, comparing, F statistic

Front 121

comparing (ANOVA)

Back 121

between group variability, within group variability

Front 122

with ANOVA results tells researcher that mean scores between at leats two groups are

Back 122

differnet but does not specify which groups

Front 123

with ANOVA you can reject null but further testing is needed to determine

Back 123

where differences lie

Front 124

post hoc tests

Back 124

turkeys HSD - which groups differ from one another

Front 125

chi squared test is nonpara or parametric

Back 125

nonparametric

Front 126

chi squared test the difference in

Back 126

proportions in categories within a contingency table

Front 127

chi squared compares observed

Back 127

frequencies in each cell with expected frequencies (expected if no relationship)

Front 128

chi squared test 2 data comparisons

Back 128

nominal or categorical

Front 129

chi squared tes stat

Back 129

X2 (squared) statistic

Front 130

multivariate statistics

Back 130

statistical procedures for analyzing relationships among 3 or more variables

Front 131

two common multivariate stats in nursing research

Back 131

multiple regression, analysis of covariance (ANCOVA)

Front 132

multiple linear regression used to

Back 132

predict

Front 133

multiple linear regression dependent variable is

Back 133

continuous

Front 134

predictor variables are (2)

Back 134

continuous or dichotomous

Front 135

correlation index for dependent and predictor variables is

Back 135

R

Front 136

regression

Back 136

strong statistical analysis

Front 137

regression is _____ not ______

Back 137

predictive not causative

Front 138

logistic regression analyzes

Back 138

relationships between nominal level dependent variable and 2 or more independent variables

Front 139

logistic regression yields an

Back 139

odds ratio

Front 140

analysis of covariance (ANCOVA) extends ANOVA by

Back 140

removing effect of confounding variables (covariates) before testing whether mean group differences are significant

Front 141

multivariate analysis of covariance

Back 141

MANCOVA

Front 142

MANCOVA

Back 142

extension of ANOVA to more than one dependent variable

Front 143

factor analysis used to

Back 143

reduce large set of variable into smaller underlying dimensions

Front 144

factor analysis used primarily in

Back 144

developing scales and complex instruments

Front 145

causal modeling tests a

Back 145

hypothesized multivariable causal explanation of a phenomenon

Front 146

causal modeling includes (2)

Back 146

path analysis, structural equations modeling

Front 147

SPSS

Back 147

most often used in nursing research, stats for bio sciences

Front 148

SAS

Back 148

more purely mathematical

Front 149

most common reported stats (4)

Back 149

descriptive stats about sample and variables, analysis of group equivalency, stats about role of error, stats to evaluate magnitude of effect, stats to determine confidence level