Research And Program Evaluation

Front 1

the process of gathering data in order to make evaluative comparisons regarding different situations

Back 1

experimental research

Front 2

the researcher uses preexisting groups and the independent variable cannot be altered

no random selection

Back 2

quasi-experiment

Front 3

"after the fact", connotating a correlational study of research in which intact preexisting groups are used

Back 3

ex post facto (after fact)

Front 4

refers to whether the Dependent variables (DVs) were truly influenced by the experimental Independent variables (IVs) or whether the factors had an impact

Back 4

internal validity

Front 5

whether the experimental research results can be generalized to larger populations.

Back 5

external validity

Front 6

interpreting the results in the simplest way, providing the easiest and less-complex explanations

Back 6

parsimony (also Occam's Razor)

Front 7

occurs when an undesirable variable which is not controlled by the researcher is introduced in the experiment. May also be called contaminating variable

Back 7

Confounding (contaminate)

Front 8

research that is conducted to advance our understanding of theory

Back 8

basic research

Front 9

research that is conducted to advance our knowledge of how theories, skills and techniques can be used in terms of practical application

Back 9

applied research (action research or experience near research)

Front 10

The variable that the researcher manipulates, controls or wishes to experiment with.

Back 10

Independent variable (controls), (I manipulate or experiment with the IV)

Front 11

The variable that expresses the outcome or the data

Back 11

Dependent variable (data)

Front 12

The group that does not receive the independent variable or the group that will not have the experimental treatment applied to them.

Back 12

Control group(no manipulation)

Front 13

The group that receives the independent variable or the experimental treatment.

Back 13

The experimental group(manipulation)

Front 14

How many subjects are is needed for a correlation Research per variable

Back 14

30 subjects

Front 15

How many subjects are need for a survey

Back 15

At least 100 subjects

Front 16

An educated guess which can be tested using the experimental model (using the IV and DV). It was pioneered by R.A. Fisher

Back 16

Hypothesis

Front 17

This hypothesis asserts that the samples will not change even after the experimental variable is applied. The control group and experimental group will not differ at the end of the experiment. (The IV does not affect the DV).

Back 17

null hypothesis

It is when the researcher rejects or attempts to disprove or "nullify" a hypothesis.

A researcher cannot reject a hypothesis without replacing it with an alternative hypothesis. The alternative hypothesis is what the researcher really believes might be going on.

Front 18

A test used to determine whether a difference in the group's scores is significant or just due to chance factors

Back 18

Test of significance

Front 19

A value drawn from a sample

Back 19

Statistic (s-sample)

Front 20

A value obtained from a population

Back 20

Parameter (p-populations)

Front 21

the accepted probability level is usually

Back 21

.05 or less

Front 22

This means that there is only a 5% chance that the difference between the control group and the experimental group is due to chance factors

Back 22

P=.05

Front 23

This error occurs when the null hypothesis (no change) is true but the researcher rejects it;if we incorrectly think we have significant evidence-strong enough to reject the null-we conclude that there is actually a change or difference in groups. as a result we incorrectly reject the null.

Back 23

Type 1 or alpha error

Hint 23

wearing your lucky socks have no effect on the outcomes of your basketball games, then our conclusion that they do is wrong. we concluded that there is a difference but there really isn't.

Front 24

type 2 errors happen when the null hypothesis (no change) is false and you subsequently fail to reject it (accept it when it is actually false);

Back 24

Type 2 or Beta error

Front 25

simplest form of analysis of variance; This test is used to ascertain whether 2 Sample means are significantly different

Back 25

T test

Front 26

One way analysis of variance is used for testing one IV; two way analysis of variance test 2 IV's

Back 26

statistical test-Analysis of variance-anova

Front 27

This statistical test will test 2 or more groups while controlling for extrenuous variables that are called covariates.

Back 27

statistical tests-Analysis of Covariance- ancova

Front 28

If a researcher uses 2 independent variables then the testing of choice is

Back 28

Multi variant analysis of variance

Front 29

To see if significant differences exist in an ANOVA you will use what table

Back 29

An F table

Front 30

A statistic that indicates the degree or magnitude of relationship between 2 variables (linear relationship) is known as a

Back 30

Correlation coefficient - it is signified by using a lower case R; how change in one variable relates to the change in another

Front 31

This correlation is evident when both variables change and the same direction

Back 31

Positive correlation

Front 32

This correlation is evident when the variables or inversely associated or one goes up and the other goes down

Back 32

Negative correlation

Front 33

Correlation ranges from what numbers

Back 33

0.00,1.0 and -1.0- 0.00 means no relation and 1.0 or -1.0 signifies perfect relation.

Hint 33

positive does not mean stronger; it just says when one variable goes up the other goes down;

Front 34

When the subject would not know whether he or she is a member of the control group or the experimental group

Back 34

Single blind study

Front 35

When the experimenter or researcher and the subjects also is not aware of what subjects are in the control group or the experiments of group

Back 35

Double blind study

Front 36

when an experiment is flawed because the experimenter may unconsciously communicate his or her intent or expectation to the subjects it's called

Back 36

Experimenter effects

Front 37

This model includes baseline (A), intervention is implemented (B), and outcome is examined with new baseline (A).

Back 37

ABA design

Front 38

When a distribution of scores Is not distributed normally

Back 38

A skewed distribution

Front 39

The middle score when the data or a ranged from highest to lowest

Back 39

The median (middle)

1,90,12,90,6,8,7; first Rank the scores from lowest to highest; 1,6, 7,8, 12,90,90; the median is 8

Front 40

It is the point where the most frequently occurring score falls

it will always be the highest point on the graph

Back 40

The mode (frequency)

1,10,19,1,10,19,19,6,54- the mode is 19 since it appears the most

Front 41

Positively skewed distribution

Back 41

Positively skewed distribution has an abundance of low scores and is asymmetrical. The direction of the tail is to the right.

i.e. all of the students scored very low on the test. the tail points you in the direction of the correct answers.

Front 42

Negatively skewed distribution

Back 42

Negatively skewed distribution reflects an abundance of high scores . The direction of the tail is to the left.

i.e. all the counselors score high on the test.

Front 43

A bar graph is also called

Back 43

A histogram

Front 44

This axis (horizontal) is used to represent the independent variable.

Back 44

The x-axis also called the abscissa

Front 45

This axis is the vertical axis which is used to scale for the dependent variable

Back 45

The y axis or the ordinate

Front 46

The measure of variance is calculated by determining the difference between the highest and the lowest score

Back 46

The range

Front 47

This is a threat to the internal validity of an experiment that occurs when subjects strive to prove that an experimental treatment that could threaten their livelihood really isn't effective at all.

i.e. counselors were asked to use computers as part of the teaching experience but were worried that the computers might ultimately take their jobs.

Back 47

The John Henry effect (livelihood)

Front 48

The peakedness of a frequency distribution

Back 48

Kurtosis

Front 49

This scale is used to distinguish separate Groups. It has no true 0 point and does not indicate order. Can only use addition and subtraction

Back 49

Nominal scale. An example is the categories for the DSM

Front 50

a behavior or circumstance that can exist on at least two levels of conditions. It is a factor that "varies" or is capable of change.

Back 50

Variable

Front 51

This scale provides relative placement or standing but does not have absolute differences.

Back 51

Ordinal scale. An example is a horse winning 2nd place.

Front 52

This scale has numbers scaled at equal distances but has not absolute zero point. Using this scale, distances between each number are equal yet it is unclear how far each number is from zero.

Example is a an IQ scale.

Back 52

Interval scale-IQ test

Front 53

This scale is an interval scale with a true point of zero. Addition, subtraction, multiplication and division can all be used

Back 53

Ratio scale. Examples are time, height, weight, temperature, volume or distance.

Front 54

When a trait which is not being evaluated, such as attractiveness or how well he or she is liked, influences a researcher's rating on Another trait, such as counseling skills.

The Halo effect occurs when a trait that is not being measured has influence on a trait that is.

i.e.a panel of investigators discovered that a researcher who completed a major study had unconsciously rated attractive females as better counselors

Back 54

The Halo effect(trait influence)

Front 55

The Rosenthal effect - the experimenter's beliefs may lead the experimenter to treat a participant a particular way to yield result and meet the experimenters expectations

i.e.a 3rd grade school counselor tells a 3rd grade teacher that a test revealed a certain number of children will excel during the school year. in reality no such test was administered. the kids excelled anyway.

Back 55

Rosenthal effect(experimenter expectancy)

Front 56

If the subject knows they are part of an experiment sometimes their performance is improved.

i.e. a counselor notes that a group of clients who are not receiving counseling, but are observed in a research study, are improving. her hypothesis is that the attention she has given them has been curative. this is known as the

Back 56

The Hawthorne effect(observer effect,performance improves)

Hint 56

Hawthorne Works of the Electric company-if subjects know they are a part of an experiment or get more attention, their performance improves

Front 57

When observations are made and the subjects behavior is influenced by the presence of the researcher it is called

Back 57

Reactive effect

Front 58

the points that divide a distribution into fourths

Back 58

quartile

Front 59

predicts that very high and very low scores will move toward the mean if a test is administered again

Back 59

statistical regression

Front 60

a within person analysis ( was your jog today faster than your jog from yesterday)

Back 60

Ipsative(inner) analysis

Front 61

implies an analysis between individuals

Back 61

normative analysis

Front 62

also known as a chain referral sample where you drum up other subjects for your study. You would get referrals from others in the same situation

Back 62

snowball sample

Front 63

this test is used to determine whether two uncorrelated/unmatched means differ significantly when data are nonparametric

Back 63

statistical test-Mann-Whitney U test-

Hint 63

U reminds me of uncorrelated

Front 64

this is a rank test used in place of a t test when the data are nonparametric and you wish to test whether two correlated means differ significantly from each other

Back 64

Statistical test- Wilcoxon signed rank tests

Front 65

when the research goes from specific to generalization

Back 65

inductive reasoning(narrow)

i.e.a counselor treats a client's phobia using a paradoxical strategy and then states that paradox is the treatment of choice for phobics.

Front 66

reduces the general to the specific in research

Back 66

deductive (sounds like reductive) reasoning

i.e.if the researched stated or observed that many clients being cured of their phobias via paradox and so assumed that Mr smiths phobi would be cured in the same manner.

Front 67

this will tell what would most likely occur if the same individual was re-tested

Back 67

standard error of measure

i.e. Mike takes a math achievement test. In order to predict his score if he takes the test again the counseller must know the standard error of measure

Front 68

this was created by Rensis Likert, this scale helps improve the overall degree of measurement. i.e. the response categories are strongly agree, agree, disagree, strongly disagree

Back 68

Likert Scale

Front 69

The flaws in research

Back 69

Bubbles

Front 70

the standard deviation squared

i.e.a popular I Q test has an SD of 15. If the mean IQ is 100 then 68% of the people who take the test will score between 85 and 115.

Back 70

variance

Front 71

68.26% of scores fall within + or -1 of the mean;95.44% of scores fall within + or -2 SD of the mean; 99.74% of scores fall within + or -3 SD of mean

Back 71

68%, 95%, 99% Rule

most scores will fall within 3 SD's or 99% of the normal curve

Front 72

The square root of the variance

The greater the SD, the greater the spread

Back 72

Standard Deviation

The standard deviation is an indication of how tightly data is clustered around the mean or "average" set of data in a bell curve.

Front 73

the most elementary type of standard score; it is simply the standard deviation

Back 73

Z score (standard scores)

Front 74

uses a mean of 50 with each SD with every 10 points landing above or below the mean

a t-score of 60 would equal +1 SD or z score while a t-score of would be -1.0

Back 74

t-score (transformed scores)

Front 75

distribution is flatter and more spread out than the normal curve

Back 75

platykurtic

Front 76

curve is very tall and thin and peaked

Back 76

leptokurtic

Front 77

divides the distribution into nine (9) equal parts with stanine 1 as the lowest ninth and 9 as the highest 9th. In this system 5 is the mean

Back 77

stanine (nine) scores

Front 78

alternative hypothesis or affirmative hypothesis

Back 78

asserts that the IV has caused a change. this is opposite of the null hypothesis where there is expectations of no change.

Front 79

descriptive statistics

Back 79

describe data- i.e. mean median and mode

Front 80

Percentile rank

Back 80

Descriptive statistic that tells the counselor what percentage of the cases fell below a certain level

Front 81

Percentage score

Back 81

Another way of stating a raw score

Front 82

between subjects design

Back 82

When our research study uses different subjects for each condition; Each subject receives only one value of the independent variable

Front 83

Within subjects design

Back 83

When a research study uses the same subjects; 2 or more values or a levels of the IV are administered to each subject

Front 84

Probability

Back 84

The level of significance

Front 85

.05 level indicates

Back 85

That the difference would occur via chance only 5 times and 100

Front 86

The smaller the value for P Or probability

Back 86

The more stringent the level of significance. .001=there is only 1 chance in 1000 that the results are due to chance

Front 87

organisms variable

Back 87

A variable a researcher cant control yet it exist such as height, weight or gender

Front 88

The probability of committing a type 1 error

Back 88

Equals the level of significance; The level of significance is called the alpha level

Front 89

the power of statistical test

Back 89

1 minus beta; power is a statistical tests ability To reject correctly a false null hypothesis

Front 90

type 1 and type ll errors are like a seesaw because

Back 90

When one goes up the other goes down; If the significance level is lowered for type one errors, it raises the risk of committing a type 2 error

Front 91

statistical test-kruskal-wallis

Back 91

used instead of the one way ANOVA when the data are nonparametric

Front 92

statistic test- Spearman correlation or Kendalls Tau

Back 92

used in place of the Pearson r when parametric assumptions cannot be utilized

Front 93

statistical test- chi square nonparametric test

Back 93

Examines whether obtained frequencies differ significantly from expected frequencies

Front 94

parametric and nonparametric

Back 94

in statistics, a set of data has a normal vs. a non-normal distribution, respectively.

Front 95

parametric tests

Back 95

make certain assumptions about a data set; namely, that the data are drawn from a population with a specific (normal) distribution.

Front 96

non-parametric tests

Back 96

make fewer assumptions about the data set; It usually means that you know the population data does not have a normal distribution.

Front 97

Pearson Product-Moment Correlation or Pearson r is used with what data

Back 97

interval or ratio data

Hint 97

uses I and R

Front 98

Spearman Rho correlation is used for what data

Back 98

ordinal data

Hint 98

rh"o"- O

Front 99

harmonic mean

Back 99

Front 100

factorial designs

Back 100

Includes 2 or more IV's. IV's are sometimes called level; its does not mean hierarchy it may mean types such as Individual therapy and group therapy

Front 101

Factorial experiments

Back 101

Several experiamental variables or factors are investigated and interactions are noted

Front 102

The benefit of standard scores such as percentiles, T scores, z scores, stanines, or standard deviations over raw scores is that

Back 102

A standard score allows you to analyze the data in relation to the properties of the normal bell curve

Front 103

Double barred histogram

Back 103

Compares two distributions of scores such as pre and post test scores

Front 104

scattergram

Back 104

Front 105

Standardized test always have

Back 105

Formal procedures for test administration and scoring

Front 106

2 types of developmental studies

Back 106

Cross-sectional- Assessed at 1 point in time

Longitudinal- The same people are studied over a period of time

Front 107

Demand characteristics

Back 107

Any knowledge whether correct or incorrect that the subject in an experiment is aware of that can influence his or her behavior

Front 108

Duncan's multiple range,tukeys, or scheff's test

Back 108

Test significant differences between group means if an anova yields a significant F value

Front 109

Counterbalancing In an experiment

Back 109

Is used to control for the fact that order of an experiment could impact upon its outcome; Switching the order in which stimula or present it to a subject in a study is also known as this

Front 110

Sampling groups

Back 110

1. random sample-each subject has the same probability of being selected

2. stratified- a special characteristic needs to be represented

3.cluster-using a selects a portion of people from the overall sample

4. horizontal-selects subjects from a single socioeconomic group

5. vertical sampling- persons from 2 or more socioeconomic groups are selected

6.quota-a specific number of cases is necessary

Front 111

operational definition

Back 111

outlines a procedure

Front 112

axiom

Back 112

universally accepted idea needing no additional proof. i.e. gravity exists

Front 113

A researcher gives a depressed patient a sugar pill and the individual's depression begins to lift. This is known as

Back 113

The placebo effect-has a positive effect

a nocebo- has a negative effect- A doctor comments that a person with such-and-such condition has only 6 weeks to live.

Front 114

qualitative vs quantitative research

Back 114

Front 115

bell shaped curve, the mean, median and mode fall in between this curve, gsussian curve

Back 115

normal curve

Front 116

an experiment always has 3 kinds of variable

Back 116

IV, DV, and controlled variable which stays constant

Front 117

is the act and method of studying something as objectively, consistently, thoroughly and systematically as possible in order to increase knowledge or understanding of phenomena or behavior.

Back 117

Research

Front 118

this means consistency. No assessment instrument is perfect but _________ is an indicator of how consistent a test instrument is in research.

Back 118

reliability

Front 119

A research project involves the weight measurement of test subjects. Researchers want to be sure that the scales will do the job of measuring the subject's weight in terms of the intention of the research. The researchers are concerned with test_________:

Back 119

Validity tells how well the test does the job. Does the instrument (scale) measure what it is intended to measure? Researchers are always going to be concerned with the test instrument's reliability and the test instrument's validity.