Research And Program Evaluation

Front 1

Research

Back 1

evidence based inquiry; the process of collecting and analyzing data for some purpose such as investigating a problem or answering a question

Front 2

Evidence based inquiry

Back 2

using empirical data which has been gathered systematically

Front 3

Quantitative

Back 3

1. assumes social fact have a single objective reality

2. ten to study samples of populations

3. researcher try no to influence collection of data (instruments)

4. Statistical methods comparing and contrasting groups occurs

5. Researchers examine for causes and relationships

Front 4

Qualitative

Back 4

1. assumes multiple reality socially constructed by the individuals and groups

2. tends to study individual units-person, family, community-in naturalistic setting

3. research may be primary instrument of collecting data (though observation)

4. researchers impressions, judgments, and feelings may be used

5. goal is to describe the nature of things

Front 5

Inductive

Back 5

begins in the real world, practical level. It tends to be descriptive, correlational, or historical and leads to the building of a theory

Front 6

deductive

Back 6

springs from a theory

Front 7

Quantitative research types

Back 7

1. survey

2. descriptive

3. Comparative

4. correlational

5. ex post facto (after the fact)

Front 8

Quantitative research types: survey

Back 8

uses questionnaires, interview ectera and is used to measure attitudes, perceptions, ect.

often has a low response rate (50 %) Unless you know the characteristics of the non-responders be cautions generalizing

Front 9

Quantitative research types: descriptive

Back 9

describes an existing state of events

Front 10

Quantitative research types: Comparative

Back 10

investigates whether there are differences between two or more groups

Front 11

Quantitative research types: correlational

Back 11

uses the correlation coefficient to determine the degree of relationship between tow or more variable or phenomena

Front 12

Quantitative research types: ex post facto (after the fact)

Back 12

aka: causal-comparative; studies possible causal relationships among variables; uses the t-test and the analysis of variance

Front 13

true experiment

Back 13

uses experimental and control group with random assignment to each. Goal is to determine a cause and effect relationship.

Front 14

Quasi-experimental

Back 14

like a true experiment but the randomization of the control and experimental group is not possible.

ex: 2 class rooms

Front 15

Qualitative research definition

Back 15

emphasizes gathering data about naturally occurring phenomena (individual and groups living experiences) and events

Front 16

Qualitative research types

Back 16

1. case study

2. Ethnography

Front 17

Qualitative research types case study

Back 17

the case may be a program, activity, or a set of individuals who are bounded in space and time

Front 18

Qualitative research types Ethnography

Back 18

which is a description and interpretation of a cultural or social group or system. Data is usually collected through observation and interviewing. Observer bias is important

Front 19

Mix method

Back 19

combines parts of qualitative and quantative

Front 20

Single-subject design

Back 20

studies the effects of a program or treatment on a an individual or group treated as an individual, usually after a baseline has been established

Front 21

Action research

Back 21

has an evaluative function; its goal is to improve services or programs

Front 22

Pilot study

Back 22

a small scale research effort often used to determine the feasibility of a large scale effort with emphasis on refining procedures and instrumentation

Front 23

Longitudinal research

Back 23

collecting data from the same group of individuals over a period of time. aka panel study

Front 24

Cross-sectional research

Back 24

collecting data from different groups at the same time

Front 25

Research outcomes may be measured

Back 25

1. within subjects: examining changes that happen to members of the group

2. between subjects- examining changes that occur between subjects or groups

Front 26

Meta analysis

Back 26

research comparing findings across studies

Front 27

Threats to internal validity

Back 27

1. selection of subjects

2. instrumentation

3. maturation

3. mortality or attrition

5. experimenter bias

6. history

7 statistical regression

Front 28

Threats to external validity

Back 28

1. selection of subjects

2. ecological validity

3. Subject reactions (hawthorn effect, demand characteristic, experimental bias, placebo)

4. Novelty and disruption effect

Front 29

Hawthorn effect

Back 29

influence on performance because of observation

Front 30

demand characteristic

Back 30

rumors, knowledge, that may effect the performance of the researcher.

Front 31

experimenter bias

Back 31

Rosenthal effect; researchers own expectations affect behaviors

Front 32

Nominal measurement

Back 32

the numbers represent the variables qualities or categories ex male female

Front 33

Ordinal

Back 33

numbers represent differences in magnitude of variable. ex. test scores

Front 34

interval

Back 34

the intervals between the number on a scale contain the same amount of variable throughout the scale

ex: Fahrenheit or centigrade

Front 35

ratio

Back 35

the number are on the scale which has a true zero.

ex weight; someone who is 200 is twice someone who is 100

Front 36

random sampling

Back 36

individuals in a population have an equal and independent change of being selected

Front 37

stratified sampling

Back 37

refers to selecting in such a way that major subgroups in the population will be sampled

Front 38

cluster sampling

Back 38

in this the unit is not an individual but naturally occurring groups of individuals such as classrooms, city blocks

Front 39

purposeful sampling

Back 39

ex; typical or extreme case selection

Front 40

Best number of sample for Correlational research

Back 40

30

Front 41

Best number of sample for ex post facto and experimental research

Back 41

15

Front 42

Best number of sample for survey research

Back 42

100

Front 43

descriptive statistical analysis

Back 43

aka summary; these techniques are used to descriptive data collected for research sample or population, and include means, standard deviations, frequency counts, and percentages

Front 44

Inferential statistical analysis

Back 44

used to may inferences form the sample to the population to determine the likelihood of something happening. Used the t-test and analysis of variance

Front 45

Parametric

Back 45

used when a sample is randomly drawn from a population and the data is normally distributed. You have para (two sided) data that yields a bell shaped curve. You assume a homogenous (similar) variance of the population

ex: t-test and analysis of variance

Front 46

Nonparametric

Back 46

you cannot make an assumption about the shape of the curve or variance of the population scores

ex: chi-square, mann-whitney u test, and Wilcoxon signed-ranks test

Front 47

independent Variable

Back 47

the variable you manipulate or vary to see what changes in the dependent variable. It precedes the dependent variable. aka stimulus variable. predictor variable or experimental variable

Front 48

dependent variable

Back 48

the variable you are measuring or trying to change. aka response variable, outcome variable, ore criterion variable

Front 49

Null hypothesis

Back 49

no differences between the variables or groups measured

Front 50

directional hypothesis

Back 50

the difference between the variables of the group in one direction

Front 51

non directional hypothesis

Back 51

there will be differences between the groups but unclear which way

Front 52

significance level

Back 52

generally .05, .01, and .001. at .5 you are willing to accept the possibly of rejecting the null hypothec in error five items out of on hundred times

Front 53

type I error (alpha)

Back 53

referring to the rejection of the null hypothesis (which states there is no difference) when it is correct

as the significance level goes up the chance of a type I error increases

Front 54

type II error (beta)

Back 54

failure to reject the null hypothesis when the is a difference

as the significance level goes down the chance type II error increases

Front 55

t test

Back 55

used to determine whether the mean scores of the two groups are significantly different fro each other. It can only be used when there are two groups (two mean scores).

Front 56

Analysis of variance (one way)

Back 56

analyzes variance for multiple variables producing and F value that must be compared to a f distribution table

Front 57

Analysis of variance (factorial)

Back 57

ANOVA to simultaneous determine whether mean scores on two or more variable (factors) differ significantly from each other and weather factor interact significantly with each other. Multiple independent and a single dependent

Front 58

Analysis of variance (multivariate) (MANOVA)

Back 58

measuring multiple independent and dependent variables

Front 59

Analysis of covariance (ANCOVA)

Back 59

measuring multiple independent and dependent variables where the dependent variable is controlled

Front 60

Post hoc or multiple comparison tests

Back 60

used to determine if the means are significantly different

Scheffes (most conservative)

Tukey HSD (honestly significant difference)

Newman-Keuls

Duncan new multiple range test

Front 61

Nonparametric tests

Back 61

when you cannon assume the distribution of scores of normally distributes (bell curve) or that the variance of the sample is similar to variance of the population (homogeneity)

ex: Mann-Whitney U test; Wilcoxen signed-rank test; Kruska-Wallis test

Front 62

Mann-Whitney U test

Back 62

when you collect data from two samples that are independent from each other and the scores are not normally distributed

Front 63

Wilcoxen signed-rank test

Back 63

when you have scores for two samples and these scores are correlated (that is you matched them ore got two score for each individual repeated measure) however the scores do not approximate a normal distribution

Front 64

Kruska-Wallis test

Back 64

when you have two mean scores on a single variable

Front 65

Chi-square

Back 65

used when you have nominal data (groups or categories) to determine whether two distribution differ significantly

Front 66

Solomon four group design

Back 66

examines the effect of any pretest used in experimental treatment