Psych Research Methods

Front 1

Descriptive

Back 1

This type of research provides a “snapshot” of variables at a single time and place

Front 2

Correlational

Back 2

This type of research examines relationships among variables but only measures them- does not manipulate them.

Front 3

Experimental

Back 3

This type of research manipulates a variable to test whether there is a causal relationship with another variable.

Front 4

Inter-individual variability

Back 4

Joe is more extraverted than dave.

Front 5

intra-individual variability

Back 5

Jane is confident in some situations, but in others she feels insecure and uncertain of her abilities. This difference occurs within a single person.

Front 6

Determinism

Back 6

Assumption that all behavior is caused

Front 7

Empirical method

Back 7

Direct observation and experimentation to gain knowledge

Front 8

objectivity

Back 8

Freedom from personal bias or feelings

Front 9

Operationalism

Back 9

A concept is only as valid as the procedures used to measure it

Front 10

Parsimony

Back 10

Several explanations? Chose one with the fewest or simplest assumptions

Front 11

Determinism

Back 11

Assumption that all behavior is caused

Front 12

Empirical method

Back 12

Direct observation and experimentation to gain knowledge

Front 13

objectivity

Back 13

Freedom from personal bias or feelings

Front 14

Operationalism

Back 14

A concept is only as valid as the procedures used to measure it

Front 15

Parsimony

Back 15

Several explanations? Chose one with the fewest or simplest assumptions

Front 16

Accumulation of Evidence

Back 16

Confidence in knowledge increased with repeated observation or demonstration

Front 17

Converging operations

Back 17

Multiple measures, multiple research designs

Front 18

Public Nature of Information

Back 18

Scientfic findings should be accessible to everyone

Front 19

Nuremberg Code

Back 19

Code developed to reduce ethical violations following WW2 (trial of Nazi's regarding "medical" research

Front 20

Belmont Report

Back 20

U.S. government document that requires that research with human participants adhere to principles of justice, benevolence, and respect for persons

Front 21

Plagiarism

Back 21

Representing someone else's work as ones own

Front 22

Measurement

Back 22

assignment of numbers to objects or events according to specific rules

Front 23

Conceptual variable

Back 23

the abstract concepts you are interested in

Front 24

operational variables

Back 24

the variables of interest, defined in terms of how they will be created, measured, or manipulated

Front 25

Nominal scale

Back 25

the number indicates the presence or absence of some quality, or the membership in some category

Front 26

Ordinal scale

Back 26

Number indicates comparison: a greater or lesser quantity of some variable

Front 27

interval scale

Back 27

each step on the scale represents an equal amount of the construct, no true zero

Front 28

Ratio scales

Back 28

Each step on the scale represents an equal amount of the construct, includes a true zero
"twice as much"

Front 29

Free Format self report measures

Back 29

uses open-ended questions such as "tell me about yourself" or "what do you think are the biggest concerns..."

Front 30

Fixed Format self report measures

Back 30

uses items with predetermined response options (a, b, c or d)

Front 31

Behavioral Measures

Back 31

things such as:
- minutes spent playing with a particular toy
-length of eye contact
-reaction time

Front 32

Psycho-physiological meaures

Back 32

Measuring an internal state, such as heart rate, brain activity, blood levels or stress hormones to infer something about a psychological variable

Front 33

reactivity

Back 33

changes in responding that occur as a result of being measured

Front 34

random error

Back 34

chance fluctuations in measurement

Front 35

Systematic error

Back 35

tendency to consistently over or under estimate the true value of the variable being measured

Front 36

reliability

Back 36

Extent to which a measure is free from random error

Front 37

test-retest reliability

Back 37

correlation between scores on Measure X at T1 and scores on Measure X at T2 using the same instrument

Front 38

traits

Back 38

Personality variables that remain stable over time; consistent characteristics of an individual

Front 39

States

Back 39

characteristics of a person that fluctuate over time, such as mood, level of motivation or energy, attention

Front 40

Internal Consistency Reliability

Back 40

•Correlations among all items on a single measurement scale
•Measure of reliability at a single time

Front 41

Inter-rater reliability

Back 41

Correlations between ratings by more than one judge
•Both rating the same behavior

Front 42

Construct Validity

Back 42

Does the measured variable actually capture the conceptual variable?

Front 43

Criterion Related validity

Back 43

does this measure predict future behavior or correlate well with current behavior?

Front 44

Validity

Back 44

Measure can be reliable without being valid.

Measure CANNOT be valid unless it is reliable.

Front 45

Ways to improve validity

Back 45

Pilot test
Converging operations
Sensitive, discriminating measures
Clear, unambiguous items Engage participants
Non-reactive measures
Cover range of content
Use existing measures

Front 46

Conducting Research??

Back 46

Ethical conduct of research with human and animal participants is responsibility of entire research team

Front 47

Consent Process

Back 47

Information
Comprehension
Voluntariness

Front 48

Debriefing

Back 48

Explain purpose
remove any harmful effects
Clear up deception

Front 49

When is deception allowed?

Back 49

When it is:
-scientifically justified
-no alternatives
-not about risks
-cleared up in debriefing
-data can be withdrawn

Front 50

Tuskeegee

Back 50

This study was unethical because the Participants were not given full information about the procedures in the study and the risks involved when they were asked to consent to participation and also the researchers allowed human beings to be harmed by not providing treatment for a disease once treatment became available

Front 51

Nazi Experimentation

Back 51

Responsibility on ALL persons involved in the research

Front 52

Sampling

Back 52

The process of selecting participants from a population for a research study

Front 53

Population

Back 53

The group that the researcher is trying to understand or study – can be people, animals, institutions or other entities

Front 54

Sample

Back 54

The group of selected members of a population that actually participate in the research (are measured in order to draw inferences about the population)

Front 55

Sampling Frame

Back 55

A list of all members of a population

Front 56

Simple Random Sampling

Back 56

the probability of being selected from the population is the same for every member of the population

Front 57

Stratified Random Sampling

Back 57

Population is broken down into subgroups
Specific number of participants are drawn at random from within each strata

Front 58

Non probability sampling

Back 58

The probability of being selected from the population is the not known
No satisfactory sampling frame exists
Samples are not representative

Front 59

Sampling Bias

Back 59

Measurements may vary from “true” measure of a population, because a non-representative sample was used

Front 60

unacknowledged Participant

Back 60

Researcher is embedded in a situation as one of the group members & studies behaviors but other group members are unaware of the research.

Front 61

unacknowledged observer

Back 61

Researcher does not interact with participants; observes behaviors without the participants’ knowledge of the research.

Front 62

Acknowledge observer

Back 62

Researcher does not interact with participants; observes behaviors and participants are aware that they are being studied.

Front 63

Archival Research

Back 63

“Analysis of …existing records of public behavior” (Stangor p. 130)

Front 64

descriptive statistics

Back 64

numbers that summarize the measured characteristics of a particular sample

Front 65

Inferential Statistics

Back 65

using measurements from a sample to draw conclusions about the characteristics of a population.

Front 66

Sampling Distribution

Back 66

A frequency distribution of a particular statistic that graphs the likelihood of every possible outcome for that statistic, based on repeated samples of a given size drawn from the population.

Front 67

Null Hypothesis

Back 67

Assumption that observed data reflect only what would be expected from the sampling distribution (Stangor p. 420)

Front 68

Experimental Hypothesis

Back 68

What the researchers predict relationship between variables will be

Front 69

Alpha or Significance Level

Back 69

preset standard for when we will consider an outcome statistically significant

Front 70

P- Value

Back 70

statistical likelihood of an observed pattern of data [calculated on the basis of the sampling distribution of some statistic]

Front 71

Fail to Reject the Null

Back 71

Based on observed outcomes, you don’t have sufficient justification to reject H0

The data don’t support HE

p is greater than α

Front 72

Reject the Null

Back 72

The H0 can’t account for the observed outcomes
p is greater than α

The data don’t PROVE your hypothesis, but support it
Always some chance (5% if α = .05) that we have made an error

Front 73

Type I error

Back 73

Reject Null when the null is true (no difference exists)

Front 74

Type II error

Back 74

Fail to reject Null when the null is false (a real difference exists)

Front 75

How to avoid Type I errors

Back 75

Clear predictions
stringent alpha
Avoid “fishing”

Front 76

Sampling Distribution

Back 76

A frequency distribution of a particular statistic that graphs the likelihood of every possible outcome for that statistic, based on repeated samples of a given size drawn from the population.

Front 77

Null Hypothesis

Back 77

Assumption that observed data reflect only what would be expected from the sampling distribution (Stangor p. 420)

Front 78

Experimental Hypothesis

Back 78

What the researchers predict relationship between variables will be

Front 79

Alpha or Significance Level

Back 79

preset standard for when we will consider an outcome statistically significant

Front 80

P- Value

Back 80

statistical likelihood of an observed pattern of data [calculated on the basis of the sampling distribution of some statistic]

Front 81

Fail to Reject the Null

Back 81

Based on observed outcomes, you don’t have sufficient justification to reject H0

The data don’t support HE

p is greater than α

Front 82

Reject the Null

Back 82

The H0 can’t account for the observed outcomes
p is greater than α

The data don’t PROVE your hypothesis, but support it
Always some chance (5% if α = .05) that we have made an error

Front 83

Type I error

Back 83

Reject Null when the null is true (no difference exists)

Front 84

Type II error

Back 84

Fail to reject Null when the null is false (a real difference exists)

Front 85

How to avoid Type I errors

Back 85

Clear predictions
stringent alpha
Avoid “fishing”

Front 86

Sampling Distribution

Back 86

A frequency distribution of a particular statistic that graphs the likelihood of every possible outcome for that statistic, based on repeated samples of a given size drawn from the population.

Front 87

Null Hypothesis

Back 87

Assumption that observed data reflect only what would be expected from the sampling distribution (Stangor p. 420)

Front 88

Experimental Hypothesis

Back 88

What the researchers predict relationship between variables will be

Front 89

Alpha or Significance Level

Back 89

preset standard for when we will consider an outcome statistically significant

Front 90

P- Value

Back 90

statistical likelihood of an observed pattern of data [calculated on the basis of the sampling distribution of some statistic]

Front 91

Fail to Reject the Null

Back 91

Based on observed outcomes, you don’t have sufficient justification to reject H0

The data don’t support HE

p is greater than α

Front 92

Reject the Null

Back 92

The H0 can’t account for the observed outcomes
p is greater than α

The data don’t PROVE your hypothesis, but support it
Always some chance (5% if α = .05) that we have made an error

Front 93

Type I error

Back 93

Reject Null when the null is true (no difference exists)

Front 94

Type II error

Back 94

Fail to reject Null when the null is false (a real difference exists)

Front 95

How to avoid Type I errors

Back 95

Clear predictions
stringent alpha
Avoid “fishing”

Front 96

Sampling Distribution

Back 96

A frequency distribution of a particular statistic that graphs the likelihood of every possible outcome for that statistic, based on repeated samples of a given size drawn from the population.

Front 97

Null Hypothesis

Back 97

Assumption that observed data reflect only what would be expected from the sampling distribution (Stangor p. 420)

Front 98

Experimental Hypothesis

Back 98

What the researchers predict relationship between variables will be

Front 99

Alpha or Significance Level

Back 99

preset standard for when we will consider an outcome statistically significant

Front 100

P- Value

Back 100

statistical likelihood of an observed pattern of data [calculated on the basis of the sampling distribution of some statistic]

Front 101

Fail to Reject the Null

Back 101

Based on observed outcomes, you don’t have sufficient justification to reject H0

The data don’t support HE

p is greater than α

Front 102

Reject the Null

Back 102

The H0 can’t account for the observed outcomes
p is greater than α

The data don’t PROVE your hypothesis, but support it
Always some chance (5% if α = .05) that we have made an error

Front 103

Type I error

Back 103

Reject Null when the null is true (no difference exists)

Front 104

Type II error

Back 104

Fail to reject Null when the null is false (a real difference exists)

Front 105

How to avoid Type I errors

Back 105

Clear predictions
stringent alpha
Avoid “fishing”

Front 106

Sampling Distribution

Back 106

A frequency distribution of a particular statistic that graphs the likelihood of every possible outcome for that statistic, based on repeated samples of a given size drawn from the population.

Front 107

Null Hypothesis

Back 107

Assumption that observed data reflect only what would be expected from the sampling distribution (Stangor p. 420)

Front 108

Experimental Hypothesis

Back 108

What the researchers predict relationship between variables will be

Front 109

Alpha or Significance Level

Back 109

preset standard for when we will consider an outcome statistically significant

Front 110

P- Value

Back 110

statistical likelihood of an observed pattern of data [calculated on the basis of the sampling distribution of some statistic]

Front 111

Fail to Reject the Null

Back 111

Based on observed outcomes, you don’t have sufficient justification to reject H0

The data don’t support HE

p is greater than α

Front 112

Reject the Null

Back 112

The H0 can’t account for the observed outcomes
p is greater than α

The data don’t PROVE your hypothesis, but support it
Always some chance (5% if α = .05) that we have made an error

Front 113

Type I error

Back 113

Reject Null when the null is true (no difference exists)

Front 114

Type II error

Back 114

Fail to reject Null when the null is false (a real difference exists)

Front 115

How to avoid Type I errors

Back 115

Clear predictions
stringent alpha
Avoid “fishing”