Methods Final Exam Key Terms

Front 1

4 kinds of univariate statistics

Back 1

1. raw count
2. percentage frequency distribution
3. bar chart
4. frequency polygon

Front 2

NONPROBABILITY SAMPLING

Back 2

- less accurate than probability sampling (used when that option is impossible)
- does not use random selection

Front 3

PROBABILITY SAMPLING

Back 3

- all possibilities have a known probability of being drawn
- best for representative sample
- applies particular reasoning to natural and social science

Front 4

RANDOM SAMPLING

Back 4

- each element has an equal chance of being selected
- representative of entire population

Front 5

RANDOM ASSIGNMENT

Back 5

- assigning the elements within the sample to control and experimental group
- happens after random sampling

Front 6

SIMPLE RANDOM SAMPLE

Back 6

- starts with sampling frames
- uses random process to select cases so each element in population has equal chance of selection

Front 7

SAMPLING FRAME

Back 7

- list that approximates elements in your target population (aka the collection of elements)
- EX: telephone directories, tax records

Front 8

SAMPLING RATIO

Back 8

- the proportion of the population in the sample
- sample size/target population

Front 9

STRATIFIED SAMPLING

Back 9

- probability sample
- population divided into subpopulations (strata) and random samples taken of each stratum
- mutually exclusive & exhaustive

Front 10

SYSTEMATIC SAMPLING

Back 10

- involves selecting sample in ordered process
- take every nTH person from sampling frame

Front 11

CLUSTER SAMPLING

Back 11

- involves randomly sampling clusters and then randomly sampling elements from within the clusters

Front 12

CONVENIENCE SAMPLING

Back 12

- nonprobability sampling
- get any cases in any manner that is convenient

Front 13

QUOTA SAMPLING

Back 13

- nonprobability sampling
- get a preset number of cases in each of several predetermined categories (using haphazard methods)

Front 14

SNOWBALL SAMPLING

Back 14

- nonprobability sampling
- get cases using referrals from one or a few cases, then referrals from those cases and so on
- example: drug dealers

Front 15

CENTRAL LIMIT THEOREM

Back 15

- the more random samples, the closer and closer you get to the true population
- allows us to generalize from one sample to the population

Front 16

SEVEN ELEMENTS OF EXPERIMENT

Back 16

1. Treatment (independent variable)
2. Dependent Variable
3. Pretest
4. Posttest
5. Control Group
6. Experimental Group
7. Random assignment

Front 17

PRE-EXPERIMENTAL DESIGN

Back 17

- lack random assignment
- are compromises or shortcuts to classical design
- make inferring a causal relationship hard

Front 18

CLASSICAL DESIGN

Back 18

- random assignment, pretest and posttest, experimental group, control group

Front 19

Why is the control group important?

Back 19

- allows us to isolate effects of the treatment
- helps eliminate alternative explanations that could harm our attempts to establish causality

Front 20

ONE-SHOT CASE-STUDY DESIGN

Back 20

- pre-experimental design
- one group, treatment, and posttest
- NO random assignment b/c one group

Front 21

ONE-GROUP PRETEST-POSTTEST DESIGN

Back 21

- pre-experimental design
- one group, treatment, and postest
- NO CONTROL GROUP or RANDOM ASSIGNMENT

Front 22

STATIC-GROUP COMPARISON

Back 22

- pre-experimental design
- two groups, posttest, and a treatment
- NO RANDOM ASSIGNMENT OR PRETEST

Front 23

QUASI-EXPERIMENTAL DESIGN

Back 23

- make identifying causal relationship more certain than pre-experimental designs
- lack a pretest
- less control of independent variable for researcher

Front 24

TWO-GROUP POSTTEST-ONLY DESIGN

Back 24

- quasi-experimental design
- identical to static-group comparison with one exception: random assignment
- all elements of classical design, without pretest

Front 25

INTERRUPTED TIME-SERIES DESIGN

Back 25

- quasi-experimental design
- measures dependent variable on one group over time using multiple dependent variables measures before & after treatment

Front 26

EQUIVALENT TIME-SERIES DESIGN

Back 26

- quasi-experimental design
- one-groupd esign
- extends over time period
- applies same treatment multiple times

Front 27

LATIN SQUARE DESIGN

Back 27

- quasi-experimental design
- allows us to observe how several independent variables in different orders affect dependent variable
- one group with random assignment
- no control

Front 28

SOLOMON FOUR-GROUP DESIGN

Back 28

- quasi-experimental design
- examines whether order or sequence of receiving treatment has effect
- random assignment and two groups

Front 29

FACTORIAL DESIGNS

Back 29

- quasi-experimental design
- uses two or more independent variables in combination to consider impacts simultaneously

Front 30

INTERNAL VALIDITY

Back 30

when the independent variable, and nothing else, influences the dependent variable

Front 31

SELECTION BIAS

Back 31

- threat to internal validity in experimental research
- groups in experiment are not equivalent or differ in regards to dependent variable at beginning making them hard to compare
- detect by comparing pretest scores

Front 32

HISTORY EFFECT

Back 32

- threat to internal validity in experimental research
- result of an event unrelated to treatment occurring during experiment and influencing dependent variable
- address threat: use design with pretest and control group; both experimental and control group will show similar changes over time

Front 33

MATURATION EFFECT

Back 33

- threat to internal validity in experimental research
- result of a natural threat within the participants during the experiment that affects dependent variable
- address threat: use design with pretest and control group; both experimental and control group will show similar changes over time

Front 34

EXPERIMENTAL MORALITY

Back 34

- threat to internal validity in experimental research
- participants fail to participate through entire experiment
- address threat: report # of participants at all stages of experiments to detect

Front 35

DIFFUSION OF TREATMENT

Back 35

- threat to internal validity in experimental research
- occurs when treatment 'spills over' from experimental to control group, making control group modify their behavior with knowledge of treatment
- address threat: isolate groups; make them promise not to disclose info to other participants; ask about diffusion in post-interview

Front 36

EXTERNAL VALIDITY

Back 36

- the effectiveness of generalizing experimental findings beyond a specific study

Front 37

NATURALISTIC GENERALIZATION

Back 37

- issue to maximize external validity
- ability to generalize from what was learned in controlled laboratory setting to "real life"

Front 38

MUNDANE REALISM

Back 38

- issue to consider to maximize external validity
- asks whether an experiment or situation is like the real world

Front 39

REACTIVITY

Back 39

- issue to consider to maximize external validity
- a general threat that arises because participants are aware they are being studied

Front 40

HAWTHORNE EFFECT

Back 40

- a reactivity result named after famous case in which participants responded to fact that they were in experiment more than to treatment

Front 41

THEORETICAL GENERALIZATION

Back 41

- issue to consider to maximize external validity
- asks whether we can generalize from concepts in abstract theory we wish to test to measures and experimental activities

Front 42

EXPERIMENTAL REALISM

Back 42

- subset of theoretical generalization
- the impact of an experimental treatment or setting on people
- occurs when participants are truly influenced

Front 43

LABORATORY EXPERIMENT

Back 43

- an experimental study in an artificial setting over which experimenter has control
- high internal validity but low external validity

Front 44

FIELD EXPERIMENT

Back 44

- a study that takes place in a natural setting
- high external validity but low internal validity
- more generalizable but less controlled

Front 45

PRINCIPLES TO AVOID WHEN WRITING SURVEY QUESTIONS

Back 45

1. Avoid jargon, slang, abbreviations (many meanings aren't universal)
2. Avoid ambiguity, confusion, and vagueness (can cause inconsistencies in answers)
3. Avoid emotional language and prestige bias (emotions can color respondent's answers; titles and status can away a person to answer a particular way)
4. Avoid leading questions (one that leads respondent to one response by wording; can lead to positive or negative answers)

Front 46

Categories of Interviewer Bias

Back 46

1. Errors by respondent (forgetting, embarassment, lying due to presence of others)
2. Unintentional errors or interviewer sloppiness (misreading question, omitting question, wrong order)
3. Intentional subversion by interviewer (altering answers, rewording question)
4. Influence due to interviewer's expectations about respondent's answers based on respondent's appearance, living situation, etc.
5. Failure of interviewer to probe (or probe properly)
6. Influence on answers due to interviewer's appearance, tone, attitude, reactions to answers, or comments made outside interview schedule

Front 47

Pros & Cons of Mail Survey

Back 47

Pros: cheap, wide geographical area, convenient, anonymity
Cons: low response rate, lack of control of conditions survey is taken, limit of questions, someone other than sampled respondent may take survey

Front 48

Pros & Cons of Telephone Survey

Back 48

Pros: most of population reached (95%), wide geographic area, high response rates (80%), interviewers can control questions and probe
Cons: moderately high cost and limited interview length, inconvenient, respondents without telephones can't be reached, potential interviewer bias

Front 49

Pros & Cons of Face-to-face Survey

Back 49

Pros: highest response rates, permit long & complex questionnaire, extensive probing
Cons: high cost (training, travel, supervision), interviewer bias

Front 50

Pros & Cons of Web Surveys

Back 50

Pros: fast & inexpensive, allow flexible design
Cons: cheap and easy so not best quality, some have no access to Internet, technical glitches

Front 51

RECENCY EFFECT

Back 51

- in survey research, when respondents choose the last answer response offered rather than considering all choices
- solution: present responses on continuum with neutral position in middle

Front 52

ORDER EFFECTS

Back 52

- a result in survey research in which a topic or some question asked before others influence respondents' answers to later questions
- appear strongest for those who lack strong views

Front 53

CONTEXT EFFECTS

Back 53

- a result in survey research when an overall tone, setting, or set of topics heard by respondents affect how they interpret the meaning of subsequent questions

Front 54

FUNNEL SEQUENCE

Back 54

- context effect
- organization of general questions to more specific ones

Front 55

ECOLOGICAL FALLACY

Back 55

- limitation with existing statistics
- means the assumption that relationships found among groups can be generalized to individuals

Front 56

CONTENT ANALYSIS

Back 56

- when you gather and analyze the content of text (in words, meanings, pictures, symbols, ideas, themes, or any communicated message)
- Useful for research questions: regarding a large volume of text, content that may be at a distance or scattered, content that is difficult to see or document with causal observation

Front 57

MEASURES OF CENTRAL TENDENCY

Back 57

- statistical measures that summarize information about the distribution of data for one variable into a single number
- mean, median & mode

Front 58

CODEBOOK

Back 58

- a document that describes the procedure for coding variables and their location in a format computers can use

Front 59

UNIVARIATE STATISTICS

Back 59

- describe a single variation in isolation

Front 60

Three Features of a Scattergram

Back 60

1. Form (Linear or Curvilinear; Independence or no relationship)
2. Direction (Positive or Negative)
3. Precision (amount of spread in the graph)

Front 61

TYPE I ERRORS

Back 61

- mistake made in saying a relationship exists when non does
- researcher claims causal relationship and real situation indicates none

Front 62

TYPE II ERRORS

Back 62

- mistake made in saying a relationship doesn't exist when in fact it does
- researcher indicates no relationship and real situation implies causal relationship

Front 63

Six Steps in Conducting Field Research

Back 63

1. Prepare oneself, read literature, and defocus.
2. Select a field site and gain access.
3. Enter field and establish social relations with members.
4. Adopt a social role, learn ropes, and get along with members.
5. Watch, listen, and collect quality data. (begin analysis, focus on specific aspects to setting, conduct field interviews)
6. Disengage and physically leave setting (write research report)

Front 64

DEFOCUSING

Back 64

- in field research, when researcher removes his or her from past assumptions and preconceptions to become more open to events in field site

Front 65

Validity in Field Research

Back 65

Ecological Validity: the degree to which the social world you describe matches members' world.
Natural History: detailed description of how you conducted project.
Member Validation: occurs when you take research back to members and they judge adequacy.
Competent Insider Performance: ability of nonmember to interact effectively as member or pass as one.

Front 66

Three Ethical Concerns in Field Research

Back 66

1. Covert Research (undermines trust between researchers and society)
2. Confidentiality (keeping information from others in field confidential and disguising members names in field notes)
3. Involvement with illegal behavior (researchers are sometimes involved with illegal activity --> guilty knowledge)

Front 67

JOTTED NOTES

Back 67

- field notes inconspicuously written while in field site
- convenient for short memory triggers

Front 68

DIRECT OBSERVATION NOTES

Back 68

- field notes that attempt to include all details and specifics of what researcher witnessed in field
- ordered chronologically
- concrete details

Front 69

INFERENCE NOTES

Back 69

- field notes involving listening and comparing and then interpreting
- create social meaning

Front 70

ANALYTIC MEMOS

Back 70

- field notes a qualitative research takes while developing more abstract ideas, themes or hypotheses
- collections of thoughts and digressions into theory

Front 71

PERSONAL NOTES

Back 71

- personal feelings and emotional reactions
- reflections that help cope with stress

Front 72

INTERVIEW NOTES

Back 72

- help make sense of notes

Front 73

MAPS, DIAGRAMS & ARTIFACTS

Back 73

- notes that organize events in field
- convey field site to others
- orient data and perform analysis
- objects serve as visible reminders

Front 74

MACHINE-RECORDED DATA

Back 74

- provide close approximation to what occurred in field
- helps recall events and observe what does not happen

Front 75

Similarities between H-C & Field Research

Back 75

1. Researcher's perspective integral in process
2. Many details to understand data
3. Grounded theory with dialogue
4. Translate meaning system
5. Meaning acted and constructed by people within structures
6. Use limited generalizations depending on context

Front 76

CRITICAL INDICATOR

Back 76

- in H-C, unambiguous evidence, usually sufficient for inferring a specific theoretical relationship

Front 77

SUPPORTING EVIDENCE

Back 77

- documents such as primary or secondary sources

Front 78

NON-SOURCE BASED KNOWLEDGE

Back 78

- general information available to a researcher based on reasoning or awareness of historical circumstances

Front 79

EXTERNAL CRITICISM

Back 79

- evaluation of authenticity
- locating place and time of document's creation

Front 80

INTERNAL CRITICISM

Back 80

- evaluation of authenticity and credibility of primary sources to determine if what is written is accurate and true

Front 81

Limitations of Secondary Historical Sources

Back 81

Implicit theories are hidden by narrative --> implicit theories come from historian's school of historiography and are influenced by his/her training --> because of theoretical bias, they only select certain information

Front 82

EQUIVALENCE

Back 82

- involves whether we can make comparisons across divergent contexts
- without it, comparison can be impossible

Front 83

LEXICON EQUIVALENCE

Back 83

- similarity of words & phrases to express identical meaning in a different language or translation of one language to another
- significant in H-C research because meaning of words change with time

Front 84

CONTEXTUAL EQUIVALENCE

Back 84

- similarity of social roles, norms, or situations across different cultures or historical periods
- example: role of religious leaders

Front 85

CONCEPTUAL EQUIVALENCE

Back 85

- similarity of ideas or concepts across divergent cultural or historical settings
- example: the class system differs across societies

Front 86

MEASUREMENT EQUIVALENCE

Back 86

- similarity of measures that will accurately represent a construct or variable in divergent cultural or historical settings

Front 87

CASE-STUDY COMPARATIVE

Back 87

- a researcher compares one or two particular cultures in depth
- does not make broad generalizations
- can be any set of cases

Front 88

CULTURAL-CONTEXT RESEARCH

Back 88

- comparative study focused on comparing small number of societies or cultures that represent theoretical types to allow generalizations to other societies of similar types

Front 89

CROSS-NATIONAL RESEARCH

Back 89

- comparative study that examines data (usually quantitative) for several variables across many nations and analyzes data
- looking just at nation-states

Front 90

TRANSNATIONAL RESEARCH

Back 90

- comparative study approach that examines and compares multinational units
- more regional

Front 91

Differences between quantitative & qualitative data analysis

Back 91

1. Quantitative uses a few standardized techniques/Qualitative uses many diverse, nonstandard techniques
2. Quantitative analyzes after all data has been collected/Qualitative begins analysis while still collecting
3. Quantitative tests preexisting theories and hypotheses/Qualitative conceptualizes and builds new theory
4. Quantitative uses precise and compact abstract data/Qualitative uses imprecise, diffuse, relativity concrete data

Front 92

OPEN CODING

Back 92

- first coding of qualitative data
- examines data to condence into preliminary categories or codes
- locate themes and assign initial codes

Front 93

AXIAL CODING

Back 93

- second stage of coding
- researcher organizes the codes, links them, and discovers key analytic categories
- reinforces link between evidence and concepts

Front 94

SELECTIVE CODING

Back 94

- last stage in coding qualitative data that examines previous codes to identify and select data that will support conceptual coding categories developed

Front 95

METHOD OF AGREEMENT

Back 95

- qualitative data analysis
- establishes that cases have a common outcome and try to locate the common cause

Front 96

METHOD OF DIFFERENCE

Back 96

- qualitative data analysis
- focuses on differences among cases

Front 97

PATH DEPENDENCY

Back 97

- an analytic ideas used in narrative analysis
- explains process or chain of events as having beginning trigger to create sequence of events

Front 98

SELF-REINFORCING

Back 98

- sequence of path dependency
- examine how events, once set into motion, continue to operate on their own or proper later events
- example: QWERTY keyboard

Front 99

REACTIVE SEQUENCE

Back 99

- sequence of path dependency
- focuses on how each event responds to an immediately preceding one
- example: assassination of MLK triggered more civil rights law enforcement & expansion of welfare

Front 100

HISTORICAL CONTINGENCY

Back 100

- an analytic idea in narrative analysis that explains a process, event, or situation by referring to combination of factors that came together in particular time and place
- example: rise of large corporation in U.S.

Front 101

EVENT STRUCTURE ANALYSIS

Back 101

- qualitative data analysis (often conducted with software) that forces a researcher to specify links among sequence of events
- clarifies causal relationships

Front 102

Major Features of Narrative

Back 102

1. Tells story or tale
2. Has sense of movement
3. Contains interrelations or connections within context
4. Involves individuals or groups engaging in action and making choices
5. Has coherence (whole holds together)
6. Has temporal sequencing of a chain of events