Test 2 Flashcards

Question

Criterion Related/Predictive Validity

Answer 1

- My questions/variables correlate in expected ways with a behavior (criterion) which is logically related to them - Show that you can predict respondents’ behaviors on the basis of how they answer your indicators

Answer 2

- That quality of measurement method that suggests that the same data would have been collected each time in repeated observations of the same phenomenon; quality of repeatability in untruths - Ex: ”Did you attend religious services last week?” would have higher reliability than question “About how many times have you attended religious services in your life?” - When it yields the same results on a number of different occasions - When it yields the same regardless of who is measuring - Do my questions/variables consistently yield the same results? - Are they dependable? - Measurements are reliable when they yield the same results on different occasions, regardless of who is measuring

Answer 3

-Measurements are reliable when they yield the same results on different occasions, regardless of who is measuring

Answer 4

- Questions are: - Too complex - Too vague - Irrelevant to respondents - Influenced by context (ex: given survey too late for doctor, and too early for bank) - Influenced by researcher - More valid → less reliable - More reliable → less valid

Answer 5

- Be careful to ask only about things respondents are likely to know the answer to - Ask about things relevant to them - Be clear in what you’re asking - Test-retest method - Split-half method - Using established measures - Reliability of research workers

Answer 6

-Ordinal level was their tool (use of reverse coded items) - Research goals: - Create a set of items that asses a domain-general level of trust in science and scientists - Establish the validity of our instrument with experts, using their feedback to refine our items - Field-test our survey to establish the instrument’s reliability - ”Trust in Science and Scientists” - Researchers developed Trust in Science and Scientists Inventory - Assesses of the level of trust in science and scientists - 21-item instrument - Has a reliability of Cronbach’s alpha of 0.86 - People aren’t trusting science or scientists - Climate change, vaccination, biological evolution, and genetically modified foods - Why? - Popular news stories (or personal perception) of researchers manipulating data, engaging in potentially unethical practices, using questionable methodologies, and withholding results - Lack of understanding of the fundamental tenets of the nature of science, such as the tentative nature of knowledge and the reliance on empirical evidence to describe the natural world - A lot of focus on trust (as an attitude) - Using subjective trust-based decision-making on scientific issues in the public realm (e.g., genetically modified foods, hydraulic fracturing, vaccines, climate change) may be at odds with the objective, data-driven nature of science, and may lead to decisions that stifle the associated scientific development, research, and applications - People do take credibility into account when determining level of trust; levels of trust based on credibility may come into play with medical procedures, pharmaceuticals, stem cell research, and diagnostic procedures, particularly under conditions in which our knowledge is limited, the conditions are personal, and individuals are potentially more vulnerable. - Worldviews, perceptions of knowledge, and thoughts about the plausibility of ideas can have a significant influence on our level of trust in others and in their associated institutions - Argue that items aligned with specific elements of trust may be responded to using other elements of trust or a combination of elements that are not representative of the targeted facet of trust. - We have found is that trust in science and scientists can have profound influence on policy, funding, and even legislation, and yet trust is a multifaceted construct that includes aspects of emotions, perceptions of credibility, perceptions of trustworthiness, world views, and knowledge. - They created a valid and reliable instrument - Goal was to develop the first instrument we are aware of that uses multiple items to assess trust in science and scientists on a general level. - Used pre-tests - Shared instrument with five external researchers with expertise in measurement… - Used preliminary pretest - Used formal pretest

Answer 7

- Decisions we make and strategies we use to collect observations about a phenomenon (people, groups, artifacts, behaviors, etc.) - We always make decisions about what will “count” as an observation. We always sample. - The scientific approach requires us to be systematic, clear, and precise about our decisions - Why? Because it is neither necessary nor efficient nor typically feasible to collect all possible observations about a phenomenon - 2 Types: - Probability - General term for samples selected in accord with probability theory, typically involving some random-selection mechanism - Non-Probability - More important with deductive path - Ex: Dr. takes blood sample and whatever she finds in sample can be said about all blood for person

Answer 8

- 1. More representative than non-probability samples - 2. Eliminate biases when selecting elements from sampling frame - 3. Allow us to apply a number of useful statistical laws - Estimate how close our sample statistics are to the population parameters -4. Used more typically in quantitative research - Rules of Thumb: - 1. The more heterogeneous a population, the larger the sample size - 2. Measuring a large number of variables requires a larger sample - 3. 50 cases for each population

Answer 9

- Avoids researchers’ conscious or unconscious biases in element selection - If all elements in population have an equal (or unequal and subsequently weighted) chance of selection, there is an excellent chance that the sample so selected will closely represent population of all elements - Permits estimates of sampling error - Although no probability sample will be perfectly representative in all respects, controlled selection methods permit the researcher to estimate the degree of expected error

Answer 10

- 1. Simple Random - Taking the last two digits of every set of numbers - 2. Systematic - Sampling Frame = 30,000 elements - Sample Size = 3,000 elements - Sampling Interval = 30,000/3,000 = 10 - Sampling Ratio = 30,000/10,000 = 1/10 - Taking every 10th on list? - 3. Stratified - Proportionate and Disproportionate to Size - Take 1, 4, 7, 10, 13, 16 from magazine set from 1960s, from 1980s, and from 2000s - Stratified: Systematic Random - Take from separate set of magazines, but instead of taking 1, 4, and 7 from each, you just number all magazines together and continue counting pattern - 4. Multi-Stage Cluster - We use this sampling strategy when a sampling frame does not exist - Step 1: List of all first-level clusters - Step 2: Sample from first-level clusters - Step 3: List of all Level 2 clusters - Step 4: Sample from Level 2 clusters - Sample of all homeless people in Las Vegas in 2019 - Sample of all the patients who were hospitalized in Vegas hospitals in 2018 - Sample of all Vegas high school seniors who graduated in May 2019 -Cluster (review questions): select two random groups and collect data on all of them

Answer 11

- That quality of a sample of having the same distribution of characteristics as the population from which it was selected. By implication, descriptions and explanations derived from an analysis of the sample may be assumed to represent similar ones in the population. Representativeness is enhanced by probability sampling and provides for generalizability the use of inferential statistics - Representative when it produces the characteristics existing in the population from which it is drawn

Answer 12

- It answers the following questions: - Is my sample representative? - Can I generalize from my sample to the population from which it is drawn? - How close is my sample statistic to the population parameter? - How confident am I that the statistics (variables) I measure in the sample represent the parameters existing in the population from which it is drawn? - In order to be able to generalize from the sample to the population from which it is drawn, the sample must be representative - It must reproduce the characteristics existing in the population from which it is drawn - The key to create a probability sample is to select its elements through a process of random selection - Every element of the sampling frame has an equal non-zero chance of being selected for the sample

Answer 13

- Selection of observations is not random | - Any technique in which samples are selected in some way not suggested by probability theory

Answer 14

- 1. Accidental/Convenience: - The researcher collects observations in terms of convenience, accessibility, etc. - Ex: malls, airports, etc. - Ex: FB experiment - We Have to Take These Guys Out 2. Purposive/Judgemental - Researcher collects observation on the basis of his/her knowledge of the phenomenon and objective - Ex: Sampling members in ethnography, interview research, sampling texts in historical research and textual analysis - Ex: We Have to Take These Guys Out (focus on men) 3. Snowball - The respondents suggest who/what else to sample 4. Quota - The researcher collects a representative number of observations, but the process is not necessarily random (representative, but not random) - Ex: of multiplying liberal arts by F by work (65 x 40% x 30% = 8) -5. Theoretical Sampling

Answer 15

- Avantages: - Good for qualitative research projects - Get inside look from informants - Good for times when probability sampling wouldn’t be appropriate even if it were possible - Disadvantages: - Informants’ marginality bias - Marginal status may also limit their access (and hence yours) to different sectors of the community you wish to study - Simply because they’re the ones willing to work with outside investigators, informants will almost always be be somewhat “marginal” or atypical within their group - Limitations: - Not accurate - Not precise representations of populations

Answer 16

- Theoretical construct - Americans, the middle class, UNLV students, TV commercials, blogs, etc. -Even though we ultimately want to make claims about the population, it cannot be studied

Answer 17

- Specifications about who/what will be studied - Ex: Bostonians: Boston residents aged 18-54, who were American citizens at the time of the study, and possessing English language proficiency - Ex: Commercials: All commercials aired between September and December 2019 on ABC, CBS, NBC, FOX channels between 12 am and 11 pm - Defining the target population typically entails deciding on criteria for membership (age, time, place) - Ex: UNLV Students: All individuals registered as students and attending classes on the UNLV main campus and online during the 2019 Fall semester

Answer 18

- Concrete representation of the target population - List of people who meet criteria - ON EXAM

Answer 19

-Subset of the sampling frame

Answer 20

-Summary measure of a variable in the population - Ex: Age, income, GPA, gender, vote - Cannot be known

Answer 21

-Summary measure of a variable in the sample - Ex: Age, income, GPA, gender, vote - Can be known

Answer 22

-Unit of the sample about whom/which we collect observations

Answer 23

- 91 Percent of UNLV Professors, but it only included 125 professors from mostly from Liberal Arts - Not representative of all professors -”UNLV Professors” -An overwhelming majority of UNLV professors are registered Democrats, according to a recent study, leading some students to believe that their classroom experience is lacking in intellectual diversity. - 91 percent of UNLV professors studied affiliate themselves with parties on the left of the political spectrum. - Study also found that some departments at UNLV don't have a single registered Republican teaching courses - Roots solution: The solution is to do away with tenure, which he believes obstructs the free market

Answer 24

- It is a non-probability sample - Accidental/Convenience -Consisted of 86% women, geographically residing throughout the country - IV = intensity of FB use - DV = life satisfaction; emotions - Classical Experiment - We hypothesize that X (IV) will cause a change in Y (DV 1) and Z (DV 2) - X: IV = stop using FB for a week - Y: DV 1 = Emotional well-being - Z: DV 2 = Life Satisfaction

Answer 25

- It is a non-probability sample - Accidental/Convenience - Purposive/Judgemental (focus on men) -Interviewed 73 men and women previously incarcerated in California prisons

Answer 26

- Step 1: Pretest both groups on variables - Step 2: Administer treatment to experimental group - Step 3: Post-test both groups on same variables and compare - Effect of treatment? There should be no difference in control group if done correctly, but there should be a difference in experimental group if treatment does have effect - When to use it: - To test hypotheses (deductive) - About causality - Among a small number of variables - 1. Two groups (experimental and control) - 2. A pre-test of variable(s) in both groups - 3. A stimulus (treatment) only in experimental group - 4. A post-test of variable(s) in both groups - 1. Does X cause change in the variable (O1 and O2)? - 2. What are x’s effects on the variable? - 3. Is it true that x will cause a change in the variable? EG O1 x O2 CG O1 O2 - Types: - Classical Design - Pre-Experimental Designs - Field Experiments - Natural Experiments - Evaluation Research - Online Experiments

Answer 27

- 1. Two groups (experimental and control) - 2. A pre-test of variable(s) in both groups - 3. A stimulus (treatment) only in experimental group - 4. A post-test of variable(s) in both groups - 1. Does X cause change in the variable (O1 and O2)? - 2. What are x’s effects on the variable? - 3. Is it true that x will cause a change in the variable? ``` EG O1(pretest) x(stimulus) O2(post-test) CG O1 O2 ``` - Ex: We hypothesize that X (IV) will cause a change in Y (DV 1) and Z (DV 2) - X: IV = stop using FB for a week - Y: DV 1 = Emotional well-being - Z: DV 2 = Life Satisfaction -Remember: conceptualize and operationalize both dependent and independent variable

Answer 28

-Experimental designs that lack pretests and/or control groups -1. Static Groups Comparison -Experimental and control group -No pretest EG x O2 CG O2 -Ex: Bobo Doll Experiment -2. One Group Pre-Test Post-Test Only: -Pre-test, post-test -No control group EG → O1 (number of tickets before) x(stimulus) O2(number of tickets after) -Ex: Milgram 3. One Shot Case Study -No pre-test -No control group EG x O2 -Ex: Milgram, Rosenhan, and Rosenthal & Jacobson

Answer 29

-1. Random Selection - 2. Matching - Single Men, Under 30 → 4 into control group, 4 into experimental group - Note: - 1. The two groups must be representative of each other, not of the general pop - 2. Sometimes, we do not know what are the variables according to which we should match both groups

Answer 30

-1. Additional Experimental Groups EG O1 x O2 EG2 O1 2x O2 CG O1 O2 -2. Change the IV EG O1 x O2 EG 2 O1 2x O2 CG O1 O2 3. Post-Test on Multiple Occasions EG O1 x O2 O3 O4 EG2 O1 2x O2 O3 O4 CG O1 O2 O3 O4

Answer 31

- Neither the observers nor the subjects know which is the control or experimental group - Compensation - Experimenter’s expectations - Use it to prevent compensation and experimenter’s expectations

Answer 32

-Are the changes observed in the DV (from pre-test to post-test) caused by the IV (x: the stimulus) or by other factors? Sources: - 1. Design: - 1. Testing (person may be alarmed that they’re being tested) - 2. Instrumentation - 3. Selection Bias (not careful in randomly selecting) - 4. Compensation (feel bad for CG and give them more attention) - 5. Causal Order - 6. Pygmalion Effect/Experimenter’s expectations (Hans horse: Hans saw farmer’s reaction; understand what experimenters want to see) - 2. External - 1. History (ex: measuring effect of movie, but day before watching movie, environmental catastrophe happened, so results are influenced by history) - 3. Subjects - 1. Maturation (people change/evolve) - 2. Attrition (some subjects leave -- those who stayed vs. left have certain characteristics) - 3. Contamination/Diffusion (EG person goes to bathroom and meets someone in CG and they talk, which changes behavior) - Compensatory Rivalry (members of CG work extra hard to go against researcher) - Demoralization (CG upset they’re not getting any rewards) - 4. Statistical Regression - 5. Hawthorne Effect (subjects feel special because they’re part of an experiment)

Answer 33

-Refers to the possibility that conclusions drawn from experimental results may not be generalizable to the “real” world - Sources: - Generalizability of experimental findings is jeopardized if there’s an interaction between the testing situation and the experimental stimulus (ex: muslim history film with classic experimental design) - Solomon Four-Group Design (solution) - Posttest-Only Control Group Design (solution) - Are the changes observed in the dependent variable (from O1 to O2) caused by - the simulus (x) or - the interactive effect of pre-testing and stimulus (O1 and x)?

Answer 34

- Addresses the problem of testing interaction with the stimulus - Involves 4 groups of subjects, assigned randomly from a pool - Classical experiment runs the risk that pretesting will have an effect on subjects, so the Solomon four-group design adds experimental and control groups that skip the pretest. Thus, it combines the classical experiment and the after-only design (with no pretest) - Benefits: - Rules out interactions between testing and the stimulus - Provides data for comparisons that will reveal how much of this interaction has occurred in a classical experiment (allows researchers to review and evaluate the value of any prior research that used the simpler design) EG1 O1 x O2 CG2 O1 O2 (from EG1 to CG2, 30%) EG3 x O2 CG4 O1 O2 (from EG3 to CG4, 10%)

Answer 35

- Situations where a stimulus is introduced in society or a group thereof, and which experimenters do not control - Natural catastrophes - Presidential debates - New laws - Terrorist acts, shootings, riots - Technological accidents - Information manipulation - May raise validity problems - Still requires ingenuity and insight - Can still have control and experimental groups - Workers in plant that had problem vs. workers in plant that didn’t have problem

Answer 36

- Experiment conducted outside the lab, in real life, in real world - Origins - Characteristics - No precise pre-test - No precise post-test - No control group - Pygmalion; My Fair Lady - tendency to see in others what we’ve been led to expect - Rosenthal and Jacobson experiment (telling teachers which students are promising → teachers thus give them more attention → of course, those students improved

Answer 37

-Research undertaken for the purpose of determining the impact of some social intervention, such as a program aimed at solving a social problem; purpose is to evaluate the impact of social interventions, such as new teaching methods or innovations in parole - Characteristics: - Main operational aspects: measurement, study design, and execution - Form of applied research (that is, it’s intended to have some real-world effect) - Methods that can be used: surveys, experiments, and so on - Appropriate whenever some social intervention occurs or is planned - Must be able to operationalize, observe, and recognize the presence or absence of what is under study - Not a method itself, but rather one application of social research methods! - Types: - Experimental Designs - Quasi-Experimental Designs - Qualitative Evaluations - Dependent variable (O1) = a condition judged undesirable - Stimulus (x) = intervention - Post-test (O2) = change in undesirable conditions - Ex: - Head-Start - Scared Straight - D.A.R.E. - Police Organization Research - PTSD Training

Answer 38

- In labs - Virtual Spaces - Social Media - Ex: FB experiment

Answer 39

- Strong validity - Weak reliability - Would the stimulus reliably produce the same effects outside of the lab? - Milgram’s research in New Haven office

Answer 40

-In a typical survey, the researcher selects a sample of respondents and administers a standardized questionnaire to them Uses: -To study large number of dispersed individuals; best method available to social researcher who is interested in collecting original data for describing a population too large to observe directly - Descriptive, exploratory, explanatory - Large number of variables - Inductive and deductive - Strengths: - Multi-Purpose: - It can measure a wide variety of issues and provide information for a wide variety of interests: - Sociological issues - Political Voting - Marketing - Health - Transportation - Education - Government assessing community and state needs - State agencies assessing programs - Legal Claims - Media - Efficient: It can - Collect large number of information - About large numbers of people - At relatively low costs - Cost of measuring many variables is not that much higher than measuring a few - Quickly - It guarantees the standardization of measurement - Large number of dispersed people have been measured/observed in the same way

Answer 41

-Large number of dispersed people have been measured/observed in the same way

Answer 42

- (1) Overall organization - Maintain constant focus - (a) Clear idea of research questions - (b) Keep population being measured in mind - (c) Integrated whole - (2 and 3) Questions and Sections - (A) Group questions into logical sections (ex: social life, information access, religiosity, attitudes, demographics -- each section has 4 questions) - a. Each section serves a purpose and is related to the research question - b. Each question within each section serves a purpose and is related to the research question - (B) Organize the order of the questions (ex. Asking about happiness in general and then marital life vs. vise versa) - a. Interesting sections first - b. Uncomfortable sections middle or last - c. Demographic section last - d. Introductions to and transitions between sections - e. Concluding statement (Thank you, etc.) - (C) 14 Rules - (4) Aesthetics - (5) Cover Letter

Answer 43

-Design → Pre-Delivery → Delivery → Post-Delivery - 1. Overall organization - 2. Questions - 3. Sections - 4. Aesthetics - 5. Cover Letter

Answer 44

- Questionnaire: Tool that measured people at distance - Sole medium of communication between researcher and respondent - (1) Overall organization - Maintain constant focus - (a) Clear idea of research questions - (b) Keep population being measured in mind - (c) Integrated whole -(2 and 3) Questions and Sections -(A) Group questions into logical sections (ex: social life, information access, religiosity, attitudes, demographics -- each section has 4 questions) Each section serves a purpose and is related to the research question Each question within each section serves a purpose and is related to the research question -(B) Organize the order of the questions (ex. Asking about happiness in general and then marital life vs. vise versa) -a. Interesting sections first -b. Uncomfortable sections middle or last -c. Demographic section last -d. Introductions to and transitions between sections -e. Concluding statement (Thank you, etc.) -(C) 14 Rules Questions: A. We ask survey questions to many people, not just one B. Our survey questions/attributes must be understood in the same way by all respondents, who may be very different from each other C. Our survey questions/attributes must be understood in the same way by all respondents, who may be very different from us D. When we deliver our questionnaire by mail or online, we can’t rephrase a question in the respondent does not understand E. When we deliver our questionnaire face-to-face or by phone, respondents do not know us and cannot catch nuances of expression - (4) Aesthetics - (5) Cover Letter

Answer 45

- 1. Provide generous space - 2. Organize different sections on different pages - 3. Distinguish questions and answers - Indents - Numbers - Fonts (different fronts for questions and answers) - Ex: Over the past two weeks, I have noticed I have more difficulty falling asleep (1) Yes (2) Not Sure (3) No - Answers are indented; answers have numbers; answers are in a different font - 4. Provide visual aid for contingency questions - Ex: Do you smoke marijuana? - a. Yes - b. No. If no, go to question 6 (include arrow to number 6) - 5. Provide clear instructions - Answering - Returning/submitting the questionnaire - Self-addressed stamped envelope

Answer 46

- 1. Correct grammar is essential - 2. Avoid jargon - 3. Questions should be precise and short - 4. Avoid vagueness about attitudes, beliefs, situations, time periods - 5. Avoid questions that fail to make a meaningful difference - 6. Ask single-issued questions - Ex: (bad) How often have you used the following sources for UNLV information such as courses, events, requirements, and the like? (Circle one number for each row) - 7. Avoid double-negatives - Ex: Most of the time, parking is NOT a problem… - Ex; I usually do NOT REFUSE requests that are NOT made at the last minute -8. Avoid questions about future intentions - 9. Ask questions that measure what we intend to: - Characteristics - Orientations - Actions - 10. Avoid biases - a. Loaded language (ex: Should the US intervene in the world for a vietnam-like situation? Should the US intervene in the world to stop a communist takeover? - b. Leading questions (ex: Don’t you agree…. Most people agree…); ask neutral questions - c. Restricted response range (ex: three positive answers and only one negative one) - 11. Mind the fence-sitters - When forced to make a choice, 10-20% of respondents who do not have a strong opinion will choose an explicit position - 12. Mind the floaters: - 30% of respondents who know nothing about a topic will choose an explicit answer in the absence of a “don’t know” option - 13. Add interpretive questions to make sure we know - What our respondents mean - They understand the question - Ex: Do you think your emotional state affected your driving on the day of the violation? Yes (explain). No. Interpretive Question: What else was going on on the day of the violation? ___________ - 14. Induce truthful answers - People should want to answer truthfully - (a) Frame deviant behavior as unexceptional - (b) Frame problematic behavior as benign - (c) Ask the same question differently, in different sections - Respondents are likely to overreport - Social desirable behaviors - Being informed - Fulfilling moral responsibilities - Having a happy life - Respondents are likely to underreport - Socially undesirable behaviors - Being uninformed - Failing to fulfill moral responsibilities - Having an unhappy life

Answer 47

- Loaded language - Ex: Should the US intervene in the world for a Vietnam-like situation? 18% say yes - Ex: the US intervene in the world to stop a communist takeover? 33% say yes - Leading questions - Don’t you agree… - Most people agree… - Restricted response range - Three positive answers, one negative

Answer 48

- 14. Induce truthful answers - People should want to answer truthfully - (a) Frame deviant behavior as unexceptional - (b) Frame problematic behavior as benign - (c) Ask the same question differently, in different sections - Respondents are likely to overreport - Social desirable behaviors - Being informed - Fulfilling moral responsibilities - Having a happy life - Respondents are likely to underreport - Socially undesirable behaviors - Being uninformed - Failing to fulfill moral responsibilities - Having an unhappy life

Answer 49

-The measurement of a DV among the subjects; subjects are measured in terms of a dependent variable; before being exposed to IV

Answer 50

Types: - 1. Preliminary Pretests - Friends and colleagues - Ex: Nadelson article sharing instrument with researchers - 2. Formal Pretests - Individuals similar to respondents - Ex: Nadelson article administering it to 75 undergraduate college -Preliminary Pretest → Formal Pretest → Delivery to Respondents → Follow-Ups

Answer 51

- Look for: - Skipped questions - Questions answered similarly by all - Questions that weakly correlate with others - Response sets - Use reverse-coded items (used in Nadelson article

Answer 52

- Credible - 17% increased response rate for government and university - Personalized - 7% increased response rate - Interesting - Voluntary Participation - Confidential or Anonymous - Identify Yourself and Research Purpose - Name, affiliation, contact info - Appeal - Science, community, society, respondents -Incentive

Answer 53

- Mail - Face-to-Face - Phone - Online

Answer 54

- Advantages: - Quick - Cheap - Sensitive Questions - Large numbers - No researcher’s effects - Disadvantages: - No control over sequence - No control over identity of respondent - Low return rate (50%) - No probing or explaining - No complex questions - Least researcher’s effect - Not expensive - Most appropriate for sensitive questions - Low response rate (50%) - Least control over the order of questions answered and the identity of the respondent - Standardization of measurement?

Answer 55

- Advantages: - Immediate entry of answers - Least skipped questions - Allows for probing - Allows for complex questions - Highest return rate (80%) - Allows for recording of additional information - Disadvantages: - Strange situation - No sensitive questions - Neutrality (tape-recorder) - Warm technician - Researcher’s effect - Most expensive - Requires training and traveling - Remember: - Make contact (letter, phone) - Non-threatening questions first - Has highest response rate (80%) - Is most appropriate for complex questions - Is most expensive - Is the least reliable one (researcher’s effect) - Tool guarantees the standardization of measurement

Answer 56

- Advantages: - Quick - Cheap - Some probing - Some sensitive questions - Immediate entry into database - Few skipped questions - Disadvantages: - 5% of population don’t have a phone - 9% response rate - Increasing number of screening devices - Telemarketing/Suspicious Calls - Time and representativeness issues - Distractions - Training -This mode of delivery is meeting most research criteria, but has the lowest response rate

Answer 57

- Advantages: - Cheap(est) - Quick(est) - Control over sequence - Control over timing - Immediate data entry and analysis - Hybrid (chat) - Disadvantages: - Low return rate (30%) - Representativeness - Distraction - Fatigue - New context, new risks - Suffers from the same weaknesses as mail and has an even lower response rate - Mail/online modes of delivery present the greatest risk of misinterpretations

Answer 58

-Phone mode of delivery is meeting most research criteria, but has the lowest response rate - Face-to-face mode of delivery: - Has highest response rate (80%) - Is most appropriate for complex questions - Is most expensive - Is the least reliable one (researcher’s effect) - The tool guarantees the standardization of measurement - Mail mode of delivery is: - Least researcher’s effect - Not expensive - Most appropriate for sensitive questions - Low response rate (50%) - Least control over the order of questions answered and the identity of the respondent - Standardization of measurement - Online mode of delivery suffers from the same weaknesses as mail and has an even lower response rate - Mail/online modes of delivery present the greatest risk of misinterpretations

Answer 59

-Post-delivery - First one: - 2-3 weeks after first mailing - Send letter and questionnaire - Second one: - 2-3 weeks after first follow-up - Diminishing return with each follow-up - Why: try to get more responses in; to remind them to complete it; try to get as many responses back as possible

Answer 60

- 1. Sponsoring organization - 2. Date on which survey was conducted - 3. Population that the sample represents - 4. Sample frame - 5. Sampling method - 6. Sample size - 7. Method of delivery - 8. Exact working - 9. Response rate - 10. Missing information - Don’t need to know?

Answer 61

- Strengths: - Strong reliability (but weak validity) - Relatively cheap - Produces results relatively quickly - Can be used to measure a wide variety of issues - Weaknesses: - 1. Communication patterns - 2. Words are context-dependent - 3. Context of response - 4. Answers created by/because of survey - 5. Respondents’ point of view - 6. Researcher’s non-intervention - 7. Assumption of correspondence - Tool guarantees standardization of measurement - a. All respondents understand questions similarly - b. All respondents understand answers similarly - c. All respondents understand both questions and answers as the researcher - 8. Validity (questions of validity - Am I really measuring what I am measuring?) - Cross-cultural (ex: of culture in Beirut)

Answer 62

- Utilizing already existing statistics to analyze them in new ways - Cross-sectional research - Longitudinal research - Cross-national research

Answer 63

- Advantages: - Cost (doesn’t cost anything) and time (doesn’t take a lot of time) - Sometimes, provides access to high quality data - More time for data analysis - May offer new insights - Disadvantages: - No control over data quality or presence of key variables that interest the user

Answer 64

- Assess unexamined relationships between different variables - Cross-sectional research - Longitudinal research - Cross-cultural research? - Analyze them in new ways - SDA-UC BERKELEY

Answer 65

- Strengths: - Cost and time - High quality data - More time for data analysis - Reanalysis may offer new interpretations - Wide obligations of social researcher - Making full use of respondents' contributions - Mining data to full extent - Weaknesses: - Lack of familiarity with data/variables - Complexity of the data - No control over data quality or presence of key variables that interest the user

Answer 66

-Research methods in which there is no interaction between the researcher and the data - Purposes: - Exploratory - Descriptive - Explanatory - Criticism - Units: - People (alone or in interaction) - Artifacts/Texts - 1. Audio (all) - 2. Visual/Pictorial (images, engravings, postcards, websites) - 3. Audiovisual - 4. Written (books, newspapers, diaries, speeches, anything written) - 5. Concrete Objects - 6. Traces - Erosion (signs of wear and tear on objects) - Ex: World soil degradation - Accretion (accumulation of objects; what people leave behind) - Ex: Graffiti; pile of cigarettes; categories of garbage we leave behind - 7. Built Spaces - Ex: Airports; churches - Focus: - Characteristics - Orientations - Effects - Time Dimension: - Cross-Sectional (past or present) - Longitudinal (trend, cohort, panel) - Sampling: - Probability - Non-Probability - 1 Single Text - Types: - Quantitative: Manifest Content Analysis (Deductive or Inductive) - Qualitative: Latent Content Analysis (Inductive) - Can be quantitative or qualitative - Can be deductive or inductive -Ex: Lee’s article “Mass Shooting and Media Contagion Theory”?

Answer 67

- Technique for the objective, systematic, and quantitative description of the manifest/overt content of a communication - ”Objective” = not subjective - ”Systematic” = - ”Quantitative” = just numbers - ”Manifest/Over Content” = systematically analyze; not about meaning

Answer 68

- Units: - People (alone or in interaction) - Artifacts/Texts - 1. Audio (all) - 2. Visual/Pictorial (images, engravings, postcards, websites) - 3. Audiovisual - 4. Written (books, newspapers, diaries, speeches, anything written) - 5. Concrete Objects - 6. Traces - Erosion (signs of wear and tear on objects) - Ex: World soil degradation - Accretion (accumulation of objects; what people leave behind) - Ex: Graffiti; pile of cigarettes; categories of garbage we leave behind - 7. Built Spaces - Ex: Airports; churches

Answer 69

- Manifest: - Quantitative - Deductive or inductive - Latent: - Qualitative - Inductive

Answer 70

- Sampling: - Probability - Non-Probability - 1 Single Text

Answer 71

- Nominal: - Is a particular word/theme/image/metaphor quality present or absent? - Does a text support or oppose? - Ordinal - Intensity of word/theme/image/metaphor/quality - Add weight to the word (ex: love = strong positive; categories by level of intensity - Interval - Quantification of intensity (add number value) - Quantification of complexity of words/texts/sounds - Ex: love = strong positive = 6 - Ratio: - Frequency - Space and time - Always record base number - Ex: 40 out of how many? - Just reporting, not interpreting

Answer 72

- Strengths: - 1. Cheapest - Can be done at home; can download software (that may cost money) to code information - 2. Democratization of what counts as legitimate “texts” - 3. Allows for endless improvement - 4. Universe of potential data - Has strong validity AND strong reliability - Weaknesses: - 1. Meaning - Frequency does not equal significance - 2. Indexicality of meaning - Meaning of word depends on context - 3. Relevance: Reader’s interpretations

Answer 73

- (E) C and D - When it yields the same results on a number of different occasions - When it yields the same regardless of who is measuring

Answer 74

- (D) All of the above - 1. Providing a working definition of the concept - 2. Developing the various dimensions of the concept - 3. Providing the indicators of the concept

Answer 75

-(A) Questions should measure only one issue

Answer 76

- (D) How frequently do you smoke marijuana? Check one - Daily - Weekly - Monthly - A few times a year - I never smoke marijuana

Answer 77

-(A) A Parameter

Answer 78

-(A) They help us assess how close our statistics are to the population parameters

Answer 79

-(D) Its elements must be randomly selected

Answer 80

-(E) C and D - Quota sample - Need to select certain number from each group to participate - Purposive sample - Purpose (ex: of TA’s study, which used snowball)

Answer 81

-(A) Las Vegans

Answer 82

-(D) A list of all registered voters in Las Vegas

Answer 83

- (B) 2,000 respondents - Population → Sampling Frame → Sample → Statistic? → Element - Las Vegans → All registered voters in Las Vegas → 2,000 respondents → A single voter - Sample → Sample Size → Usually means a number

Answer 84

- (D) Stratified Sample | - 30 years of movies → All top-winning movies → Randomly selected

Answer 85

-(E) Convenience Sample

Answer 86

-(B) History

Answer 87

-(C) A double-blind experimental design

Answer 88

- (D) A field experiment - Field → natural environment, but researcher is adding stimuli - Natural → natural environment, and stimulus occurs regardless of researcher

Answer 89

-(C) One group pretest-posttest only

Answer 90

-(B) A natural experiment

Answer 91

-(C) Mailed questionnaire

Answer 92

-(A) When the questionnaire contains complex questions

Answer 93

-(A) Face-to-face

Answer 94

- (E) All of the above - It is relatively cheap - It is strong on reliability - It produces results relatively quickly - It can be used to measure a wide variety of issues

Answer 95

-(B) It allows for constant fine-tuning of the measuring instrument

Test 2 Flashcards

(123 cards)