Reading...
Play button
Play button
Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
80 Cards in this Set
- Front
- Back
|
Advantages of survey research
|
versatility- can be associate with multiple types of populations, efficiency- can be conducted quickly and with minimal post-collection effort, generalizability- allows for multiple populations, allows for more accurate image of the population studied
|
|
|
research methodology focused on responses to questions
|
typically utilizes fairly large numbers of subjects, Questions can be focused on self or others
|
|
|
Split-Ballot Design
|
Experimentation within a survey setting; Allows for greater variety of questions; Can be used for testing effect of question differences; Provides greater coverage at lower cost for researchers
|
|
|
Concerns raised with survey research
|
poor measurement, nonresponse, inadequate coverage, sampling error
|
|
|
poor measurement
|
Questions that are unclear, Difficult questions receive poor answers, Respondents “satisfice”
Techniques to minimize poor measurement Know your population, Use simple-ish language, Proper English isn’t necessary best, Organize questionnaire thematically |
|
|
nonresponse
|
People not participating in surveys, People not answering questions when participating, Costs of participating outweigh perceived benefits
|
|
|
inadequate coverage
|
Sampling problems, Unrepresentative samples don’t reflect reality,
Generalizability of study significantly diminished |
|
|
sampling error
|
Proper sampling makes a study→Improper sampling sinks a study
|
|
|
double-negatives
|
Questions with two negative clauses, Double negatives are confusing for respondents
|
|
|
double-barreled questions
|
One physical question with two or more questions within it
Impossible to determine what question is being answered |
|
|
Skip Pattern
|
Linked to filter and/or contingent questions; Specific pattern of question(s) asked of a specific respondent; Questions asked or not asked based on previous question responses; Allows for more detailed responses of appropriate respondents; Also allows for more efficient interviews
|
|
|
fence-sitters
|
people who are neutral on many of the topics within a surveyVast majority of respondents do not fall into this category; More or less problematic based on topic of survey
The more esoteric the topic, the more likely to find fence-sitters |
|
|
forced-choice questions
|
Pre-defined answers to a question; Used when a lot of information is known about a topic; Allows for rapid administration of the survey; Minimizes amount of respondent variation
|
|
|
floaters
|
Like fence-sitters when given the opportunity; Folks choosing a neutral option when provided
|
|
|
part-whole effects
|
Questions influenced by earlier questions; General questions influenced by earlier, specific questions
Starting with questions about domestic violence activities within the home Moving later onto questions about marital satisfaction DV questions will almost certainly influence responses about marital satisfaction |
|
|
considerations for translation
|
reliability, completeness, accuracy, cultural appropriateness, equivalentness
|
|
|
reliability
|
Regardless of different translations, text must be equivalent across all
All translations need to be asking the same questions |
|
|
completeness
|
Need to make sure all versions contain precisely the same amount of information
Any unintended additions/subtractions introduce systematic bias into the data |
|
|
accuracy
|
No grammatical or spelling errors
Standard efforts to ensure professional presentation of material |
|
|
cultural appropriateness
|
Different populations have different standards of acceptability
|
|
|
equivalentness
|
Assures that everything remains as equal as possible across versions
Appropriate modifications for cultural sensitivities |
|
|
total design method
|
Social Exchange Theory guiding survey administration
Brief letter alerting respondents to forthcoming survey Quality questionnaire, quality paper, sufficiently personalized, etc. Reminder post card ~2 weeks following initial mailing Replacement questionnaire ~4 weeks out Final replacement ~6-8 weeks using different USPS delivery method (e.g. priority or certified) |
|
|
CATI
|
Surveys read off of a computer screen; Computer programming handles skips and most data recording
Limits mistakes; More efficient, data not recorded multiple times |
|
|
CAPI
|
Mixture of CATI and in-person interviewing
Some questions asked & recorded by interviewer Some questions asked via recording, recorded by respondent Costly as in-person interviews Limits data entry errors and replication Allows for accurate responses to sensitive questions |
|
|
Interactive Voice Response (IVR)
|
Surveys still completed with computer assistance; No live interviewers; Allows for asking deeply personal questions more accurately; Some people don’t like “talking to” a computer; Difficult to ask follow-up or clarifying questions
|
|
|
Mixed-Mode Surveys
|
Survey using multiple methodologies; Strengths of multiple methods to secure completion
A bit more money up-front (configuring multiple administrations); Better completion rates |
|
|
Confidentiality
|
Respondents assume confidentiality as a condition of participation
Identifying information removed from data publically available: Name, Address, Height, Etc. Condition for open and honest answering of questions Prevents tracking someone down/ using their answers against them |
|
|
Anonymity
|
More stringent standard than confidentiality→No identifying information is collected at all
Typical with highly sensitive survey topics: Sexual activities/practices, Drug Use, Illegal Activities, Etc. Prevents any possible follow-up surveys; Might be necessary given research questions |
|
|
Idiosyncratic Variation
|
Variation in responses to questions that is caused by individuals’ reactions to particular words or ideas in the question instead of by variation in the concept that the question is intended to measure.
|
|
|
Cognitive Interviewing
|
A technique of diving deeper into question meaning
People read questions and then are asked about what it means to them Allows researchers to dive deeper into what questions mean to potential subjects Serves as a way of establishing validity for a given population |
|
|
Context Effects
|
Occurs in a survey when one or more questions influence how subsequent questions are interpreted.
|
|
|
Filter Questions
|
Question used to determine if follow-up questions should be asked; Allows for proper follow-up/expansion of a topic; For example: Questions about marital satisfaction
Requires that respondents are married If respondents are married, additional questions are asked If respondents are not married, these questions do not make any sense to ask |
|
|
In-Person Interview
|
Surveys completed face-to-face; Allows for extensive probing/clarification of answers
Extremely costly; Sensitive questions difficult to ask |
|
|
Omnibus Survey
|
-covers a range of topics of interest to different social scientists, in contrast to the typical survey that is directed at a specific research question.
A survey that covers a range of topics that are of interest to different social scientists. |
|
|
Participant Observation
|
Gathering data by actively engaging subjects; Requires mixture of watching and action; Requires significant amounts of time and effort; Usually done with an open mind; Requires informants to help comprehend subject’s perceptions/impressions
|
|
|
Intensive (In-depth) Interviewing
|
Open-ended questioning of respondents; Basic ideas/questions start process; No set structure of interview; more organic; Time intensive
Interview transcriptions 1.5-2 hrs. per hour of interview time Requires entre to the population of interest |
|
|
Focus Group
|
Unstructured group interviews; Discussion among interviewees encouraged; Yield more info than 1-on-1 interviews
|
|
|
Case Study
|
In-depth analysis of a small organization or individual’s experience→Holistic look at case
Deep and wide examination of case; Can be taken as indicative of others Common technique in business and/or law |
|
|
Covert Observer
|
Participant observation where no discussion of the researcher’s goals and objectives
Researcher’s presence blends into the background & watches All data collection are from the researcher’s perspective Ethical considerations if not conducted in completely public settings |
|
|
Complete Observer
|
Researcher’s presence is known to subjects
Researcher simply watches, does not actively engage in activities of the subjects; Full disclosure, limited ethical risk |
|
|
Reactive Effects
|
Possibility of changes because of researcher
Challenge in all research efforts where researcher alerts subjects to presence “Hawthorne Effect” a possibility→Generally fades over time |
|
|
Jottings
|
Notes in the “heat of the moment”→Very superficial accounts, Only key words and phrases→Jog your memory
Typically recorded in private: Bathroom, Around the corner, In the car |
|
|
Field Notes
|
Detailed description of events→The greater the detail the better, “Blow-by-blow” account of observations
Serve as the foundation of data for later analysis Ideally, written as soon as possible following leaving the field→Written using one’s jottings |
|
|
Gatekeeper
|
Person(s) controlling access to research site; Requires their buy-in; One of the hardest parts of qualitative research
|
|
|
Key Informant
|
Provides important information about setting(Mentor of sorts)
Most observational work done from subject’s perspective Requires researcher as naïve needing insider knowledge |
|
|
Snowball Sampling
|
One person and building out; Focused on similar subjects; Typically efficient sampling technique
|
|
|
Theoretical Sampling
|
Sampling based upon emerging needs;People or settings selected as fitting or counterfactual to current themes
Cases selected to bolster or test validity of data |
|
|
Systematic Observation
|
Set observational days and times; Use of standardized coding schemes
Typical of teams of researcher not individuals |
|
|
Saturation Point
|
No new or useful information is obtained, Stopping point of data collection
|
|
|
Confidentiality
|
Subjects should not be identifiable; Use fake names; Use fake locations; Change significant details
Especially important for illegal or illicit research settings |
|
|
Thick Description
|
A rich description that conveys a sense of what it is like from the standpoint of the natural actors in that setting.
|
|
|
Participant Observer
|
Most involved role for researcher; Combines explicit observation, active participation in subjects’ activities
Requires well-developed personal relationships; Requires substantial time commitment Raises ethical and perspective problems for researcher |
|
|
Experience Sampling Method ESM
|
A technique for drawing a representative sample of everyday activities, thoughts, and experiences. Participants carry around a pager and are beeped at random times over several days or weeks; on hearing the beep, participants complete the report designed by the researcher.
|
|
|
Descriptive Statistics
|
Preliminary Step→Basically just a tally of the variable
Useful because allows for checking the variable for problems: Missing Values, Miscoded Values, Etc. |
|
|
Data Cleaning
|
Resolution of any data irregularities
Coding/Recoding of variables for analysis, Construction of scales, indexes and latent variables This continues until all issues are resolved Arguably this is the hardest and most difficult stage of data analysis |
|
|
Univariate Statistics
|
Statistics involving only one variable
Three prominent univariate statistics: Central Tendency, Variability, Skewness |
|
|
Central Tendency
|
Basically, the most common aspect of the variable
Correct type depends on the level of measurement of the variable There are three major measures of central tendency (discussed in more detail later)→Mean, Median, Mode |
|
|
Variability
|
Basically, shows the amount of difference within the variable
Consider beer; How many different varieties of beer are there? |
|
|
Skewness
|
Refers to how the variable is clustered
Think about a “bell curve” and how it is perfect You can think of skewness in relation to the perfect bell curve When things deviate from perfection it is either positive or negatively skewed |
|
|
Bar Charts
|
VERY DIFFERENT from a histogram→The bars should not touch!
|
|
|
Histograms
|
Graph for numerical/quantitative variables
Distribution of the variable illustrated by bars touching; Touching bars indicate the continuity of the variable |
|
|
Base Number (N)
|
Total number of cases for a variable, From previous example, N = 1016
|
|
|
Range
|
Represents a measure of variation within a variable
Shows researcher amount of difference between highest and lowest values Problematic because can be swayed by extreme value(s) (outlier) |
|
|
Interquartile Range IQR
|
Another measure of variation; Quartiles = 4ths; breaking up distribution into 4 equal parts 25%, 50%, 75% & 100%
IQR = 75%-25% Beneficial because it uses middle of the distribution; Less influenced by outliers than Range |
|
|
Variance
|
A more precise measure of how much variability in a variable
Measure of how much each case differs from the average |
|
|
Standard Deviation
|
Another way of expressing variability of a variable
Provides a common way of thinking about the variability of a variable Commonality allows us to compare apples to apples across variables regardless of how they were originally measured |
|
|
Monotonic
|
Idea that two variables follow a consistent pattern
When values on one variable increase, the values on the other variable are consistently increasing or decreasing A form of linear relationship between two or more variables |
|
|
Curvilinear
|
Pattern between two variables that isn’t consistent
At first it will increase, hit a certain point and then turn downward |
|
|
Measures of Association
|
Descriptive statistics summarizing the strength of a relationship between two or more variables
Numerical value to ascertain the strength of the relationship Useful and necessary because looks can be deceiving |
|
|
Extraneous Variable
|
Variable that distorts actual relationship between DV and IV
Important to guard against these in research Ruling out alternative explanations (extraneous variables) improves your argument |
|
|
Correlation Coefficient
|
A summary that varies from 0 to 1 or -1, with a 0 indicating the absence of a linear relationship between the two variables and a 1 or -1 indicating that the relationship is completely described by the line representing the regression of the dependent variable on the independent variable.
|
|
|
Frequency Distribution
|
Numerical data presentation
Should show at least: number of cases and percentages Can, and frequently does, contain considerably more information |
|
|
Marginal Distribution
|
The summary distributions in the margins of a cross-tabulation that corresponds to the frequency distribution of the row variable and of the column variable.
|
|
|
Mean
|
average
|
|
|
Median
|
middle number
|
|
|
mode
|
probability average
|
|
|
outlier
|
exceptionally high/low
|
|
|
normal distribution
|
A symmetric, bell shaped distribution that results from a chance variation around a central value.
|
|
|
Unimodal
|
Variable distribution with only one mode to it
|
|
|
Bimodal
|
Variable distribution with two modes that are not adjacent to one another
|
