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
Quantitative Research
|
How much of something? Surface, many people
|
|
Qualitative Research
|
Why? Depth, less people
|
|
Three Qualitative Assumptions
|
Natural, Phenomenological, Interpretive
|
|
Natural
|
Context: everyday life circumstances affect interaction
|
|
Phenomenological
|
attempts to bracket (separate) their assumptions
|
|
Interpretative
|
Researcher and participants perspectives are woven together when reported (“giving voice”)
o NOT just the researcher, participants speak o Exemplar quotes: group findings into themes and look for best example(s) of the theme |
|
Participant
|
Full immersion, group members are unaware that researcher is an outsider
|
|
Participant-Observer
|
Consultant role, collecting information for a company, involved but also studying, group members are aware that researcher is an outsider
|
|
Observer-Participant
|
More removed than the last, participants know you are an outsider, not part of decision making, just sitting and observing
|
|
Interviewer
|
Talk and ask questions, not in lived experience
|
|
Ethnography
|
exploring the shared understandings necessary to be a member of a particular group (Full participant)
|
|
Critical Ethnography
|
exploring the shared understandings necessary to be a member of a particular group with a goal of reduction of and emancipation from oppression (very interested in power structures and social equality)
|
|
Auto-Ethonography
|
reporting on one’s own immersion and experiences within a group
|
|
Ethnomethodology
|
exploring the taken-for-granted, daily actions people use to socially construct their world (e.g. conversation analysis)
o I.e. what is a common day for a migrant farmer (framework/structures) |
|
Interviews
|
are in-person surveys where you can alter the content to meet researcher needs
• Most removed from participant involvement • Must be sensitive to setting (where to conduct?) • Must be aware of sensitive topic areas (funnel v. inverted funnel) |
|
Descriptive Questions
|
mini-tour: Tell me about your typical day…
|
|
Structural
|
How often do you call your spouse in a day?
|
|
Compare/Contrast
|
Which is more important to you A or B?
|
|
Focus groups
|
• Bringing people together usually 6-12 people
• Require a moderator, facilitator, or a conversation regulator • Must be aware of those who monopolize or are hesitant • Advantage: spurs thinking through others’ comments • Disadvantage: Must avoid “group-think” |
|
Grounded Theory
|
Glaser and Strauss 1967; it is a systematic methodology in the social sciences involving the discovery of theory through the analysis of data
|
|
Analytic Induction
|
looking for emerging patterns (meaning units)
|
|
Open Coding
|
(inductive): first read, passing, first attempt at looking for meaning units, group common concepts
|
|
Axiel Coding
|
(deductive): once themes are established look for specific examples to support main themes
|
|
Constant Comparative Analysis
|
thematic groupings
|
|
Validity Checks
|
Intercoder reliability, reader confirmation, cross-check with the participant
|
|
Quantitative Analysis Questions
|
• How do people typically behave?
• What is common (average)? • What is “typical” communication? • What do people “normally” experience? • What happens most of the time? |
|
Descriptive Statistics
|
summary characteristics of a data set
o Tell us about our data, lots of numbers (reduction) & reduce them o Deductive |
|
Inferential Statistics
|
used to generalize and make claims from the data set
o Inductive |
|
Central tendency
|
How do people tend to group together?
|
|
Dispersion
|
how spread out are peoples' thoughts from the middle
|
|
3 Main Summary Statistics
|
Mode, Median, and Mean
|
|
Mode (Mo)
|
the most frequent (most often) value in a data set
o Bi-modal: when two modes exist o Multimodal: when more than two modes exist |
|
Median (Mdn)
|
divides the sample exactly in half
|
|
Mean (M)
|
o Add values in data set and divide by total number of values
o Outliers: people on the far spectrum; sometimes are dropped to determine a more accurate mean. o M-Population often not possible b/c it requires a census |
|
Bell Shaped Curve
|
o Even distribution around mean
o Important for generalizing |
|
Peaked Curve
|
o Agree more, not much variation
|
|
Negative Skew
|
o Opposite of what it is—where the tail goes
o i.e. TV watching |
|
Positive Skew
|
o Opposite of what you would assume
o i.e. US Income |
|
Range
|
• Distance between the highest and lowest score
|
|
Variance
|
• Average distance from the mean insquared units
• Formula: sum of squares/number of scores • Only works if it is a bell curve! o Bell curve can be established if the Mean and Standard Deviations are available |
|
Standard Deviation
|
how much scores vary from the mean in the original unit of measurement (= variance check)
|
|
Standard Scores (z-scores)
|
• How much scores vary from the mean in common units (ex; p. 303)
• Taking out the variance and shifts everything to the center • Allows researchers to be more standardized |
|
Interval Statistics
|
used to generalize and make claims from a data set
|
|
Estimation
|
generalizing sample characteristics (sample) to population characteristics (parameters=M)
|
|
Characteristics of a Normal Distribution
|
• Center point is the exact middle & highest point
• Center point is the M, Mdn, & Mo • Scores on both sides are symmetrical (in SD) • As a result, value probabilities (% of population) can be estimated • Most variables are assumed to be normally distributed in populations o i.e. low, below average, average, above average, high o unless known to be otherwise (i.e. income) |
|
Central Limits Theorem
|
• For random samples of sufficiently large size, the distribution is almost always approximately normal even if smaller samples are not
|
|
Law of large numbers
|
the larger the random sample, the closer it approximates the true population and the small sampling up to 1,600-2,000 people (little error reduction beyond that)
|
|
Random Sample
|
the population parameters are “x”
|
|
Non-Random Sample
|
the sample characteristics are “x” and we believe the population parameters are like “x” (but cannot know for sure)
|
|
Significance Testing
|
• Significance testing=null hypothesis testing
• Significance Testing: the process of testing • Based on previous research (literature review), researchers ask for questions or make predictions about differences between groups or relationships between variables |
|
Research Question/Two-tailed hypothesis
|
explores or predicts if any difference or relationship exists between variables
|
|
One-tailed hypothesis
|
predicts a specific difference or if a relationship exists
H: Males disclose more than females |
|
Null Hypothesis
|
states no difference or relationship exists (any found are simply random chance)
H: No difference between males and females regarding of self-disclosure |
|
Statistical Tests
|
do NOT prove anything, they show percent of likelihood of occurring
|
|
Type I Error
|
rejecting a null hypothesis that is probably true, when generalized to the larger population it does not fit
|
|
Type II Error
|
accepting a null hypothesis that is probably false
|
|
Statistical power
|
the probability of rejecting a false null hypothesis (a correct assessment for the population)
|
|
Difference Analysis
|
examines differences between categories of an independent variable on another variable
|
|
Nominal Variable Tests
|
Chi Square
|
|
Chi-Square Test
|
examines the differences between categories of nominal variables
|
|
One-Variable Chi-Square Test
|
differences between categories of one variable
• IV=stadium food—Are preferences different for hotdogs or metts? |
|
Two-variable Chi-Square test
|
differences between categories of two variables
• IV= Sex (Male/Female) DV (High, Low) |
|
Ordinal Data Tests
|
Median Test
|
|
Median Test
|
differences between categories on ranked variable
|
|
Interval/Ratio Data (Likert) Tests
|
t-tests, ANOVA, Multiple Comparison Tests
|
|
t-tests
|
differences between 2 means (average) on interval/ration data
|
|
Independent Sample t-test
|
different groups (males, females) and groups are independent (nominal)
|
|
Related Measure t-test
|
same group, means at different times (pre/post tests)
|
|
ANOVA
|
(Analysis of Variance): differences between 2 or more means on interval/ratio (typically 3 or more)
|
|
Multiple Comparison Test
|
(Post hoc comparisons): are required to determine which differences are significant
|
|
Types of Relationships between Variables
|
Unrelated, Linear, Non-Linear
|
|
Correlation
|
a statistical relationship between 2 variables (i.e. indicates how similar are peoples’ shared meanings on 2 given variables (operationalized groups). NO ASSUMPTIONS OF CAUSATION ARE MADE!
|
|
Correlation Coefficient (r)
|
indicates the level of association
|
|
Pearson's r
|
calculates a correlation by comparing how much a score on a measure deviates from a mean compared to how much each score on another deviates from a sample mean. The more similar the variation (the M) the more correlation
|
|
Correlation Strength
|
.90= high, .70=strong/standard, .40=moderate, .20-.30=weak, .10=very weak
|
|
Coefficient Determination
|
o Squares correlation
o Indicates the truest strength of association (i.e. just how similar is the variation across variables) o .81=high, .49=strong, .16-.25=moderate, .04-.09=weak, .01=very weak |
|
Correlation Matrix
|
presenting correlations and significance levels in table format
|
|
Partial Correlation
|
partitioning out (controlling) the variance of 1 or more variables, when multiple variables are studied
|
|
Regression
|
Predicts or explains the change of one variable (criterion variable) based on the assumption of a causal relationship with a precursor (predictor) variable. Assumes causation
|
|
F-test
|
is used to calculate the statistical significance of causal relationships
|