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;
48 Cards in this Set
- Front
- Back
Direct observables
|
o Number of students, etc.
|
|
Indicatory observables
|
• Characteristics of what’s being observed
o Clothing of number of students observed |
|
Constructs
|
• Theoretical creations based on observations
• BUT cannot be observed directly or indicatory o Prejudice, anger, guilt |
|
Conceptualization
|
• Specifying what we mean when we use particular terms
o Prejudice, anger, guilt o Defining a feeling |
|
Operationalization
|
DETAILS
• Procedures/techniques used to find out something about your topic o To what extent is the research willing to combine attributes in fairly gross categories? • Income → how much details? • Attitudes → strongly agree and strongly disagree o 11 point scale → 0-10 |
|
• Nominal
|
just name, no rank
|
|
• Ordinal
|
rank order
|
|
• Interval
|
fill in the blank
o IQ scores o Temp o Zero doesn’t mean anything |
|
• Ratio
|
fill in the blank
o Income o Age o Zero means something |
|
• Test-Retest Reliability
|
o Give it now, give it later
|
|
• Split-Half Reliability
|
o Two parts
o 20 questions → one half has 10, and so does the other o Split it up between groups/people |
|
• Alternate-Forms Reliability
|
o Two different forms of the same test to the same person
|
|
• Internal-Consistency Reliability
|
o Overall degree of relatedness of all test items or raters
|
|
Validity
|
• How well does it measure what its supposed to measure
|
|
• Face Validity
|
o The quality of an indicator that makes it seem a reasonable measure of a variable
o Looks like it’d be ok |
|
• Criterion Related Validity/Predictive Validity
|
o The degree to which a measure relates to some external criterion
o Driving test → real driving o Design a test → relate to real life |
|
• Construct Validity
|
o The degree to which a measure relates to other variables as expected within a system of theoretical relationships
|
|
• Convergent Validity
|
o 2 measures that measure the same construct should be related
o happy people → high self-esteem |
|
• Divergent Validity
|
o 2 measures that measure different constructs should not be correlated
|
|
Index and Scale Similarities
|
o Both are ordinal measures of variables
o Both are composite → more than one data items (several indicators) |
|
Index and Scale Differences
|
o Indexes accumulate the scores assigned to individual attributes (support, stress)
• All items have same weight o Scales assign scores to patterns of responses • Items differ in the level of intensity • Strong topics |
|
o Uni-dimensional
|
(one aspect)
|
|
o Variance
|
(do you go to church?)
|
|
Empirical Relationships
|
• When respondents answer to one question help us predict how they will answer other questions
• “say yes to one, say yes to all” • If 2 items are empirically related, we can argue that each reflects the same variable and both can be included in the same index |
|
Handling Missing Data
|
• Few cases → excluded them from analysis
• Replace the missing data with the mean for a continuous variable o Income, Age. Etc. • Replace the missing values with middle value |
|
• Pairwise Deletion
|
o Use only completed data on only those variables selected for a particular analysis
|
|
• Thurstone Scale
|
o Large number of statements
|
|
• Bogardus Social Distance Scale
|
o How would you feel about having a member of X group…
• Close to my kids • In my club • As my neighbor |
|
• Semantic Differential Scale
|
o Rank answers between 2 extremes
o Good _ _ _ _ 5 Bad o Strong _ _ 3 _ _ Weak |
|
• Likert Scaling
|
o Use standardize response categories
o Really Strong, Strong, Neutral, Weak, Really Weak |
|
• One of most visible uses of surveys
|
political poll
estimating the votes of 100+ million by asking 2000 people |
|
• Population
|
o The theoretically specified aggregation of study element
|
|
• Sampling
|
o Selecting subgroups from a population
o People, objects, events |
|
• Sampling Frame
|
o List of elements from which a probability sample is selected
|
|
• Sampling Unit
|
o Element or set of elements considered for selection in some stage of sampling
o “Parents of children with autism”, “students attitude of teacher” |
|
• Random Selection
|
o Each element has an equal chance of selection independent of any other event in the selection process
|
|
• Parameter
|
o Description of the variable in population
|
|
• Statistic
|
o Description of the variable in sample
|
|
• Rep Sample
|
o Usually more important than the size of sample
o Needs to represent population in the best way possible |
|
• Probability Sampling
|
o Likelihood of any member of the pop being selected is known
|
|
• Non Probability Sampling
|
o Likelihood of any member of the pop being selected is not known
|
|
o Convenience Sampling
|
• Easy access to sample
• Quicker cheaper method |
|
Problems with convenience sampling
|
Nonrandom
Potentially biased Results cannot be generalized |
|
o Purposive
|
• Selecting a sample on people who have knowledge of a population, its elements, and the purpose of the study
• Participants has characteristics that the researcher wants |
|
o Snowball Sampling
|
• Collect data on members you can locate → then ask to locate other members of that population
|
|
o Quota Sampling
|
• Start with table describing characteristics of target population
Gender, Age, Education • Data is collected from people with the characteristics of a given cell |
|
• Probability Sampling
|
• All elements of a population have an equal chance of inclusion
“Fair” |
|
o Simple Random Sampling
|
• ID all elements of population
• List elements • Randomly select a sample from the list • Only works with a simple group of people |