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;
26 Cards in this Set
- Front
- Back
|
Independent vs. dependent variable
|
Ind variable: cause
Dep variable: effect |
|
|
3 Reasons to conduct exploratory studies
|
1. Examine new interest to gain a better understanding
2. Test feasibility of larger study 3. To represent methods for study and establish questions about the subject |
|
|
3 Criteria for causal relationships
|
1. Variables must be correlated
2. Cause takes place before effect 3. 3rd variables must be eliminated (no confounding variables) *Can't be proven with one study **All 3 must be satisfied |
|
|
4 units of analysis (who/what studied)
|
1. Independent
2. Group (can take sample of groups based on answers of ind) 3. Organization (involves people) 4. Social artifacts (anything other than people) |
|
|
3 Types of longitudinal studies
|
1. Panel: follows same people over time
2. Cohort: follows different people over time 3. Trend: same type of people surveyed annually, but not followed |
|
|
Concept vs. Conception
|
Concept: can't directly observe; term or word; what people hear
Conception: mental image the summarizes collections of related observations & experiences; what people think |
|
|
Conceptualization vs. Operational Definition
|
Conceptualization: process to specify meaning of term; how people understand or define
Operational Definition: how concept is measured (ex: # minutes; day/week doing _____) |
|
|
Descriptive vs. Inferential statistics
|
Descriptive: Specific group; summarizing large amount of data using mathematical analysis "data reduction"
Inferential: Large Group |
|
|
4 Levels of measurement
|
1. Nominal: to name or classify into categories; no difference in value
2. Ordinal: rank # 3. Interval: equal distance between points, but zero isn't = to nothing (temp) 4. Ratio: True zero point |
|
|
Reliability vs. Validity
|
Reliability: not equal to accuracy; applied repeatedly to same object yields same results each time. (ex: scale)
Validity: extent to which measure adequately reflects the real meaning of the concept |
|
|
3 ways to reduce social desirability bias
|
1. minimize sense of judgement & maximize importance of accuracy
2. use self-admin questionnaire 3. assure anonymity & confidentiality |
|
|
Ways to make sure participants understand possible study risks
|
Benefits must outweigh harms
"Informed Consent" |
|
|
Anonymous vs. Confidential
|
Anonymous: no identity info shared
Confidential: ID info shared, but no released outside of study |
|
|
Term used to tell subjects truth after initially deceived
|
"debriefing"
Failure to inform or be honest about study; who is conducting/funding; events occurring |
|
|
Panel whose members review research proposals involving human subjects so rights and interests are protected
|
Institutional Review Boards (IRB) 5+ members; non-scientists
|
|
|
Statistic vs. Parameter
|
Parameter: Summary description of given variable in population (ex: Avg. %)
Statistic: summary description of given variable in sample |
|
|
Sampling Frame
|
List of sampling units from which sample is selected (ex: phone book; teacher/class registry
|
|
|
4 Types of probability sampling techniques
|
1. Simple Random
2. Systematic 3. Stratified 4. Cluster |
|
|
Using a table of random #s for Simple Random Sample
|
#s 1-40 and every 10th person picked; everyone has equal chance of being picked
|
|
|
Systematic Sampling
|
Every Kth element in total list chosen for sample; not everyone has = chance once Kth element chosen
|
|
|
Sampling Interval vs. Ratio
|
Interval: population size / sample size (formula for Kth element)
Ratio: sample size / population size |
|
|
Simple Random vs. Cluster Sampling
|
In Simple Random sampling, each observation element selected is viewed as an individual. In Cluster sample, units/groups of elements are selected (households)
|
|
|
Simple Random vs. Stratified Sampling
|
In Simple Random sampling, elements are chosen from total population. In stratified sampling, elements are chosen from subsets of the population (males vs. females, etc..)
|
|
|
4 Types in non-probability sampling
|
1. Available or Convenience
2. Purpose or Judgmental 3. Snowball 4. Quota *Not equal chance for selection |
|
|
Under which conditions would Snowball Sampling occur?
|
When members of special population (homeless) are difficult to locate. Collect info on a few members of target pop and then ask for location of others.
|
|
|
2 problems with Quota Sampling
|
1. Proportions of population with certain characteristics not commonly accurate
2. Bias exists in selection of sample elements within a given cell |
