Term
|
Definition
| when a hypothesis specifies a direction of outcome |
|
|
Term
|
Definition
hypothesis must have possibility that it won't gain empirical support
ex. tomorrow it will rain (falsifiable) |
|
|
Term
| two-tailed/non directional hypothesis |
|
Definition
| when a hypothesis predicts a relationship of some sort will be observed or that a difference will be found |
|
|
Term
|
Definition
| the more ____ a hypothesis the better |
|
|
Term
|
Definition
|
|
Term
|
Definition
| the implicit presumption that there is no difference between groups or no relationship between the variables being studied |
|
|
Term
|
Definition
| the body of statistical computations relevant to making inferences from findings based on sample observations to some larger population |
|
|
Term
| deductive-to-inductive-to-deductive process |
|
Definition
first, inferential statistics are used probalbilistically
second, rival theories are tested to prove similar outcomes
often, the results of empirical observation are used to revise a theory |
|
|
Term
|
Definition
| process of (re)producing societal structures |
|
|
Term
| deductively derived theories used heuristically |
|
Definition
| researcher reasons specific research questions to guide the kinds of things looked for in the empirical study - Giddens believes groups create structures unique to their group - structuration theory allows us to understand the process by which groups reach decisions |
|
|
Term
| Nuremberg Code ethical principles |
|
Definition
-voluntary participation
-no harm to participants
-avoidance of research where death or disabling injury was likely
-commitment to end any experiment if its continuation would likely kill, etc. participants
-research only conducted by highly qualified researchers
-commitment to research only for the good of society |
|
|
Term
| ethical issues in conducting research |
|
Definition
-voluntary participation
-informed consent
-anonymity & confidentiality
-deceiving is unethical & deception within social scientific research needs to be justified by compelling scientific or administrative concerns
-debriefing-entails interviews to discover any problems generated by research so those problems can be corrected |
|
|
Term
| ethical issues in analyzing research |
|
Definition
-objectivity & ideology (researchers must try to be objective, even though human nature is subjective & researchers must face scientific facts)
-protecting study participants (use pseudonyms for patients)
-honesty & integrity (researcher bears the ethical obligation to be true to his data- must make shortcomings known to reader - researchers must resist temptation to save face and ego -
-avoiding plagiarism |
|
|
Term
|
Definition
| entails interviews to discover any problems generated by the research experience so that those problems can be corrected |
|
|
Term
|
Definition
| institutional review board - est. to ensure risks faced by participants are minimal. researchers must form proposal for research and submit to IRB. |
|
|
Term
| exemptions from IRB proposals |
|
Definition
-research conducted in educational settings
-research involving use of educational tests
-research involving the collection of existing data
-research conducted by or subject to approval of department or agency heads
-taste & food quality evaluation |
|
|
Term
|
Definition
| man researched homosexual encounters in public restrooms - posed as "watchqueen" - invaded privacy - deceit - it was unethical for Humphreys to trace participants to their homes & to interview them under false pretenses |
|
|
Term
| Stanley Milgram's experiment |
|
Definition
| brought in 40 men & had them play "teacher" and "pupil" where teacher sent shocks to pupil if they got the question wrong. they were encouraged to continue shocking the pupil even if they heard them scream or pass out. caused pain & trauma to "teachers" - unethical |
|
|
Term
|
Definition
| process of coming to an agreement |
|
|
Term
|
Definition
| things we can observe simply & directly - like color |
|
|
Term
|
Definition
| more subtle, complex, or indirect observations - like a person's gender |
|
|
Term
|
Definition
| theoretical creations that are based on observations but cannot be observed directly or indirectly - like scholastic aptitude |
|
|
Term
|
Definition
| describes what something means by relating it to other abstract concepts |
|
|
Term
| types of conceptual definitions |
|
Definition
a. daily - generally accepted by society
b. poetic - figurative interpretations
c. scholarly - specific, technical |
|
|
Term
|
Definition
describes what is to be observed by specifying what researchers must do to make observations
*the specific steps to be followed - the "recipe" |
|
|
Term
|
Definition
| moving from conceptual definition to operational definition |
|
|
Term
|
Definition
1. respect for persons - treated fairly & openly with choices
2. beneficence - systematic assessment of risks and benefits
3. justice - who should receive benefits of research - who should bear its burdens |
|
|
Term
| levels of measurements for measuring variables |
|
Definition
-nominal - numbers used to "name" a category (eye color)
-ordinal - rank order used to determine differences (armed forces)
-interval - establishes equal distances between points but zero is arbitrary (fahrenheit)
-ratio - extends interval to include absolute zero (age) |
|
|
Term
|
Definition
| develops many statements about variables - and people rate statements 1(poor) - 11(great) statements rated consistently are assigned value based on avg. scores - final statements used to measure variables |
|
|
Term
|
Definition
"method of summated agreement"- create statements that are clearly positive or negative towards topic - ask people to indicate degree to which they agree/disagree with statement
-sum responses too all items
-infer overall attitude from summed ratings |
|
|
Term
|
Definition
all research projs. should be valid and reliable.
validity = accuracy
reliability = consistency |
|
|
Term
| types of measurement validity |
|
Definition
-external - generalization (results can be applied to larger sets of people)
-internal - study should be designed to lead to accurate findings
-measurement - accuracy of measures (technically also an internal validity) |
|
|
Term
|
Definition
reliability (consistency) is key to measurement validity
types:
-content validity - measure looks like what it's supposed to be measuring
a) face validity
b) expert validity
-criterion validity - assessing a measure's validity through its relation to some external criterion
a) predictive validity - predicts future behavior
b) one variable is compared with an established variable
-construct validity - measure should relate appropriately to other theoretically consistent variables
a) convergent construct validity- converges positively with related variables
b) discriminant construct validity - measure diverges from diff variables |
|
|
Term
|
Definition
researchers measure variables in consistent & stable manner:
consistent across time
consistent across individuals
consistent across questions |
|
|
Term
| reliability (how it's measured) |
|
Definition
| usually measured with correlation called coefficient alpha -1 (perfect neg.) one goes up, one goes down - 0 (no relationship) - +1 (perfect pos.) both go up or down |
|
|
Term
|
Definition
to obtain reliability statistics diff. methods can be used:
1. test - retest method
2. split-half method (randomly halve your questions & see if they both reach similar results on your subj)
3. intercoder reliability - where two coders agree on how they classify data
4. use established measures - what's worked in the past
5. use alternate form (create 2 & see how results compare) |
|
|
Term
| representational validity |
|
Definition
| whether a measure's categories are meaningful to the people who are being assessed |
|
|
Term
| threats to internal validity |
|
Definition
1. history - changes in the external environment affect people
2. sensitization (testing effect)- initial measurement influences later measurements
3. instrumentation - changes in methods used to collect data
4. maturation - physical or psychological changes within participants over time
5. mortality - loss of research participants
6. selection - methods used to obtain participants |
|
|
Term
|
Definition
| convenience, purposive or judgmental, snowball, quota |
|
|
Term
|
Definition
| nonprobability - limits sample by convenience (classroom) |
|
|
Term
| purposive or judgmental sampling |
|
Definition
| nonprobability - actively selecting particular participants based on the study |
|
|
Term
|
Definition
| nonprobability - finding a participant and asking if they know someone who's appropriate & so forth |
|
|
Term
|
Definition
| nonprobability - convenient sampling until you've met the quota for your group |
|
|
Term
|
Definition
| simple random, systematic, stratified |
|
|
Term
|
Definition
a sample will be representative of the population if aggregate characteristics of the sample closely approximate those same aggregate characteristics in population
*if population has 50% women - sample should have close to 50% women |
|
|
Term
|
Definition
| THE BEST* -everyone has fair chance of being picked |
|
|
Term
|
Definition
where every X element is chosen to study:
sampling interval = population size/sample size
sampling ratio = sample size/population size |
|
|
Term
|
Definition
| -separate groups & pick one from each group |
|
|
Term
|
Definition
| initial sample of groups of members (clusters) is selected first then all members of each cluster is selected first - then the members listed in the selected clusters are subsampled - providing final sample of members |
|
|
