2017-09-21 01 Exam Flashcards
primary research article
(A & L, 32)
- author(s) report original research they conducted
steps in the research process
(A & L, 11-25)
Aka the scientific method
1) identify your topic: interests you and others based on an established opinion/belief, no special/specific participants, have q’s but read past research to develop a strong hypothesis/design
2) find, read, and evaluate past research to develop hypothesis/research q: constantly look at past research
3) further refine topic and develop a hypothesis/research q: testable hypothesis
4) choose research design: feasible with resources and time, ethical
5) carry out your study: approval from professor, IRB
6) analyze the data: chose analysis based on hypothesis
7) communicate results: fit/don’t fit with past research, limitations
correct format for research article
(A & L, 48-58) (in-class 8/31)
1) Abstract
2) Introduction
3) Method
3) a) Participants
3) b) Procedure
3) c) Measures
4) Results
5) Discussion
abstract
(A & L, 51) (in-class 8/31)
- 150-200 words
- like a movie trailer (but spoils the end)
- always first, but created last
introduction
(A & L, 53) (in-class 8/31)
- introduces topic, why it’s important (why people should care)
- build case for study: describe past research, gaps in it, and limitations (organize broad to narrow)
- introduce new study, how it’s addressing past limitations/gaps
- state hypothesis/research questions
method
(A & L, 54) (in-class 8/31)
a) Participants
- how many, who are they, consent?, was anyone excluded
b) Procedures
- what did you do step-by-step, data collection process, include anything relevant for replication
c) Measures
- what were key variables?, how were they measured?, example items, scale scoring
results
(A & L, 56) (in-class 8/31)
- “just the facts”
- objective description of the results
EX: X correlated with Y
discussion
(A & L, 57) (in-class 8/31)
- restate hypothesis: if findings supported past research, relate to previous research
- describe limitations
- why results are interesting/useful
correct APA format and references format
(in-class 8/31) Dutton, H., & Shen, H. (2017). Title: No more uppercase letters. Just Journal in Italics: 40(5), 223-225. dio:00000
steps to ensure a study is ethical
(A & L, 3-11) (in-class 8/29)
- informed consent: explain study purpose, can withdraw at any time, risks
- confidentiality: only researchers and participants know defining characteristics (anonymity: only participant knows)
- incentive: nothing unreasonable (a million dollars)
- deception: none or ethical
- debriefing, answering questions: asap after the study
- approval from IRB
descriptive research design
(A & L, 106, 117)
- Survey
- Interview
- Questionnaires
- Observational
- Archival
correlational research design
(in-class 8/29) (A & L, 20)
- looks for relationship of 2 observed variables
- all about causation
experimental research design
(A & L, 19)
- determines causal relationship by manipulating IV, measuring DV, and random assignment
independent variable (and levels)
(A & L, 21) (in-class 8/29) - variable that's manipulated in an experiment Levels: a control group and then 1 or more other assignments/groups
dependent variables
(A & L, ) (in-class 8/29)
- variable that’s measured in an experiment
- expected to change based on IV
study reliability and replication
(A & L, 69-71, 76) (in-class 9/4)
- how generalizable is the study?
- extent to which a set of findings is reproducible
- does the measure have similar results in many trials
- are there relatively low levels of measurement error?
internal validity
(in-class 9/4, 9/7) (A & L, 71) (in-class 9/4: 351)
- all about causation
- allows researchers to state that they’ve identified causal associations
- was the IV the sole cause of changes in the DV
EX:
- high: well controlled experiment (ideally with random assignment)
- medium: correlational study with statistical controls
- low: did not rule out potential third-variable explanations
scales of measurement
(in-class 9/4) (A & L, 79-83)
- nominal
- ratio
- interval
- ordinal
nominal scale
(in-class 9/4) (A & L, 80)
- identity (each number has a specific meaning)
- used to measure categories
ratio scale
(in-class 9/4) (A & L, 80)
- identity (each number has a specific meaning), order (numbers on a scale, in ordered sequence), equal intervals (distance between numbers on the scale is equal), true zero (fixed-point)
- used to measure quantities
interval scale
(in-class 9/4) (A & L, 80)
- identity (each number has a specific meaning), order (numbers on a scale, in ordered sequence), equal intervals (distance between numbers on the scale is equal)
- used to measure ratings
ordinal scale
(in-class 9/4) (A & L, 80)
- identity (each number has a specific meaning), order (numbers on a scale, in ordered sequence)
- used to measure rankings
operational definition
(in-class 9/4) (A & L, 77)
- specifics of how the variable is measured
- so it can be exactly replicated
construct validity
(in-class 9/4: 351)
- abstract psychological phenomenon
- inferred from observable behavior
EX: love, attraction, engagement
– need to be specific in how you’re going to measure (unlike a ruler)
- did the authors measure or manipulate ~all facets of the concept~ that they claim to be measuring or manipulating?
- can’t measure directly
measurement validity
(A & L, 76)
Measurement is accurate and measure’s what it’s supposed to
- Construct validity (content, divergent, criterion[predictive, concurrent])
content validity
(in-prac 9/11) (A & L, 93)
construct validity
- have to measure all aspects of the construct
EX: only measuring one part, but now the full range of what’s being measured
divergent validity
(in-prac 9/11) (A & L, 94)
construct validity
- negative or no relationship between 2 scales measuring different constructs
- 2 measures that only really measure one thing
criterion validity
(in-prac 9/11) (A & L, 94)
construct validity
- positive correlation between scale scores and a behavioral mesaure
measurement reliability
(in-class 9/4) (A & L, 90) - internal consistency (Cronbach's Alpha, split-half reliability), test-retest reliability, alternate forms reliability, inter-rater reliability
internal consistency
(in-class 9/4) (in-prac 9/11) (A & L, 91)
Consistency of participants responses to all items in a scale
- Cronbach’s Alpha, split-half reliability
Cronbach’s alpha
(in-prac 9/11) (A & L, 91)
measures internal consistency
- computes inter-correlations of scale items
- values >0.7 are acceptable internal consistency
- when alpha is <0.7, sometimes items deleted to reach 0.7 standard
split-half reliability
(in-prac 9/11) (A & L, )
measurement reliability
- correlations between 2 halves of the items on a scale (ex: even number items correlated with odd-numbered items
- values >0.7 are considered acceptable reliability
test-retest reliability
(in-class 9/7) (in-prac 9/11) (A & L, 91)
measurement reliability
- check again if the measurement could be unstable
- same question, different answer (means question is picking up something else)
interrater reliability
(in-class 9/7) (in-prac 9/11) (A & L, 91)
measurement reliability
- for when people are being observed, field work with complicated behaviors (not relevant for survey data [questionnaire])
- multiple people coding
3 criteria of an experiment
(A & L, 19)
1) random assignment
2) manipulation of IV
3) measurement of DV
descriptive study
(A & L, 19)
- describe variables, but don’t examine relationship or causation
Survey research
(A & L, 106)
Descriptive research design
- interviews or questionnaires where participants report attitudes or behaviors
- Advantages: insight into how participants see themselves, can be administered easily (online, a lot at a time)
- Disadvantages: social desirability bias, interviews time consuming, interviewer bias, questionnaire don’t get as much in depth info
interview
(A & L, 106, 117)
Descriptive research design, survey
- 1-1 conversations directed by researcher
- phone, in person, email
- can’t be anonymous, but can be confidential
questionaire
(A & L, 85, 108, 117)
Descriptive research design, survey
type of measurement
- allow for anonymity (reduce social desirability bias)
- can be administered easily (online, a lot at a time)
- asses one or more construct
observational study
(A & L, 109)
Descriptive research design
- recording behavior
- can be in addition to other research methods
- Advantages: reduce social desirability bias, time consuming to record and code data, potential observer bias)
social desirability bias
(A & L, 106)
- in self-reports, people responding in what they thing is the most desirable or ideal
Convert observation
(A & L, 111)
Descriptive, observational
- observations made without participants knowing
- to capture participants natural and spontaneous reactions
- can be unethical (can be in a public place)
overt observation
(A & L, 111)
Descriptive, observational
- no attempts made to hide observation
- participants could change behavior if they know they’re being watched
- researchers usually give time for participants to acclimate to situation