Flashcards
Research methods
Cards in this deck (48)
- 1qualitative researchquantities/ words numbers
images
meaning - 2Qualitativesnumbers stats/ drawing a general conclution
- 3Basic• Aimed at anything intresting
• How does the wolrd work
• Resraches have freedom and intrinsic motivation
• Success if results impact science community
• Ex does prosciial behavior incres happiensss - 4aplied research• Aimed at practical problems
• Can we solve this problem
• Resercher as part of the job and at the emploeys demand
• Success if result are usavle
○ Ex do our costomers becom more happy and buy more if we greet them - 5Exploratroy• To explore
• What
• Idea generating
Feasibility stidues - 6Descriptive• To describe
• How? Who?
Caregorizing and classifying - 7Explanatory• To explan
• Why?
• Theory testing
• Model comparison - 8Case study• A study based around on thing
• Can be measerd more than once but always at least once - 9Cross sectionalA study of a group of people in one moment of time
- 10LongitudinalStudies a groupf of people at least twice across time
- 11what are the two state of behaviorautonomous state: direct thier own actions / pass of respncibility
agentic state: allow other to direct thier actions/ take responcibility - 12social theories vs every day theoriesupdated with light of new evidence
falsifiable
logically consistent - 13what are thoeris built from and what types are therebuilt from concepts
Variable not variable
dimension vs type
unidimentioanl vs multidimentional
concreate vs abstract - 14what is the deference between deductive and inductivedeductive--> general to specific
inductive--> specific to general - 15what are the range of theoriesempirical generalization: direction between observations and concepts
middle range: often guides new research
theoretical frame work: general frame work involving assumption guilding the dev of mid range theo - 16exaspling: Fabrication/faslificaiton/ quetionable research practices1 makng up data
2 disotrting data
3 cooking data/ concrealing data/ conflicts of intrest - 17explain : popular/ pro/ acedeminc sources1ads easy appealing
2 withing fields and between academic
3 only and for exsperts ina filed - 18what is an association in a studytwo events that have a relation. when one changes so does the other: when one happens the other will be affected/ varies in strength
- 19what is a ecologicla phalacyEcological fallacy is an error in reasoning that occurs when conclusions are drawn about individuals based solely on aggregate data of a larger group or population
- 20what is a sampling framea lsit of cases in a popilatio or the best approx of it
- 21what is a good sampling raitosomething with the smae characteristics as a population ( the percetage of teh group will differ based on the size of teh population)
- 22what is sampling errordeviation in a sampled vale versus the true population
- 23what is the law of large numbersit is an average result of one or multiple samples will become closer to teh true valie as your smaple size increases/ or the more studies are conducted
- 24central limit theoremas you increae the number of smaple the distrib of the average of tehose smaple will tend to becomecome normal
- 25probalilty samplingbased on probablalistic (math principles)
used in quanttative research
law of large number
when drawing a representative sample is essential - 26non probability samplingno probability theory
qualitative
whe you ain to get a new insight in a specific social phenomenon
when representatiatieness is less importatn - 27Non probability: convienince samplesampling is determinded how closle the participant s are close to the researcher/ not representative/
- 28Non probability: Voluntary response sampleex. posting on socail people who come forward to particiapte
- 29Non probability: purposivesearching for a certin type of person or chareteristic
- 30Non probability: deiveient casesearching out people that do not fit a dominatnt pattern
ex males in domestic abuse cases - 31Non probability snowball sampleone person tells another aquntince
larger number of people
will only get one type of person - 32Non probability respondent driven sampleyou look for one person and then they ask them to spread the word
- 33Non probability quota samplingyou determine befor hand what you are looking for and find interesting and sample from that. can be representaive but can fuck up in some areas while sucreading in others
- 34probability: random number generation methodsuses a random table (oldschool) or online coding/wesite
- 35probability: simple random samplerandomly draw a nuber of people from a group
as the population get more diverse it gets harder to get each subset - 36probability: strarified sampleseperate people into groups that you want to sample then do the randoms ample from that (can be 40% from one and 60% from another)
- 37probability: systematic samplestart out with a random sample then proceed with sampleing in a set interval ( 1 every 4 )
risk there could be an order for you sampling frame ( could make you only chose one charecteristic) - 38probability: cluser sampleusefull for larger geographic area
list all countries that have what you need
then pic a prediterminded number of them then use them as a your sampling frame
a problem could be that the country that you chose is bigger than others - 39face to face intervew pro and conleast confusing
good fo rlong survays
probes
best responce rate
-
highest cosr
most interiew bias
slow
socail desieability - 40telephone interview pro-conlow sampling bias
fleaxable
proves
good respmce rate
-
high cost
less anonymity
interviewer bias
less questions types - 41Self adminisrered paper surveyanon
easy
low cost
no interviewer bias
-
lowerst responce rate
responder unclear
situation unclear
no clarification - 42self amninistered web surveyfast
flexable lowerst cost
many question types
-
privacy
responder unlcear
situation unclear
sampling bias - 43three rules whne writing survey questionsclear: avoid ambiguity/no double questions (double barrel) / or negatives/ clear response options
simple: avoid jargon/ dont ask about reality/ sont ask about future/ no false premises
perspective taking: help respon dents fell that they understand the question and that thier answres are meaningful - 44issues in designing survey questions: order effectorder effect: An effect in survey research in which
respondents hear some specific questions before
others, and the earlier questions affect their
answers to later questions. - 45issues in designing survey questions: context effectscontext effects: As a practical matter,
for the same topic you want to use a funnel
sequence of questions—that is, ask more general
questions before specific ones - 46issues in designing survey questions: response set biasresponse set bias: response set bias—a tendency of
some respondents to agree and not really decide - 473 diffrent tpyes of filter questions'no option' choise not offered
'no option' choise offered
'full filter' ask whether respondents has opinion - 48open vs closed questionsOpen-ended question A type of survey research question in which respondents are free to offer any answer they wish to the question.
Closed-ended question A type of survey research question in which respondents must choose from a fixed set of answers.
Information
- Description
- Flashcards on Research methods , created by Gavin REID on 15/12/2022.
- Created by
- Gavin REID
- Views
- 146
- Rating