Research Methods 2
Terms
undefined, object
copy deck
- APA ethics code
- -represents the consensus of the psychology profession about what is considered acceptable practice
- What are the 6 general ethics "rule of thumb"
-
1. be explicit about research project
2. treat subjects w/ respect
3. find out about previous guidelines
4. don't lie about procedure b/c could be fined or have research abilit revoked
5. don't harm or put subjects at risk
6. dont' perform unnecessaryly hard/difficult procedure - Institutional Review Board (IRB)
-
-appropiate constituted body of ppl that look through research project to decidie if appropriate
-studies present as little risk to subjects as possible and have scientific merit (risks and payoffs)
-each institution has own and is required by federal govern if recieve federal funding - Responsibilities of researcher
-
-see ethical principles are followed
-protect subjects from harm
-protect subjects from stress
-conflict bwt values of expanding knowledge and cost of research to participants - What factors must be considered to protect participant from harm?
-
-any new situation is stressful and concievably could be harmful so impossible to aviod risk of harm entirely
-stress can physical or psychological (deceit, mental stress)
-to judge acceptability of stress assess how stressful situation is likely to be compared w/ activities of everyday life - Characterisitcs of informed consent
-
-ensures participant is taking part voluntarily and is aware of what is about to happen
-purpose of research, duration, and procedures, rights to decline or withdrawl and the consequences, potential risks, discomfort, or adverse effects, prospective benefits, limits of confidentiality, incentives, and whom to contact - Privacy and freedom from coercion
-
-balance right to privacy against welfare of society as whole
-need to know individual subject results wont be released to the public
-names must be kept seperate from subject's data/results
-results can't be used against subject in any way
-serious sums of money cannot be offered for participation and ppl cannnot be induced to participate w/ false promises - When and what type of deception is permissible
-
-no alternative procedures are feasiblt and the payoffs exceed the risks
-deception cannot be used w/ research that may cause physical pain or severe emotional distress
-explanation of deception must be given to participant as soon as feasible
-have to tell what research is really about at end
-if show emotional stimuli should give subset to see if subject is okay w/ them before - confederate
-
-actors in expt
-form of deception
-subject must be told about them at conclusion of experiment - what is debriefing
- -process of informing the subjects after the session of the experiment's true purpose to increase their understanding and to remove possible harmful effects of deception
- What info is included in debriefing?
-
-general and theoritical debriefing
-how and why deception was used and its relation to the theoritical construct
-contact info of researcher and reference list - what is the role of the participant in research
- -consent = contractial agreement so should do what expected to do
- Guidelines for animal research and for insuring animal welfare
-
-usually not in psychology but different standards b/c can be invasive
-only 50% of legal dose can be given to animal
-should be treated humanely (aniseptics, clean living conditions, can't be too stressed) - animal rights
-
notion that animals have the same sort of rights as people, including legal rights (not generally accepted)
-unethical to use animals for research, food, pets, recreation, etc - animal welfare
-
generally accepted term for concerns about the care and use of animals
-humane treatment of animals - speciesism
-
term used by analogy w/ racism and sexism by those who claim that it is unethical to treat animals differently from humans, particularly in research
-reject any special moral status for humans - invisible college
-
informal communication network of ppl having common scientific interests
-new ideas and results are usu. first discussed through this - discourse community
-
group of ppl who share common goals, a public forum, common knowledge, and a specialized language
-psychologists (audience) - Goals of scientific writing
-
-convey a message clearly, concisely, and interestingly
-persuade as well as inform - agrument
- a set of reasons in support of a proposition
- thesis
- the proposition that is supported by an agruement
- Characterisitcs of good scientific writing
-
1. Clarity: say exactly what you mean as directly as possible
2. Brevity: brief (good communication)
3. Felicity: pleasingness of style (liveliness and grace) - Heurisitcs for good writing (esp. clarity)
-
1. prepare outline
2. use technology (spell check)
3. be prepare to "kill your baby" b/c it'll take several tries to make it perfect
4. have someone else proofread
5. read paper aloud - Science Journal
-
-includes many science areas like psychology and physics
-for up and coming research - APA guidlines and their benefits benefits
-
-avoid sexist language and ethnic bias in writing (ambiguity and stereotyping)
-standardized format that's helpful for publisher
-don't have to reformat for each journal - Parts of APA paper
-
title page
abstract
introduction
method
results
discussion
references
foot notes/ author notes if any
tables
figure caption
figures - Title page
-
-title (centered, include keywords, convey main idea)
-author(s): listed in order of importance of their contribution and last author is usually most prestigious (money source)
-affiliations
-running head in upper right corner - abstract
-
-summary of most important elements
-self-contained
-max 120 words
-block form, on pg 2, heading abstract centered - Body
-
-intro, methods, results, discusion
-begin w/ title centered
-centered heading organize subsections - Introduction
-
-not labled, on pg 3
-state general problem examined
-discuss relevant literature
-indicate hypothesis
-state how study will contribute to understanding the problem
-expected results - Method (heart of the paper)
-
-in past tense
-allows someone to repeat experiment exactly in all essential detials
-judge validity of conclusions by comparing them w/ the method section
-subheadings: participants, materials, design (logic/ varibales), procedure - Results
-
-past tense (what you found)
-indicate any data transformations made before analysis
-describe statistical procedures
-displace/summarize data in table, figure, or graph - Discussion
-
-state relationship bwt findings and hypothesis
-present tense
-interperate results and relate them to literature (similiarities and differences) - References
-
-ties paper to literature
-only lists papers cited in text
-3 componets: author, title, yr
-hanging indent
-new page - Author notes
-
-provide mailing address, acknowledgements (of financial supportors and technical assitance)
-distracting--aviod - Foot notes
-
-content and copyright permission types
-aviod footnotes to content of paper
-copyright footnotes are used for reprinted tables or figs - Tables
- -supplement material in text not duplicate material
- Figure captions
-
-each figure has one
-describes content of figure
-come after tables but before figures themselves - Figures
-
-professionaly rendered (high quality)
-simple as possible (clear/brief)
-aviod color
-stylistically pleasing - Steps in the publication process
-
1. before write manuscript
-decidie which journal going to submit to
-impact factor of journal (circulation, how many ppl read it, how many submissions it receives, etc.)
-choose journal that publishes articles similiar to yours and the one you cited the most
2. before submit manuscript
-know requirements of journal (hard copy or electronic submission)
-know turnaround time (how long it takes to review)
-balance prestige of journal w/ difficulty of getting published
-prepare cover letter - Steps in publication process continued
-
3. after submission
-paper goes to reviewers (subjective process)
-4 options
-resubmittL need another cover letter and address every reviewers pts and whether followed suggestion and if not why - cover letter
-
-addressed to editor
-include title and some details like theory, methods, results, significance
-give return address and # - If published what happens
-
-"in press"
-you get copyeditied version of manuscript and make corrections
-get proofs of what paper will look like
-can take 12-18 months from acceptance to publication - Oral Presentations
-
1.Talks: shorten paper, simplify material, organize talk to main pts of paper w/ 1 or 2 pts in each section, most of time should be spent on results w/ emphasis on main findings
2. Posters: place copies of abstract or paper in pocket on board, stay near board to talk to ppl - Validity
-
-an indication of accuracy in terms of the extent to which a research conclusion corresponds with reality
-truth of observations or experiment
-dependent on validity and reliability - Reliability
-
-observations were obtained on multiple samples (many times)
-only one measurement - Internal Validity
-
-extent to which a study provided evidence for a cause-effect relationship bwt the IV and the DV
-degree to which 2 or more
variables are related
-variables important hypothesis must be controlled - Confounding variables
-
-covaries w/ the IV
-their effects cannot be sorted out
-big threat to validity in an experiment - Subject variables
-
-differences bwt subjects that cannot be controlled but can only be selected
-gender, age - Threats to internal validity
-
1. history: events that take place in time could change results
2. maturation
3. testing (order) effects
4. regression effects
5. selection-experimenter bias
6. mortality - Construct validity
-
-extent to whcih the results support the theory behind the research
-measure what suppose to
-experimental results have to support theory motivated by experiment
-if measurement used in some research lacks construct validity the research as a whole will also
-rule out other possible theoretical explanation - auxilliary hypothesis
-
-better theory supports results
-theory x-hypothesis-expt-results
-theory y supports hypothesis - External Validity
-
-how well the findings of an experiment generalize to other situations or populations
-whole pt of research - Statisitical Validity
-
-extent to which data are shown to be the result of cause-effect relationships rather than accident
-statistical tests guarantee only low probability not definite cause-effect relationship
-statistical significance (type of test) - Maturation
-
soure of error related to the amount of time bwt measurements
-development of subjects (esp w/ children)
-subjects change - Regression effects
- tendency of subjects with extreme scores oon a first measure to score closer to the mean on a second testing
- random error
- that part of the value of a varibale that can be attributed to chance
- Mortality
- the dropping out of some subjects before an experiment is completed causing a threat to validity
- Threats to construct validity
-
1. poor operational definitions
-loose connection bwt theory and method
2. ambiguous effects of IV's - Examples of ambiguous effects of the IV's
-
1.Hawthorne effect: individuals behavior may be altered b/c somebody know they are being observed
2. good-subject tendency: participants act according to what they think the experimenter wants
3. evaluation apprehension (ie social desirability)
-alter behavior to appear socially desirable (anxiety, want to look good) - Threats to External Validity
-
1. population: generalizing from sample to population
2. time: population may change w/ time so conclusions not valid anymore
3. other settings: move from lab to another or real world setting - Threats to statistical validity
-
1. erroneous stats (wrong tests, not correcting for multiple occurences)
2. not enough power (not enough subjects or observations to detect effect) - Control
-
any means used to rule out threats to the validity of research
-not every experiment requires a control group
-providing a standard against which to compare the effect of a particular IV
-control for as many variables as possible for best validity (internal validity)
-ability to restrain or guide sources of variability in research - control conditions
-
-condition in a within-subject design experiment that does not contain the experimental manipulation
-serves as baseline - experimental conditions
-
-treatment
-what you change to affect measurement
-allows for strong inference about casaulity - experimental group
- subjects in an experiment who receive treatment
- control group
- subjects in a between-subjects design experiment who are like the experimental group in every respect except that they do not recieve treatment
- With-In Subjects Experiment
- -research design in which each subject experiences every condition of the experiment
- Between-Subjects Experiment
- -research design in which each subject experiences only one of the conditions in the experiment
- Methods of Experimental Control
-
1. laboratory setting: control everthing in lab setting (social research is done in field setting)
2. laboratory preparatons: method/procedure is same for all subjects (also allows for replication)
3. instrumentation: make response reliable, improve the measurement of the behavior (DV) - Modularity Theory (Fodor, 1983)
-
-mind is seperated into different processing modulars
-domain specific: incoming info is specific (1 type of processing )
-information encapslated: no other input from other modules
-automatic: automatically process incoming visual info a specific way - Design 1 : Between-Subjects
-
-2 conditions: control group and experimental group
-participant assignment: random to avoid confounds like order effect
-instrumentation (how test)
single trial or multiple trials
-counter-balancing: order effects, control number of each type of response - counterbalancing
- controlling for order and sequence effects by arranging that subjects experience the various conditions in different orders
- Design 2: Within-Subjects
-
-same ppl run through each condition
-powerful control technique b/c variation caused by differences bwt ppl is greatly reduced
-basic logic: control condition vs experimental condition
-pros: more power, more control for bwt subject variability
-cons: not always feasible - When to use subjects as own control
-
1.logically possible
2.participating in all conditions will not destroy the naivete of the subject
3.serious contrast effects bwt conditions will not be present - Random Assignment
-
-unbiased assignment process that gives each subject an equal and independent chance of being place in every condition
-confounding of subject-related variables can only occur by chance w/ random assignment - Design 3: Matched Design
-
-match participates on certain variables
-justification conditions
1. important variable subjects differ on that can be controlled by matching
2. feasible to present a pretest to the subjects before assigning them to the condition
-still randomly allocate the members of the pair conditions - Matching
-
-yolking
-control procedure to ensure that experimental and control groups are equated on one or more variables before the experiment - Matched Design: pros/cons
-
pros: more statistically power
cons: more complicated, possibly less power when stats are based on fewer degrees of freedom - nuisance variables
- -a condition in an experiment that cannot easily be removed and so is made an IV as a means of control
- Statistical Control
-
mathematical means of comparing subjects on paper when they cannot be equated as they exist in fact
-broadly synonymous w/ inferential statistics - Replication
-
repeating an experiment to see if the results will be the same
-method of control - Direct Replication
-
-someone repeats essentially the identical experiment in an attempt to obtain the same results
-seldom carried out - Systematic Replication
-
-similiar experiment but with different type of subjects, different values of the stimulus, or with different ways of measuring the theoritical concept
-tests external validity - Independent sample t-test
-
assume have different SD for 2 groups
-2 groups come from different populations - Matched sample t-test
- can relate SD b/c matched individuals