definition of an abstract used by researchers to measure or manipulate teh concept in a research study

Nominal

Ordinal

Interval

Ration

( No One Is Real)

Involves non ordinal categorical responses (Qualitative Data )

Ex: measuring mood they might ask ppl to choose from categories such as anxious , happy , depressed , and angry

Ex:College Majors - Math English Psych 

Ordered categorical responses ( qualitative)

ex: How often you drink- never , 1week, 2week,3week

ex: Ranking a product- first second third

Numerical responses that are equally spaced, but scores are not ratios of eachother ( quanitative)

Response from surveys: 1 lest like me , 3 somewhat like me, 5 most like me ( Likert Scale)

Temperatue ( F or C)

Involves numerical responses , where scores are ratios of eachother ( quanitative)

ex: amount of time to complete a task ( reaction time)

ex:distance , time , accuracy , weight , height, kelvin scale

indicates that a survery measures the behavior it is designed to measure

Ex: Poor operational def is used innacurate conclusions from the data may also result= LOW construct validity

survey response scale that involves pictoral response categories for participants with low verbal skills ( kids)

ex: mood responses

:( :[ :/ :] :)

• Presence/ Abcence variables
• Type Variables
• Amount Variables

• variable that involves a manipulation with a level that invloves the treatment and a level that dosnt involve the treatment
• Bivalent IV because they contain two levels
• Presence group: experimental group
• Absence group: control group
• ex: therapy , drug treatment , time constraint for a task

• variable that involves manipulation of types of treatment or different type of the something being manipulated
• ex: different types of drugs or therapies can be compared , different versions of an ad or product can be compared , different types of instructions for a task can be prepared

• manipulations of a the amount of a factor in each level
• ex: amount of drug to releive a symptom so each level of a variable includeds a different amount of the drug given 
• ex: amount of time passing btwn the study of the material and the test on that material. Each study has diff time delay ( 5 min 30 min 1 day)
• Multivalent Variable: IV variable contains three of more levels

• variable that allows comparison of groups of participants w/out manipulations ( no random assignment)
• gender , age , personality type, ethnicity
• any characteristic you can measure or observe from the participants can be used to group those ppl and create this variable
• Without manipulation its possible additional factors are causing differences to occur between groups
• ex: study that compares diff age groups on completion of a memory task to see if memory declines as ppl age. If they used computers in the experiemtn a factor could be older people were not as farmiliar with computers as the younger ones

• degree to which a study provides causal info about a behavior
• Study with good internal validity will provide a good test of causal relationship by removing alternitive explanations of the data
• well designed experiments have higher internal validity

• a source of bias in a study created when a researcher treats groups differently ( often unknowingly) based on knowledge of the hypothesis
• Test effectivness of new assignment: half class gets old half gets new. If the new assigment scores higher there are possible reasons: 

1. new assigment is more effective than the old one in helping kids learn

2. There are group differences ( better kids in new assign group)

3. experimentor bias where bc she expected the new assign to work better she way have been more enthusiastic  or encouraged them more than the other section. 

Procedure used to hide the group assigment from the participants in a study to prevent their beliefs about the effectivness of a treatment from affecting the results

ex: when effectivness of a drug is being tested the control group receives the placebo

• occur when participants are tested more than once in a study with early testing affecting later testing
• Occur more in within-subjects

natural changes that occur to the participants during the course of a study that can result in bias

• a source of bias that can occur in a study due to participants changing their behavior because they are aware they are being observed
• Landsberger 1955 as he was analyzing a set of studies conducted at the hawthorne plant

• a source of bias that can occur in a study due to participants changing their behavior based on their perception of the study and its purpose
• occurs when partic try to figure out purpose of the study

• observe obtrusively
• naturalist observations
• annonymous responses or confidential

• Dependent Variables are measured from individuals and are present in all research studies
• Confounding variables: variables that may affect the DV in a study but are not of interest to the researcher in an experiement ( can bias results)
• Independent Var: variables manipulated by the researcher
• Quasi Independent : present in quasi-experiments and experiments. they are similar to IV in that they involve a comparison of groups, but are not manipulated ( levels are not randomly assigned)

• having good internal validity in a study means reducing bias that can affect the results and providing a good test of the hypothesis
• having good external val means studying behaviors and obtaining results that generalize to individuals beyond the study in everyday life
• having reliable data in a study means having results that would occur again in similar situations

• a system of moral principles 
• no right or wrong answer ( grey areas )
• used in every part of reasearch process

• 1932-1972 Tuskagee alabama
• 4 african men attracted syphallis and it can lead to dimensia
• doc wanted it to run its course to see its effeccts
• ISSUE: 1947 penacillin could treat it but doc didnt administer meds bc they were in a rural area
• when experiment was relseaed ppl were mad bc it was not ethical

• set of ethical guidlines developed for research with human participants based on info gained during Muremberg trials after WWII
• focuses on issues of informed consent and coercion

1. participation is voluntary
2. informed abour risks involved
3. research must contribute to scientific knowledge
4. performed by qualified researchers
5. no unnecessary physical / psychological harm
6. take precautions about the risks
7. ensue benefits outweigh risk
8. discontinue if harmful
9. ability to withdawl

• based on nuremburg code

1. confidentiality
2. informed consent
3. right to withdrawl
4. reduce harm to deception
5. risk vs. benefit

• 1963 during nuremburg trials defendants argued they werent responsible for their wartime actions b/c their actions were carried out to follow orders of their superiors
• What happened in Nazi concentration camps
• Study examined the effect of an authority figure on participants behavior
• Participants were recruited to administer a memory task to a second participant. The second participant was a confederate, not an actual participant but acted like one to make the participant beleive he was just another participant in the study
• Participant administered a memory test , when the person got it wrong the participate had to administer a shock and for each one wrong the level of the shock increased ( not really being shocked) they were able to hear yelps
• Study showed that the presence of an authority figure greatly influences peoples behavior, to the point where people may harm another person when ordered to do so. 
• They were fully debriefed so no long term harm
• critics said the stress and deception was to great and psycholog. harm

• 1973 critisized for stretching ethical standards for research by philip zimbardo at stanford university
• Interested in how the roles we are given in a society affect our behavior toward others
• Puts out ad saying needs participants ( random assignment) where some are prisoners and some gaurds.
• made sure they all were of same in intellegance and personality characteristics
• PLanned for 2 week experiment , lasted few days bevause study became harmful to the participants
• Prison guards got to big headed bc of the power they had and were doing unethical things making prisoners break down which showed how people get swept up in roles in society
• Ethical: considered alteritave ways of studying the effects of the roles, received informed consent , and debriefed, also stoped it early
• critics say the participants shouldnt have been put in such a stressful postion

• lists the responsibilites of researchers as they conduct research with human participants and the rights of those participants before , during and after a study
• ethical guielines researchers must adhere to

1. Respect for persons
2. beneficence
3. Justice

• provide info about study before it begins ( nature of participation , purpose , risk, benefits)
• Informed consent: obtain voluntary consent after they are informed
• give opportunity for questions
• inform them of the right to withdraw
• consent form

• reduction of the risk of harm to the participants as compared with the benefit of the study
• risk benefit analysis
• confidentiality

• fair selection of participants
• all participant groups must have opportunity to receive benefits of research
• no participant groups may be unfairly selected for harmful research

• committe of knowledable individuals who oversee the ethics of research with human participants conducted at an institution
• made up of 5 members with backgrounds sufficent to ensue qualified review of the research proposal
• 3 categories based on the amount of risk there is to participants

1. exempt studies
2. expedited studies
3. full review studies

• samllest amount of risks involved and brief reveiw
• studies involving educational or cognitive testing of participants
• archival studies where part arnt identified 

• involve moderate risk
• studies reviewed by only one member
• noninvasive medical procedures ( heart rate, brain wave activity ) collection of blood ( finger stick ) video/audio taping of observations, survey/questionaire with minial emotional impact

• Highest amount of risk and receive full review by IRB
• invasive medical procedures, high psychological risk, questionaires about behaviors that can cause emotional distess

• researcher will minimize unnecessay risk to participants
• risk is justified by the potential benefits of the study
• selection of the particiapant is fair and appropriate for the study
• informed consent procuess is included and documented
• researcher will monitor collection of the data to ensure the saftey of the participant
• privacy and confidentiality will be protected
• if a special participant group is included , procedures must be included to protect the rights of these participants

IACUC 

Institutional Animal Care and Use Committee

• Justification: use of animal , benefits , species chosen
• Care and Treatment: trained researcher , animal kept in good health, housing meets fed guidlines
• Aqusition:bred hummanley , obtained humanly if taken from the wild

1. minimun dicomforst
2. greater justification when greater discomfort
3. surgical procedures carefully overseen
4. animals cant be released in the wild

1. errors in the data are reported
2. plagerism: claiming anothers work or ideas as ones own
3. properly cite sources

• how we choose subjects
• affects validity of study
• learn about a large group by testing a small group

the difference in the observations btwn the sample and the popluation. 

• this happens bc the sample will never give you the same observations as the entire population would
• to reduce this: probability sample and convienence/purposive sample

individuals are chosen at random from the population such that the chance of any one individual being chosen is known

• ex: Shu college students out of 14,000 you know probability because its 1/14000

• sample chosen such that the probability of an individual being chosen cant be determined, because the individuals arent chosen randomly.
• May not work as well as probability sample because not chosen at random

• each member of the population has an equal probability of being selected using random sampling
• ex: students chosen randomly from list of all students at a university. Population is 20,000 kids and sample 1,000 kids so 5% chance of being chosen
• reduced sampling error by choosing from all members of the population to best represent the population
• difficult to ensue that each member of a large pop can be chosen in a sample

• individuals are chosen at random from groups within a population 
• ex:population is all college kids and each group represents a different college. 1 cluster is chosen at radnom = 1 school and you sample students chosen randomly from each cluster
• makes it easier to choose members randomly from smaller clusters to better represent the pop
• can ignore segments of the pop that are not in the clusters chosen for the sample

• members of a pop are selected such that the proportion of a group in teh sample is equal to the proportion of that group in the pop using random sampling 
• ex: 60% of US college kids are Female and 40% are male. So  you would conduct your experiment 60% are F and 40% are M so out of 100 kids 60 are F and 40 are M.
• reduces bias due to an identified characteristic of the pop by equating proportions in the sample and the pop for that characteristic to better represent the pop
• similar to simple random - can be difficult to ensure equal probability of being chosen from a large group pop

• Individual is chosen at NON random from a pop such that available individuals are chosen and the chance of any one individual being seleceted is NOT known

• member of pop are chosen based on convienece and on who volunteers
• ex: sample is chosen from students who volunteer to complete an extra credit assignment in their psych class
• easier to obtain than probability samples
• may not represent the pop well due to selection bias bc random sampling isnt used

• members of the pop are selected such that the proportion of a group in the sample is equal to proportion of that group in population 
• non random choosing so the proportion of individual with a particular characteristic is equivlent in the pop and in the sample
• like stratified random sample without the random selection
• ex: volunteers are recruited from students in a psych class such that the gender break down of all students at the school
• easier to obtain than probability samples and allows for better representation of a characteristic of the pop
• may not represent the pop well due to selection bias bc random sampling isnt used

measures that help us summarize data sets.

1. central tendancy
2. variability
3. graphs/tables

measure that indicated a typical score in the distribution

1. Mean: average! outliers, saleries, grades and used to compare
2. Median: middle score
3. Mode: most common score

• Indicates how much the scores in the distribution differ from each other across the response scale.
• low variability is better bc it shows clearly whats going on ( skinnier curve ) 

a measure representing the average difference btwn the scores and the mean of a distribution

1. difference btwn each score and the mean
2. square those values
3. after squaring add them all up
4. divide by the n-1
5. THAT GIVES YOU THE VARIENCE
6. take square root of that # to get SD