Emotional-unintended stress response

Physical-Bleeding

economic- job loss because identity is revealed

"Ethical Principles and Guidelines for the Protection of Human Subjects of Research".

Prompted in part by problems arising from the Tuskegee Syphilis Study and based on the National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research (1974–1978). Continues as an essential reference for institutional review boards (IRBs) that review HHS-conducted or -supported human subjects research proposals involving human subjects, in order to ensure that the research meets the ethical foundations of the regulations.

Getting permission and ensuring the right of a person by

1. giving a description of study
2. tell participants what they will do in the study
3. descibe risks and benefits
4. compensation
5. confidentiality
6. giving voluntary freedom to withdraw
7. giving contact information of researcher and IRB

Simulation Studies- "real world phenomenon"

&

Honest experiments

ensuring reasonable, non-exploitative, and well-considered procedures are administered fairly — the fair distribution of costs and benefits to potential research participants — and equally.

Bad ex. Tuskegee study

Deceptive study- Participants were told that the study was about learning when it was really about obedience to authority.

Shock was delivered by "the teacher"/participant to the student/ actor. No matter how much the student suffered, the teacher was told to continue by the researcher/authority.

AKA Stanford prison experiment was a study of the psychological effects of becoming a prisoner or prison guard.

24 subjects were randomly assigned to prison or prison guard group and acted out the roles for 6 days. 

Problems:caused psychological and emotional stress, deception (surprise arrest, prisoners believed they were held against their will), questions of generalizability and repeatability

No risk-no interaction (observational)

Minimal risk

 Greater than minimal risk-always place under scrutiny

Extreme risk- could lead to death

Perry School Project

Did the cost outtweigh the benefits

T-Test

Intro/Lit Review

Methods

Results-What happened, significance, drawbacks, surprises, went as planned

Discussion-What does it mean to in the study and to the world

What went on in past experiments

Focus on efforts

meta analysis-comprehensive review of literature

Sample:Who, What and how did you get subjects?

Data: variable, sources of info, operational definition

Analysis:

1. Are results significant,
2. What happens
3. What doea it mean
4. What influence does it have on society

t-test

ANOVA

f-test

Chi ^{2 -}

1. only relationship association
2. nominal level data
3. (if from experimental design, then results can be correlation)

Regression-

1. Relationship among lots of variables-only one outcome variable
2. has to be interval or better
3. Good for linear associations only

An F-test is any statistical test in which the test statistic has an F-distribution under the null hypothesis.

Ex.The hypothesis that the means of several normally distributed populations, all having the same standard deviation, are equal. Doesn't tell the difference, only that a difference exists

One-way analysis of variance

H₁ & H₀- each represent a population; only tells that there is a difference

Position isn't defined one way or another.

(ethical basis for research; you must stop if one side is clearly better than another)

Revised in 2002

Benefice, responsibility, justice, respect, integrity, dignity of others

10 standards that address specific issues concerning the conduct of psychologists

How sure are you about the data represented (95% or better)

Upon repeated draws, the means of each falls within the specified interval

Larger sample size=precision +power, but not necessarily truth (bias and error variance are factors)

Sample Size

Variance-Error &Bias

Treatments (randomly equivalent)

1. Every member of the population has an equal chance of being chosen
2. Has a sampling frame/list of those eligible
3. Simple random-random numbers
4. Systematic random- uses every n^th #

All AU students

1. Population divided into strata-meaningful criteri (Sex, race, age, education level)
2. Small populations are guaranteed to be in sample, making variable show up

All girls at AU

1. Do not need a list
2. Population could be defined by geography or location
3. individual elements can be defined as posessing chacteristics at each level

All female AU students living in DC

Probablity-Simple, stratified, and cluster

Non-probability-convenience, quota, purposive

1. subjects close to you
2. saves money, time and easy to access
3. used in pilot studies
4. drawbacks:-not representative of population

Purposive

1. elements are chosen to serve a preconceived objective
2. done by marketing research or focus groups
3. target is selected because they satisfy a criteria of interest (people who drink coke after exercise)
4. will often answer questions and possibly aid in development of drugs
5. drwback:sample is not considered representative

Use methods of convenience sampling, but researcher builds composition based on a broader representation

(Minority pops may be sample more until each piece of info is met)

Makes a statement about a population based on sample statistics.

Guess the significance

If two groups are randomly significant, differences in the dependent variable are due to manipulation of the independent variable. 

Describes how you are going to measure the subject of interest. 

Measures like blood pressure and weight are manifested.  Stress and perception is not.

Nominal

Ordinal

Interval

Ratio

Exhaustive and mutually exclusive. 

Male, female, banana, right, left (if you belong to one group, you cannot belong to the other).

Categorical variable-individuals are put into a group but there is no explanation as to how the groups relate

Exhaustive, rank order, mutually exclusive

1st, 2nd place, degree

Can only say one level is higher than the other

Exhaustive, mutually exclusive, rank order, significant difference.

Ex. temperature-the difference between 70 and 80 degrees and 40 and 50 degrees is 10, but you can't say one 10 has more heat than the other.

Exhaustive, mutually exclusive, rank order, significant difference, meaningful 0.

Amenable to mathmatical operations. ex. money, time, height, weight

Other factors that could have an effect on the outcome are controlled for.

Experimental- representative sample can be randomly place in control and treatment group

Non-experimental-control variables can be collected and used in a way that group differences can be minimized. (does not take the place of randomize control trial

Causality

1. Statistical association
2. Temporal Order
3. Absence of Rival Causal Factors

Manipulates the independent variable on the dependent variable.

Soley depends on randomization being done correctly

drawback:  confounder may cause a nuisance effect

Targeted to isolate the effect of manipulation of the independent variable.

Pretest allows for some Assessment of whether groups are randomly equivalent.

Can be used as a covariate to increse statistical power to detect treatment effect.

Drwback: pre-experimental observation may effect the experimental treatment

Extension of pretest post test design.

Observations ate taken at multiple points before and after the treatment is given.

can be used to provide understanding  of the relationship among different variables

Cross-sectional, large, usually survey based

Cannot claim causality

Data is made up of information or observation from respondents.

No random assignment to a treatment or control condition.

Control group members may work extra hard to see that expected superiority of the experimental group is not demonstrated.

ex: a pedometer challenge, where the controls begin walking more

Goal of Sampling

How is it Done

Minimiz bias and sampling error

random assignment and sampling is important, but hard