Loading presentation...

Present Remotely

Send the link below via email or IM


Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.


Intro #5: Research & Ethics

No description

Cyd Skinner

on 6 September 2016

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of Intro #5: Research & Ethics

We're just talking about describing your results.
Normal curve
Very few things work
on a bell curve.
IQ tests are forced into a bell shape
It happens more often that there are more scores toward the top or the bottom.
Height does
Most don't
grades in college classes
Statistical Significance just means that your results did not happen by chance.

It does not mean your results are important!
3 things that may indicate your results are important:
effect size
How important is the difference?
sample size
Larger samples are more likely to be accurate - to match the people (as a group) we want to compare them to.
Some data varies a lot, both in the sample and in reality. But, the less variation in both groups, the more sure we can be of our results.
central tendency
mean and median
standard deviation
frequency distribution
bar graphs, histograms
demand characteristics
biased sample
observer expectancy
subject expectancy
Can we trust the results?
get the same results each time we test?
Ethical Issues
1. Right to privacy
2. Do no harm
3. Deception
Institutional Review Board
A new drug may be shown to help in a statistically significant number of people...


It helps them only a little - the difference is barely noticable.
A new drug may be shown to help in a statistically significant number of people...


It helps a lot.
Statistical significance does not mean that the results are meaningful - just that they didn't happen by chance!
Some, but not a lot of variability in this sample of house pets.
A lot more variability, still house pets
Milgram's subjects
thought this behavior was needed by the experiment
baseline study 40 white, employed, males
Was the test done correctly?
Can we generalize?
Use the right tool?
Look right?
Match other results?
In the Penn State community, an alumni group is pushing for a bronze statue of Paterno to be restored on campus, and for the university to repudiate a 2012 report by former FBI director Louis Freeh that blamed Paterno, other university leaders and a “culture of reverence for the football program” for Sandusky’s rampant sexual abuse.

Paterno died of lung cancer in 2012, just months before Sandusky was convicted of sexually abusing 10 boys and sentenced to 30 to 60 years. Sandusky, 72, is appealing his conviction.

The 1976 victim, identified in court records as John Doe 150, said that while he was attending a football camp at Penn State, Sandusky touched him as he showered. Sandusky’s finger penetrated the boy’s rectum, Doe said, and the boy asked to speak with Paterno about it. Doe testified that he specifically told Paterno that Sandusky had sexually assaulted him, and Paterno ignored it.

“Is it accurate that Coach Paterno quickly said to you, ‘I don’t want to hear about any of that kind of stuff, I have a football season to worry about?’ ” the man’s lawyer asked.

Washington Post, 7/12/16
A culture that "discourages discussion and dissent"
Louis Freeh
Full transcript